Some aphorisms on software development methodology

The net benefit of having anything that can be called a software development methodology is in inverse proportion to the quality of your developers – and possibly inverse-squared.

Any software-development methodology works well on sufficiently small projects, but all scale very badly to large ones. The good ones scale slightly less badly.

One thing all methodologies tagged “agile” have in common is that they push developers away from the median. Competent developers work better; mediocre developers don’t change; incompetent ones get worse.

Software metrics have their uses. Unfortunately, their principal use is to create the illusion that you know what’s going on.

Structured development methodologies have their uses, too. Unfortunately, their principal use is to create an illusion of control.

Trust simple, crude metrics over complex ones because the simple ones are less brittle. KLOC is best, though a poor best.

Agile development is efficient only in environments where the cost of flag days is low. Otherwise, slow down, and take time to think and write about your architecture.

Good programmers are difficult to control; great ones are nearly impossible to control. Different methodologies require different kinds and degrees of control; match them to your developers wisely.

Process is not a good substitute for judgment; when you have to use it as one, your project is probably too large. Sometimes this is unavoidable, but don’t fool yourself about what that will cost you.

The difference between O(n**2) and O(n log n) really matters. It’s why lots of small teams working coordinated small projects works better than one big team building a monolith.

A million dollars is roughly a 55-gallon oil drum full of five-dollar bills. Large-scale software development is such a difficult and lossy process that it’s like setting fire to several of these oil drums and hoping a usable product flutters out of the smoke. If this drives you to despair, find a different line of work.

385 comments

  1. In a commercial setting, illusions can be a useful way to help the people who sign the checks remember why they hired specialists, and thereby resist the temptation to interfere too heavily.

  2. @esr: “Different methodologies require different kinds and degrees of control; match them to your developers wisely”

    I’d also posit that different projects require different methodologies, and both selection of methodology and selection of developers should be driven by the requirements of the project.

    Both are architectural decisions, but of course, you never get time to properly consider architecture.
    ______
    Dennis

  3. @baylink

    Lol…when do you know some technology is passé?

    When the federal government adopts it.

  4. “Any software-development methodology works well on sufficiently small projects, but all scale very badly to large ones. The good ones scale slightly less badly.”

    Meh…classic software engineering methodologies are mostly useless on small projects and are geared toward large ones. Software engineering practices work about as well as managing any other large scale engineering effort.

    We’ve been building dams and bridges for thousands of years. The first bridge made from a fallen log over a creek was probably 50% late and required double the original manpower estimate to complete. IMHO large modern civil engineering projects have about the same batting average as large software projects. For every “software disaster” trotted out to show that software development methodologies suck I can show a nuclear plant, bridge or dam project that had a similar outcome. Massively late, massively over budget and then cancelled.

    However, that’s not to say there isn’t a huge difference between “scales badly” and “scales slightly less badly”.

    I would also say that “process is not a substitute for good developers” rather than judgement to make the point more clear. No emphasis on process will beat good hiring practices despite the desire to treat developers as mostly interchangeable parts in most software process philosophies.

  5. This post makes about as much sense as judging a weight-lifting competition using a stop watch, but I’m pretty sure that’s the intent, lol!

  6. Comparisons between construction and software are generally poor because software development is more making a construction project, i.e., what the architects do; the actual construction work is like deployment in an already known environment. It is predictable, and the influence of weather and other such factors is well-studied and known.

    Rolling out, say, RPMs into a system, in case the software is not broken in the first place, is quite a predictable process, too. The actual development, well, not so much.

  7. @nht

    For every “software disaster” trotted out to show that software development methodologies suck I can show a nuclear plant, bridge or dam project that had a similar outcome. Massively late, massively over budget and then cancelled.

    Please, do, by all means.

  8. @nht: ” For every “software disaster” trotted out to show that software development methodologies suck I can show a nuclear plant, bridge or dam project that had a similar outcome. Massively late, massively over budget and then cancelled.”

    Which simply makes a case that *all* development methodologies suck, and the problem isn’t unique to software.
    ______
    Dennis

  9. >”One thing all methodologies tagged “agile” have in common is that they push developers away from the median. Competent developers work better; mediocre developers don’t change; incompetent ones get worse.”

    This assertion is highly interesting to me. I’ve worked in agile and non-agile shops, and have found the former to be more functional than the latter. In particular, the concept of the daily stand-up seems beneficial to all levels of developers: “This is what I did yesterday, this is what I plan to do today, and this is what I’m having a problem with.” That last seems most important, particularly for the incompetent developers – in order to get feedback from the competent ones. No?

  10. Over the years I’ve come to believe the old-fashioned top-down modular method worked the best. A detailed specification, flowcharts, modularization, and detailed descriptions of how the bits and bobs interact.

    Lots of boring planning and documentation, but it tends to produce working code that can be effectively maintained after the original developers have moved on.

  11. Meh…classic software engineering methodologies are mostly useless on small projects

    Well of course they are. Just like a rigidly and strictly defined social structure is useless at family scale. But they still work at that scale.

    and are geared toward large ones.

    Indeed. Sadly they then must cross the legendary Chasm of Intent vs Reality. A yawning gap which has claimed many, many lives.

  12. @ esr

    Different methodologies require different kinds and degrees of control; match them to your developers wisely.

    I get a vague Sun Tzu vibe from this. Did you intend it that way? :-)

    1. >I get a vague Sun Tzu vibe from this. Did you intend it that way? :-)

      Not consciously, but I can’t answer for my unconscious mind.

  13. None of the methodologies solve the real problem, which is that the software team comes in totally ignorant of the the subject of the software, while the ‘domain experts’ really don’t understand it either (they just think they do.) Under such circumstances, estimation is useless, version 1 is off-the-wall. You can only settle down and hope that everyone can learn what’s needed before the whole thing is cancelled.

  14. “None of the methodologies solve the real problem, which is that the software team comes in totally ignorant of the the subject of the software, while the ‘domain experts’ really don’t understand it either (they just think they do.)”

    This is not true if the team has done the same kind of work in the past. This is why CMMI level 5 organizations exist and why they are successful in making the next similar thing.

    When they move to a different domain, arguably they can’t be level 5 anymore…or possibly not even 4. Which is, in the end, why CMMI has kind of petered out even in the most rigorous software domains.

  15. “Which simply makes a case that *all* development methodologies suck, and the problem isn’t unique to software.”

    I would rephrase as “all large engineering efforts are hard so singling out large software projects for failures is silly”.

    Especially since civil engineering has existed for a much longer time than software engineering.

  16. Another important function of software process is to limit the damage an incompetent developer can do. I would argue that this is the principle use, but then I mostly try to ignore illusions unless they’re working toward my goals.

    Furthermore, we’re all incompetent at least some of the time. Caffeination level, emotional state, amount of sleep, and other things can easily combine to cause days of incompetence. (Incompetence, of course, which humans are very poor at recognising in oneself.) Software process can limit the damage we can do by slowing things down, adding reviews, adding testing steps, adding sign offs, and so on.

    When I’m looking at process, it’s mostly a question of “how bad will it be if this blows up in the field”. The more severe the answer to that question, the more process I try to wrap around the development.

  17. @yaroslav

    “Please, do, by all means.”

    What? Google doesn’t work for you? In any case I said I could show one civil engineering project with the same issues as any software disaster project provided. Given you have provided no software failures what do you want me to provide at this point? A bunch of links to civil engineering problems?

    Sure thing.

    Google search: “bridge behind schedule”

    http://www.express.co.uk/news/uk/677076/Billions-project-Firth-of-Forth-Scotland-delay-schedule-Queensferry-Crossing

    http://uawire.org/news/kerch-strait-bridge-project-is-falling-behind-schedule

    http://www.njtvonline.org/news/video/pulaski-skyway-project-behind-schedule/

    http://windsorstar.com/news/local-news/howe-bridge-project-appears-stalled

    Google Search: “power plant behind schedule”

    http://www.nytimes.com/2016/07/05/science/kemper-coal-mississippi.html

    http://www.ctvnews.ca/business/first-u-s-nuclear-plant-in-decades-is-over-budget-behind-schedule-1.2212928

    http://www.postandcourier.com/article/20140812/PC05/140819838/sc-nuke-plant-project-delayed-1-year-sceg-says

    Google Search: “dam behind schedule”

    http://www.desmog.ca/2016/06/30/site-c-dam-already-cost-314-million-more-expected-behind-schedule-new-documents-show

    http://cdapress.com/news/local_news/article_ee80a3bc-6291-563a-996d-2700f1d4f553.html

    Google search “dam cost overrun”

    https://www.internationalrivers.org/blogs/227/large-dams-are-uneconomic-scientific-study-finds

    Money quote:

    “Large dams suffered average cost overruns of 96%. The degree of cost overruns tended to increase with the size of projects. Even without considering social and environmental costs, large dams on average don’t make economic sense.”

    “Project implementation suffered an average delay of 44%. The implementation schedule does not include the lengthy lead time required to prepare projects.”

    Builders DO build buildings like programmers write software.

  18. “Comparisons between construction and software are generally poor because software development is more making a construction project, i.e., what the architects do; the actual construction work is like deployment in an already known environment. ”

    Most software development teams have more folks than most architectural teams. Managing large software projects isn’t very different from any other large systems engineering problem.

    A large architecture firm might have 300-1000 or so employees and work multiple projects at once.

    Large software projects are bigger than most architectural firms in their entirety and managing their work is far more complex than design but also covers implementation which is a significant part of major civil engineering projects. Architects don’t do the implementation part.

    We do. And we do it to a schedule and our costing is a lot more like costing a major construction project than just the architectural phase because shit happens during implementation and estimates are mostly wrong (but tend to cancel out a bit on larger projects until they don’t).

  19. > Agile development is efficient only in environments where the cost of flag days is low. Otherwise, slow down, and take time to think and write about your architecture.

    What do you mean by ‘flag days’ in this context? I assume you mean something like ‘unproductive days’ rather than national holidays (the more common meaning of the term). If so, are you meaning that agile is only efficient in environments where you have relatively consistent productivity?

  20. @lliamander

    A software change that is neither forward- nor backward-compatible, and which is costly to make and costly to reverse. “Can we install that without causing a flag day for all users?” This term has nothing to do with the use of the word flag to mean a variable that has two values. It came into use when a change was made to the definition of the ASCII character set during the development of Multics. The change was scheduled for Flag Day (a U.S. holiday), June 14, 1966.

    The change altered the Multics definition of ASCII from the short-lived 1965 version of the ASCII code to the 1967 version (in draft at the time); this moved code points for braces, vertical bar, and circumflex. See also backward combatability. The Great Renaming was a flag day.

    http://www.catb.org/jargon/html/F/flag-day.html

  21. I would add:

    Metrics are not deliverables. High “velocity” is not a deliverable. A nice looking “burndown chart” is not a deliverable. Estimates and task breakdowns are not deliverables. Working code — THAT’S your fucking deliverable. Documentation and tests are supplementary deliverables. Your process either keeps overhead low enough to stay out of the way of programmers getting working code where it’s needed, or it’s shit.

  22. Reminds me of the saying: Those who are convinced it cannot be done must not disrupt those who are doing it.

  23. @Jeff Read:

    Metrics are not deliverables. High “velocity” is not a deliverable. A nice looking “burndown chart” is not a deliverable. Estimates and task breakdowns are not deliverables.

    Goodhart’s Law: “When a measure becomes a target, it ceases to be a good measure.”

  24. @Jeff Read

    Epic comment. Made me smile.

    I always, always remember to tell the managers when we’er starting a project that the code is the damn deliverable. Actually a fully working system is the deliverable, but your pithy sentance captures the feel better.

    1. >The optimum number of developers per project, any project, is equal to 1.

      True, but sometimes this is unachievable. Fixing NTP, for example, is not even possibly a one-man job. It has a surgical team because it needs one.

  25. Comparisons between construction and software development are indeed silly. Tit for tat ‘for every one of these, I’ll show you one of the other’ are even sillier. What matters is the percentage of projects that end up being functional. Civil engineering projects are often late, overrun their budget, yes. But it’s pretty rare that they are completely useless and have to be scrapped, unlike large software projects which end up in that position with distressing frequency.

    I’ve worked on road and rail projects for 15 years (I’m an electronics hardware/systems engineer with an early background in industrial control, so I’m generally at the interface between the devs and the civil/mech/elec cohort – talk about a clash of cultures) and I would estimate that about 50% of the projects I’ve worked on have been on time and budget, with the rest suffering some degree of overrun on both. Also some poor engineering and management decisions that led to expensive fixes. But, and this is a big but, they all have trains/cars/buses running on them (or in them in the case of tunnels). I’ve never seen construction project completely abandoned. I have seen a year’s worth of development on the operation and control software completely thrown out and started from scratch a few times. And this is for a fairly well understood problem space. Software development can go invisibly off the rails in a way that construction projects generally don’t.

    The exception to this is construction projects involving new technologies, which tend to map better to software engineering failures. The ‘clean coal’ power plant nht linked to above is a good example. Another I would nominate in that field is Geodynamics and their deep geothermal efforts in South Australia. Sounded nice and simple – drill down five kilometers to hot rocks, inject water, get steam, power! They originally planned to have a pilot plant up and running 10 years ago. They may yet solve the massive engineering problems they’ve encountered since, but I wouldn’t hold your breath.

  26. In any engineering project, the more precisely and definitely the questions “what functionality is required” and “what elements / features are necessary to provide that functionality” can be answered before development starts, the more successful that development process is going to be. Consider a building, and an operating system. I think you would have to be looking at a very large and complex building before the complexity and uncertainty of the answers to those questions would anywhere approach that for even a simple OS. So I’m in the camp which considers the comparison between buildings and programs rather facile.

  27. Precisely trying2b – requirements analysis is not generally a major factor in Big Engineering projects. I wouldn’t even compare them to an operating system. GPSD alone is probably orders of magnitude more complex than tunneling.

  28. @Jeff Read:
    Is there a good RFC for sending a standing ovation over the ‘net? I’ve seen enough dysfunctional workspaces that needs that list of rules or their equivalent nailed over the door. I’ve also seen enough broken rollouts to have one further nitpick:

    Working code — THAT’S your fucking deliverable. Documentation and tests are supplementary deliverables.

    IMHO documentation is not “supplementary deliverables”. Code nobody can figure out how to use is as valuable as broken code, and both are arguably worth less than no code at all. Code, user documentation, and docstring comments[1] are all core deliverables to me.

    If you never ship code, the project obviously failed. If you ship code but nobody uses it, the project obviously failed. If you ship code, it goes into use, but nobody can maintain it, that’s a future failure in the making. In all three cases, we can easily identify the missing element which would prevent such failure, so we have good cause to make it a day-one guaranteed deliverable.

    [1]: Please don’t read my saying “comments” as limiting this to in-line code annotations. From context, I do mean to include any technical documentation targeting future system maintainers: the sort of things that are typically inward-facing, and address the architecture, implementation, or interface of your program. I just feel saying “technical documentation” would be too broad, as many of the things currently under that umbrella are not the sort of product I feel is critical.

  29. I’ll just add my two cents. From what I see, far and away the most common cause of software failure is not technical, but requirements. The problem is that what the users want or need is not clearly known (and in some cases is even unknowable.)

    However, it is a great plan by managers to blame the wrong people. Run a software project for a year, constantly change the requirements throughout that year, throw a bunch of new features and major requirement changes in at the last minute, then blame the crappy programmers for not meeting their deadlines.

    Of course, anyone who reads Dilbert already knows this.

    This is why I am a fan of agile. At its core it is about the dynamic discovery of requirements, and the design of an architecture that is not brittle. This, from what I can see, best meets the reality of a constantly changing requirements space.

    All that planning — I have rarely seem it be anything other than a hindrance. Rather than planning what you need are programmers and designers who have an innate sense of what is a good way to do stuff. (There are exceptions to this for sure — one would be technology selection, another regular reviews and communication session about “where we are at and where we are going” are definitely valuable, and big picture interaction diagrams definitely have value. But there are a billion pounds of unread design specs out there that cost more than their weight in gold, and have less value than their weight in poop.)

    Having said that though, I have never been involved in a truly massive project, so perhaps my observations don’t apply there. For sure, BIG programming is qualitatively different.

    I’ll also add, those mediocre programmers? They don’t matter, they just don’t contribute all that much. And the bad ones? Better to send them home to watch Oprah, or maybe have then write code coverage unit tests until they quit.

    However, I do think that two or your aphorisms contradict. The first suggests that if you double the quality of your programmers you half the benefit of a methodology, whereas the third suggests that agile significantly improves if you have good programmers.

    Of course all good aphorisms have an opposite aphorism that is also true. But it just struck me. I’d suggest that in future when creating a list like this that “many hands make light work”, not forgetting that “too many cooks spoil the broth.”

    1. >However, I do think that two or your aphorisms contradict. The first suggests that if you double the quality of your programmers you half the benefit of a methodology, whereas the third suggests that agile significantly improves if you have good programmers.

      I think both things are true. You have to freeze all the other variables to observe the effects, though.

  30. > Adjacent to topic: http://siderea.livejournal.com/1241996.html
    “What Software is Made Of”, by siderea

    I really liked the following quote from this:

    There is a joke in science, “If we knew what we were doing, we wouldn’t call it ‘research’.” I have long wished we had an expression for a similar sentiment about programming. Because programming naturally, regularly, and just about always involves an encounter with the unknown, and having to reckon with what you didn’t reckon with.

  31. @LTW

    “But it’s pretty rare that they are completely useless and have to be scrapped, unlike large software projects which end up in that position with distressing frequency.”

    Citation needed. People claim this all the time that civil engineering is FAR superior to software engineering and hand wave the proof as “obvious”. Well, it’s not obvious if you actually look into it.

    The primary difference that I can tell is that most of the time the incomplete project gets bought and completed by someone else. We don’t do that in the software industry but if every disaster project that got cancelled was released as open source I bet we would see a lot of repurposed code that is analogous to someone buying assets for pennies on the dollar and finishing a failed construction project.

    When you read civil engineering papers they sound JUST like ours:

    “At least one project failed or underperformed last year for more than 60% of organizations that spent $10 million or more on capital construction projects—in large part because of personnel issues, according to KPMG’s 2015 Global Construction Survey.”

    http://www.constructiondive.com/news/why-do-so-many-major-construction-projects-fail/387382/

    And for construction project failures it’s not just for projects that were cancelled but also for structural failures that kill people. Something that software some times does but typically not for most IT projects.

    http://www.fep.up.pt/disciplinas/PGI914/Ref_topico1/Doc_Analyzing_Cons_Failures.pdf

    Civil engineering project failure root causes they look very similar to software project failure root causes:

    Scoping issues – Project scope does not fully address organizational business requirements
    Inexperienced or unqualified project team – Project team lacks appropriate skills and expertise to manage the project
    Poor estimating – Project estimates are incomplete or insufficiently detailed for budgeting
    Lack of integrated budgeting & planning – Project business requirements are not aligned with budget and execution plan
    Incomplete & fluid design – Construction commences based on an incomplete design and project scope is continually in flux
    Lack of proactive risk management – Project risks are not fully understood or vetted prior to project approval
    Unrealistic schedules – Project delays during planning and approval result in compressed schedule milestones and unrealistic completion targets set by management

    https://www.kpmg.com/BE/en/IssuesAndInsights/ArticlesPublications/Documents/avoiding-major-project-failure.pdf

    http://web.usm.my/jcdc/vol19_1_2014/JCDC%2019(1)%202014-Art.%203%20(35-52).pdf

    https://www.researchgate.net/publication/271906893_PROJECT_FAILURE_FACTORS_AND_THEIR_IMPACTS_ON_THE_CONSTRUCTION_INDUSTRY_A_LITERATURE_REVIEW

    http://www.ijirset.com/upload/2014/icets/2_CE101.pdf

    Replace “construction” with “software development” in those papers and other than weather factors it would sound like the “software projects suck” articles in IT journals

    So citation needed.

    I’ve given you software engineering naysayers plenty of examples that when project complexity reaches a certain point both civil and software projects start looking the same. The technical details differ but the root cause of failures are the same.

    We do about as well as they do especially given they’ve been doing it thousands of years.

    PMBOK for civil engineering looks pretty much the same as SWEBOK for software (it made my Systems Engineering professor irate when I pointed out the Software Engineering was on par or better in many areas than Systems Engineering) because the issues faced in any large engineering effort are more similar than different.

    A Roman project manager would look at the root cause lists above and nod his head and say “Yep, I was telling Maximus the other day that the &*(^&*)^ customer didn’t have a clue what he wanted and kept changing the damn specs. What the heck? He wants to have BOATS in the Collesium now? That’s going to be one beast of a change order.”

    Folks that think the construction requirements are well understood before construction begins has never built a house for/with his wife.

  32. @jessica “However, I do think that two or your aphorisms contradict. The first suggests that if you double the quality of your programmers you half the benefit of a methodology, whereas the third suggests that agile significantly improves if you have good programmers.”

    Good coders trump good processes.

    The caveat is that while everyone claims to be only hiring the top 5%ers as staff, 95% are wrong.

  33. >Process is not a good substitute for judgment; when you have to use it as one, your project is probably too large. Sometimes this is unavoidable, but don’t fool yourself about what that will cost you.

    This is probably true of management of businesses and government organizations, as well. Process has a role to play, but when taken too far it can result in a lot of time and expense being spent on simple tasks. For example, performing tasks through a project management framework which starts by forming a project steering committee which meets for a project initiation meeting where they agree on such things as timescales, project milestones, project scope, and the procedures for reviewing progress on the project and whether it still comports with organizational goals. You can see that all of that stuff could be useful for large or complicated projects, but in simple cases someone should just make a decision. Similar is recruitment, which HR people and managers pretend is a scientific process which if you apply precise rules for recruitment in a mechanistic fashion, you can get employees who will be good at their jobs, without any real need to be a good judge of character or of someone’s likely competence, or even understanding what the job entails; when actually it is mostly superstition achieving good results only by accident. Large organizations are full of people operating a process and lacking in any intelligence or sense of creativity, achieving some results thereby, but slowly and at huge expense. Unfortunately, there may not be an alternative (other than smaller organizations, maybe?).

  34. Jessica’s point about agile’s place in a world full of shifting requirements may make more intuitive sense if you think of agile as not so much a methodology for managing programmers and rather as a methodology for managing their managers.

  35. Ms. Boxer:
    > This is why I am a fan of agile. At its core it is about the dynamic discovery of
    > requirements, and the design of an architecture that is not brittle. This, from what I can
    > see, best meets the reality of a constantly changing requirements space.

    The problem is that Agile doesn’t have any real meaning any more.

    I am currently working in a “Waterscrumfall” shop where we have daily standups, sprints, lessons learned sessions at the end of the sprints etc.

    Every single sprint to the release date has already been planned out. Even when I (the “Dev/Ops” guy[1]) point out stuff that the Super PMs missed, haven’t planned on, don’t know about (guys, invite Ops to your planning meetings. It’s cheaper that way).

    I’m a fan of what Agile is *supposed* to do. And I suppose somewhere it does.

    I shore wish that place was hiring though.

  36. [1] No, not that sort of Dev/Ops. The sort who gets to be ignored by the Devs when he says “this is how Ops works” and the sort who the ops team listens to, but then does what Management says anyway, and is *SHOCKED* when it falls over.

  37. >Paul Brinkley on
    > if you think of agile as not so much a methodology for managing programmers and rather as a methodology for managing their managers.

    This is a valuable point, however, it doesn’t capture the whole thing. Another extremely important aspect of agile is disciplined delivery — the forced closure points where a whole delivery is made at the end of each cycle, every two weeks for example.

    This is crucially important to programmers because it deals with that old aphorism “The first 90% takes 90% of the time and the last 10% takes the other 90% of the time.”

    Moreover the much ignored principles of refactoring — high coverage automated regression testing — is a crucial barrier for programmers to meet. By insisting on this programmers are put in a position to do much better unit testing and produce higher code.

  38. > William O. B’Livion
    > The problem is that Agile doesn’t have any real meaning any more.

    Yes, William I agree with that broadly speaking. However, even if the plan is laid out to the end, there is a process for adapting at least. Which is a step in the right direction. And even the use of phoney artifacts, such as daily stand ups, do make a positive difference. And again, although much ignored, regular deliveries, whether they are used correctly, do have a lot of benefits firstly by forcing delivery (even if 50% of them are missed and blended into the following) and it also gives constant feedback from users, even if informally, which offers the benefits of preventing projects from going too far off into the weeds.

  39. nht on 2016-08-19 at 18:15:18 said:
    > Good coders trump good processes.

    Given my previous interactions with you I expected you to give me shit about the whole “documentation costs more than its weight in gold and is worth less than its weight in poop.”

    Perhaps it was funny enough that what I imagine you believe is its lack of veracity was compensated by its humor?

    LOL.

  40. Some documentation does cost more… and is worth less… and I thought that quip was funny. (But I’m not nht, so.)

    Most of my exposure to agile comes from employers, as opposed to programmers, so that probably informs a lot of my thinking about it. It’s clear to me that they’re learning it in fits and starts. The first lesson seems to be releasing early and often; a large software project is broken into many smaller ones, and that’s easy to digest, so they do that. A few have gotten as far as scrums. They haven’t been good at those. (One shop out of a few dozen I’ve seen managed to do them standing up without anyone having to sit down, and it still took over half an hour.) Lots of lip service. Like I said: fits and starts.

    It hurts the programmers too. This kinda mindset has been going on long enough to percolate down to that level, so there’s enough clock punchers there to make it hard to get something going. In a few places, hiring even (inadvertently) encourages that behavior. (Most of this is in the public sector. That’s probably what I should have started with.)

    This means William O’s not wrong either; a lot of people say “agile” who don’t really know what it means. Hell, *I* don’t completely know what it’s all about, but I’m just here to build something, so my support of agile is contingent on that. If some shop says agile and half-asses it and still manages to put something together on their budget, I’m may not be thrilled with that, but I am at least okay with that.

    And of course, in the larger scheme, a lot of people espouse all sorts of labels and don’t quite live up to someone else’s definition of it.

    1. >And of course, in the larger scheme, a lot of people espouse all sorts of labels and don’t quite live up to someone else’s definition of it.

      Alas, the problem is not as innocent as a failure of understanding.

      Agile is fundamentally incompatible with the illusion of control that so many managers crave above all else. So there’s a kind of constant political pressure felt by people selling it and developers doing it, to bend it onto something almost but not completely unlike what it ought to be, often preserving the superficial features while emptying them of any actual meaning.

      Whenever I hear that a developer who’s been through agile hated it, on closer inspection the process always turns out to have been this kind of parody of agile. I need to qualify this by observing that I don’t think I ever talk to developers far enough down the bell curve to hate correctly-done agile because it left them floundering; that is another plausible reason.

      I also don’t want to leave the impression I think agile properly done is a panacea. The aphorism about agile being appropriate only when flag days are cheap is the only one I didn’t write – Daniel Franke threw it in the pot – but I agree with it. The problem with processes that are all a fast loop of trials and corrections is that is that while they’re good for preserving adaptive flexibility, they can also leave you stuck in a local optimum in design space, unable to climb out because you can’t see a better optimum at distance and couldn’t organize to get there if you did.

      (While I’m crediting where credit is due, the last one with the burning barrels is loosely based on something Mark Atwood muttered about software development at HPE.)

  41. @esr
    > Agile is fundamentally incompatible with the illusion of control that so many managers crave above all else.

    I think you are basically right here Eric, but I also think that is only half the story. The truth is that agile actually gives program managers vastly larger amounts of actual control than they do in waterfall or other methodologies. Which is why the word “illusion” is so important.

    However, I have managed many software projects, and when done right agile has always helped me have much better insight into what is going on. Of course, your point is correct, not perfect control. But it gives way more visibility than the other techniques I have been involved with.

    And one thing that I particularly appreciate is that it is less brittle. With big software projects you can get in a situation where you either fail completely or succeed completely (though obviously that is a bit of a hyperbole.) Agile is much better at hitting the gray area in between. Which is of course far more realistic.

  42. One of things that I like about old-school Unix philosophy is that it almost forces development to be small scale, almost single-developer scale, and relegates large system design to the realm of gluing together components with well-understood semantics, rather than maintaining a Big Ball of Mud™.

    But if we’ve learned anything from Windows, it’s that enterprise developers and sysadmins (or their managers) prefer the unifying model of “components with well-defined, possibly async IPC interfaces” over processes, sockets, and pipes. And to keep up with modern developments this must become the unifying model for Linux also — hence D-Bus, hence systemd. Both of which make me throw up in my mouth a little, especially when they’re far enough past the manageable-complexity threshold to give Linux weird, Windows-like failure modes.

    I’m still not convinced that the async-IPC and structured interface model is not better overall. There is still no well-supported standardized way in Unix (except maybe Mac OS X and Solaris) to do the sort of async I/O that Windows devs consider as natural as breathing. And sorry, but PowerShell beats any Unix shell in terms of power, flexibility, and ease of use. Imagine DCL with tab completion so you don’t wear your fingers to stubs, and the ability to call into any component’s COM API and perform transformations and reports on the returned typed objects, right from the shell.

  43. ESR writes: The problem with processes that are all a fast loop of trials and corrections is that is that while they’re good for preserving adaptive flexibility, they can also leave you stuck in a local optimum in design space, unable to climb out because you can’t see a better optimum at distance and couldn’t organize to get there if you did.

    I’ve noticed this problem in a lot of places. It’s especially distressing because there does not appear to be any way to tell a local optimum from a true optimum without a Smart Person, and there isn’t always a way for a non-Smart Person (or in many cases, a person who is Smart, but not smart in the domain in question) to tell a Smart Person from a person who’s bullshitting you in order to justify sitting on a tough problem for the next umpty-ump cycles while drawing a paycheck. Or even a person who’s geniunely trying to be Smart but just isn’t finding the flash of insight needed to solve a specific tough problem.

    Which means people keep searching for that silver bullet methodology, that source of control.

    I should say that your quote about agile and flag days doesn’t quite make sense to me, even now, and I think it’s because I’m thinking of flag days as major rewrites. If major rewrites are easy, why not do those and dispense with agile? I thought agile encouraged infrequent major revisions punctuating an otherwise constant stream of smaller refinements…

    1. >there does not appear to be any way to tell a local optimum from a true optimum without a Smart Person and there isn’t always a way for a non-Smart Person (or in many cases, a person who is Smart, but not smart in the domain in question) to tell a Smart Person from a person who’s bullshitting you in order to justify sitting on a tough problem for the next umpty-ump cycles while drawing a paycheck. Or even a person who’s geniunely trying to be Smart but just isn’t finding the flash of insight needed to solve a specific tough problem.

      This is all true. Welcome to reality. It bites.

      Now consider how this mess looks from my point of view as one of the few who reliably can see long distances and around corners in design space. I know I have an immensely valuable ability, but I don’t get to use it nearly as often as would be efficient for everybody, nor do I get paid what would be its clearing price in a more rational world.

      Any why not? First, because managers seeking the illusion of control have trouble giving up actual control to people who can exercise it more effectively than they can, even when they know that. They’d rather, as you say, search for a silver-bullet methodology that preserves that illusion.

      Secondly, my market value is (in effect) lowered by the manager’s need to discount for (and protect himself from) the existence of fakers and timeservers that he has little ability to distinguish from the real thing. Generally speaking the only people who can accurately evaluate the capabilities of system architects are…other system architects.

      It’s quite frustrating.

      >I’m thinking of flag days as major rewrites.

      No, that’s not what a flag day is. Flag days can be caused by a change that is quite minor, but fails to preserve compatibility forward or backwards of it. Worst case you lose access to old data, or it’s garbled. The original flag day was a change in the ASCII character encoding.

  44. Much like “pair programming” almost always rapidly degenerates into a version of old-school Visual Basic’s “can’t move off a line without fixing your syntax error” feature, but implemented on a smelly human constantly looming over your shoulder, “agile” too often rapidly degenerates into a way to tie up programmers in busywork so they look more productive while distracting them from what they’re there to do which is programming, motherfucker. And browbeat them at performance review time for not being agile enough when they balk at the kafkaesque rules and requirements implemented by outside business consultants and prove to be insufficiently “team players”.

    I like the spirit behind agile, but as commonly implemented, any specific agile methodology has a good chance of being more a hindrance to a good team than a help.

  45. @Jeff Read: And sorry, but PowerShell beats any Unix shell in terms of power, flexibility, and ease of use. Imagine DCL with tab completion so you don’t wear your fingers to stubs, and the ability to call into any component’s COM API and perform transformations and reports on the returned typed objects, right from the shell.

    http://www.hanselman.com/blog/AnnouncingPowerShellOnLinuxPowerShellIsOpenSource.aspx

    https://github.com/PowerShell/PowerShell
    ______
    Dennis

  46. @Jeff Read
    > I like the spirit behind agile, but as commonly implemented, any specific agile methodology has a good chance of being more a hindrance to a good team than a help.

    I don’t agree at all. Of course it is poorly implemented, often so because it is implemented by people who are pretty clueless about what it is. But the reality, of what I have experience anyway, is that even poorly implemented it deals with the really dreadful situations that often prevail in development shops. In a couple of different ways:

    1. It forces a daily meeting to get people talking about what they are doing.

    2. It makes the management team make a list of short term, small chunk expectations. And provides a mechanism for developers to communicate about it, moving it around in a realistic manner.

    3. I provides a reasonable demand of cadence of delivery.

    4. It at least makes programmers feel guilty about not writing unit tests that they might actually do it a little.

    Not ideal, but am improvement. Of course properly implemented the benefits are better and more profound.

    Thing I think is interesting is that many of the principles underlying agile are derived from early Unix programming principles, perhaps codified differently. And I think that bespeaks again of the genius of those early guys.

  47. >I’ve noticed this problem in a lot of places. It’s especially distressing because there does not appear to be any way to tell a local optimum from a true optimum

    But an important principle here is that optimum architecture is often unknowable until you are done, because architecture ultimately proceeds from requirements, but requirements are often unknown in entirety until the end. We have surely all been involved in projects which had a beautiful architecture, until a last minute requirement came along that utterly underminded all our assumptions.

    My view is that the solution to this is really three pronged.

    1. Great architects that have design patterns embedded in their thinking (and I don’t mean the GoF principles, but just a general experience as to what works and what doesn’t.)

    2. The use of well know patterns that have that experience baked in.

    3. A design methodology that factors in reality that architecture needs to be changed. The greatest flaw in most agile processes is the poor implementation of this. At the heart of architectural refactoring is the ability to regression test both black box and white box with isolation of functionality (which means a properly implemented IoC architecture.)

    You need all these things. Like I said, great programmers are the key to successful projects. The mediocre ones are a wash and the bad ones should be sent home before they break something.

    Shame they are in such short supply. Which is actually an interesting question. Why is it that there are few really good architects? Nature or nurture?

    1. >Why is it that there are few really good architects? Nature or nurture?

      Both, I think. It’s not a capability anyone really knows how to teach, getting good at it takes a long time building on a lot of field experience, and you probably need some kind of at least mildly unusual neurological talent to be capable of the learning process in the first place. Parallel to theoretical mathematics, though I don’t think it’s the same talent.

  48. @Jeff Read: And sorry, but PowerShell beats any Unix shell in terms of power, flexibility, and ease of use. Imagine DCL with tab completion so you don’t wear your fingers to stubs

    Bash has context-aware tab completion support, zsh has it even better, and (mostly) built in.

    And isn’t using shell on MS Windows not popular because of horrific cost of fork on this system?

  49. @Jeff:
    >And sorry, but PowerShell beats any Unix shell in terms of power, flexibility, and ease of use. Imagine DCL with tab completion so you don’t wear your fingers to stubs…

    I’m sorry, but on the ease of use front:

    1) Everything I have encountered that is not zsh, Unix or non-Unix, has crap tab completion. Specifically, ambiguous completions are universally poorly handled: both PowerShell and cmd will cycle through possible completions with each press of the tab key, but won’t give you the whole list at once, bash gets it very nearly right, but throws a wrench in the works by giving nothing on the first tab press, then giving a list on the second. Only zsh does the right thing and gives the full list of potential completions the first time tab is pressed.

    2) PowerShell is way too verbose to be a comfortable environment without decent tab completion (which I understand was a problem with DCL too).

    1. >PowerShell is way too verbose to be a comfortable environment

      That’s a surface problem. The underlying, much more serious one is crappy discoverability.

      Yeah, passing around objects instead of text streams sounds great – until you grapple with the difference between text streams, where you can often figure out the format and semantics by eyeball, versus object blobs that give you no hope of this. Unless they’re well-enough documented (which is basically never) you’re screwed. You can’t compose what you can’t comprehend.

      True message-passing-OO languages like Smalltalk get around the discovery problem in various ways, including powerful introspection facilities and an essentially social agreement that all source code for the objects must be browseable. Lacking this, the ecology around PowerShell must essentially reduces you to either banging rocks together hoping you get a spark of the right color or relying on collections of canned recipes. The real advantage of shell-like environments – fruitful unexpected combinations – almost never materializes because the object interfaces are too opaque, specific, and brittle.

      How do I know this never having used PowerShell myself? Because it’s not a new idea. Systems like this get reinvented, usually in academia, fairly often. They look brilliant in doctoral theses. They never scale, and text-stream shells have outlasted them all, because their potential is smothered by discovery costs.

  50. > Why is it that there are few really good architects? Nature or nurture?

    The role of architect requires flexibility in social styles. This does not come naturally and may be impossible for many people.

    On the one hand the architect must be a humble learner, seeking domain experts and eliciting not only what they can easily articulate but also the implicit, unspoken knowledge the domain practitioners don’t even consciously notice as they do their work.

    On the other hand the architect must be a strong leader, able to command the respect of the developers. This often means facing down developers that would love to try fashionable new coding techniques or oversimplify the architecture to conform to some elegant Platonic ideal.

    Failure of the architect to be a humble learner in the domain space is often at the root of project failures blamed on “changing requirements”. Failure of the architect to cast a unifying vision is often at the root of project failures due to massive cost overruns.

    The exception to this rule is when the architect already deeply understands the domain space because the architect is also a consumer of the product. (E.g. RMS and EMACS.)

    But too many budding architects learn the wrong lessons from success in these special situations. They may never have had to seek understanding before, and their arrogance blinds them from realizing when they are out of their depth in the domain space. Result: disaster.

    1. >They may never have had to seek understanding before, and their arrogance blinds them from realizing when they are out of their depth in the domain space. Result: disaster.

      NTPsec could have been this kind of disaster. Fortunately, (a) the domain knowledge it requires is so arcane that I couldn’t have fooled myself into thinking I knew it all if I wanted to, and (b) I’m not disposed to fool myself that way to begin with.

      I just wrote an article for Linux Journal (pub probably in October) in which I explain in detail how the main driver of the project’s technical strategy was my knowledge of my own ignorance. I had to bet that massively reducing the attack surface through pure software engineering would be enough, because I didn’t know enough to do anything else.

  51. Parallel said:

    The exception to this rule is when the architect already deeply understands the domain space because the architect is also a consumer of the product. (E.g. RMS and EMACS.)

    This reminds me of an observation Paul Graham made in a short piece titled “Java’s Cover”:

    Historically, languages designed for other people to use have been bad: Cobol, PL/I, Pascal, Ada, C++. The good languages have been those that were designed for their own creators: C, Perl, Smalltalk, Lisp.

    Is this phenomenon common enough in IT – and perhaps other fields – to consider it a pattern?

    (Incidentally, the article also reflects Jessica Boxer’s concern about “too many cooks”.)

  52. @jessica

    Given my previous interactions with you I expected you to give me shit about the whole “documentation costs more than its weight in gold and is worth less than its weight in poop.”

    Perhaps it was funny enough that what I imagine you believe is its lack of veracity was compensated by its humor?

    Why would I give you shit about something I agree with (humorous wording or not)?

    I think I have always described (and dated) myself when I state I’m on the James Bach side of good enough software and a proponent of DeMarco and Lister’s PeopleWare.

    But not for large complex projects…software or bridges.

  53. > A Roman project manager would look at the root cause lists above and nod his head and say “Yep, I was telling Maximus the other day that the &*(^&*)^ customer didn’t have a clue what he wanted and kept changing the damn specs. What the heck? He wants to have BOATS in the Collesium now? That’s going to be one beast of a change order.”

    That Roman project manager bears much of the responsibility if he didn’t take the time to survey the body of scripts and production techniques for which the new venue was likely to be used. As an empire based around the Mediterranean sea, he should have anticipated marine stories would be part of the cultural oeuvre.

    So yes, this is a larger pattern. There must be at least one person who viscerally understands both the user domain and the development process. Otherwise nobody is competent to evaluate and defend rational tradeoffs in the union of both possibility spaces.

  54. @ESR

    On the subject of pending essays, is there any word on the HBD post?

    @Jorge

    Ah good! I was beginning to worry that you had been scared off.

    1. >On the subject of pending essays, is there any word on the HBD post?

      Needs a rewrite before I’ll consuder shipping it.

  55. @esr
    > That’s a surface problem. The underlying, much more serious one is crappy discoverability.

    I’m not much of an expert on powershell, but I do know a lot about the .NET platform that it is based on, and the fact is that this type of discoverability, reflection, is baked into the heart of the platform. So I don’t much must powershell so I am not familiar with its debugging capabilities, but it is certainly feasible to have a very rich view of the interacting data streams.

    Having said that I did have one experience recently that gave me pause to reconsider my thoughts on this. I have this project that involves spreadhsheets being emailed in (I have long contended that much of American business runs not on Windows or Linux, but on Excel…) However, we have had constant problems with edits from multiple sources getting lost, because our tools (git in this case) doesn’t do well diffing these big blobs.

    Of course, you could argue, and I have argued, that that is a fault of git, which has an architecture based on the premise that it will be managing text files, and isn’t really designed to handle non text well (perhaps a plug in architecture where Excel provides a diff engine). But the reality of this bites, for sure, especially since doing stuff in Excel is already a bit of a pain in the ass. Do-overs ain’t much fun.

    1. >the fact is that this type of discoverability, reflection, is baked into the heart of the [NET] platform

      That might be true without being helpful, if PowerShell doesn’t have easy access to the reflection features.

      I’m basing my pan of it partly on some on-line comments I’ve seen describing the ecology round PowerShell as opaque blobs. So at least some people with experience don’t think discoverability is working.

  56. @esr: I’m basing my pan pf it partly on some on-line comments I’ve seen describing the ecology round PowerShell as opaque blobs. So at least some people with experience don’t think discoverability is working.

    The keyword is “opaque” above. They’re opaque because those complaining don’t know what the blobs are. All the discovery in the world won’t help you if you don’t know what you’ve discovered.

    Text based shell like bash can suffer from similar issues if you lack the background knowledge the shell implicitly assumes you have.
    ______
    Dennis

  57. @JEssica Boxer: Of course, you could argue, and I have argued, that that is a fault of git, which has an architecture based on the premise that it will be managing text files, and isn’t really designed to handle non text well (perhaps a plug in architecture where Excel provides a diff engine). But the reality of this bites, for sure, especially since doing stuff in Excel is already a bit of a pain in the ass. Do-overs ain’t much fun.

    Years back, there was an outfit called Component Software that offered a version of RCS for Windows with a diff that could operate on binary Word and Excel files. They aren’t around any longer, but it was a worthy attempt.

    git is the most recent in a line of VCS’s that derived from things like sccs and RCS, whose implicit assumption was that you were dealing with versions of source code that would be in text format. That’s still the dominant assumption in such tools. Blaming git because it doesn’t do well at versioning binary blobs when it was never designed to strikes me as blaming the tool when you are using the wrong tool for the job.

    Now that Word and Excel use a flavor of XML as the underlying storage medium unless you specify otherwise, I’d expect git to be able to do something with it, though I’ve no idea if it would be useful.

    Excel 2010 and 2013 have a Track Changes feature akin to Word’s. Are they being used by those sending the spreadsheets?

    It’s not clear to me git should be in this loop.
    ______
    Dennis

    1. >It’s not clear to me git should be in this loop.

      On the other hand, I don’t know of any VCS that deals gracefully with cases like this, so others would probably suck worse. Problems like how you browse – or even define – a diff get trickier when the data is not structures as text lines.

  58. Jessica is right.

    COM and .NET both have discoverability and reflection as a fundamental principle of the platform. It’s why we even have a COM. Yes, the reflection is accessible from PowerShell.

    The “opaque blobs” you may have heard of are probably the cmdlets. PowerShell works more like DCL than bash; to extend the shell, you write a class in a .NET language like C# that conforms to the cmdlet interface. Cmdlets are no more opaque than any other .NET assembly (they support full reflection).

    What’s really intriguing is that command line parsing, output formatting, and error reporting are all handled by the shell; you don’t even have to call anything to get your command line arguments. They are dependency-injected right into the cmdlet when the shell launches it. Thus — again much like DCL — the shell is smart enough to know which parameters a cmdlet takes, and can provide help and even tab completion for them.

    Enterprise developers have stuck with the Microsoft platform and toolset lo these many years because in many respects it really is leagues ahead of Unix. Bitter pill to swallow, I know.

  59. @esr: >It’s not clear to me git should be in this loop.

    On the other hand, I don’t know of any VCS that deals gracefully with cases like this, so others would probably suck worse. Problems like how you browse – or even define – a diff get trickier when the data is not structures as text lines.
    I’m not aware of one either.

    You can make a case that one should be able to, though precisely how you do that will depends on the binary blob, and that gets ugly fast.

    Jessica thought of a plugin that could deal with Excel files called from git, which is one possible method.

    My question is just what changes need to be tracked? If the answer is “numbers recorded in the spreadsheet”, another approach might be a script that dumped the Excel file in a format that could be diffed.

    For that matter, instead of slinging spreadsheets around, I’d investigate cloud based versions where multiple folks could work on the same sheet collaboratively. (Google Sheets can do this, and track changes made by collaborators, but it might not be sufficient for what Jessica needs.)
    ______
    Dennis

    1. >My question is just what changes need to be tracked? If the answer is “numbers recorded in the spreadsheet”, another approach might be a script that dumped the Excel file in a format that could be diffed.

      An even better approach would be to use a spreadsheet that natively stores its data and cell formulas in a textual format, ideally one in which at least a portion of the line structure corresponds to data rows. With modern computers we are long past the point where this would impose any serious time or space overhead.

      If this were done, version-control diffs would bear an overall relationship to the content that is actually scrutable to humans. Score another win for textual formats!

  60. > Git is the most recent in a line of VCS’s that derived from things like sccs and RCS, whose implicit assumption was that you were dealing with versions of source code that would be in text format. That’s still the dominant assumption in such tools. Blaming git because it doesn’t do well at versioning binary blobs when it was never designed to strikes me as blaming the tool when you are using the wrong tool for the job.
    >
    > Now that Word and Excel use a flavor of XML as the underlying storage medium unless you specify otherwise, I’d expect git to be able to do something with it, though I’ve no idea if it would be useful.

    Git has something called diff and merge drivers, allowing for an external program (like XML diff, or XML merge, plus unzip) to calculate differences, or provide a 3-way merge for file contents. Assuming that one uses XML-based XLSX, and not binary XLS, though even then you can write diff and merge drivers.

    > My question is just what changes need to be tracked? If the answer is “numbers recorded in the spreadsheet”, another approach might be a script that dumped the Excel file in a format that could be diffed.

    A simpler solution would be to use ‘textconv’ driver to convert file to textual format before diffing, for example converting to CSV, or Ethercalc text (?) format.

    > For that matter, instead of slinging spreadsheets around, I’d investigate cloud based versions where multiple folks could work on the same sheet collaboratively. (Google Sheets can do this, and track changes made by collaborators, but it might not be sufficient for what Jessica needs.)

    And there is Ethercalc… but I guess the problem is the use of Excel macros, isn’t it?

    > Excel 2010 and 2013 have a Track Changes feature akin to Word’s. Are they being used by those sending the spreadsheets?

    As far as I know Track Changes features, be it in MS Office, or LibreOffice, deal almost purely with linear history. That’s the easy parts of version control. (BTW. I like the MS Office solution better, as it describes changes textually, including formatting only changes).

  61. It has been written several times above that the problems in managing software projects is not in the coding that has to be done, but in finding out what has to be done.

    I would like to add an anecdotal illustration.

    A decade ago I got involved in the development of an educational application (accidentally, as someone misdialed a phone number). After the end of the project I kept maintaining it as a hobby. I ported it to a new platform to cure the build-guru disease.

    Early this year a user wanted a mobile version and had some professional perform a cost analysis for porting. The outcome was an estimated 800 person hours. Too much so the effort was abandoned.

    I decided it had to happen anyway and constructed a functioning web-app in 10% of the time.

    The difference between the professional estimate and my actual time spend is not due to my L337 skills (I am mediocre at best), but the fact that I only had to translate the old code into Javascript. Writing the code is easy, designing the stuff is the hard part.

  62. Folks that use excel for business tend to have an bunch of macros, scripts and whatnot. Google sheets typically won’t handle those same use cases (yet).

    You can argue that it’s the wrong approach but it’s one that non-programmers can do to get fairly significant functionality.

    Presumably Google sheets will get to the same point eventually…although I wouldn’t want my business sensitive info in someone else’s cloud.

    Share point will do some rudimentary revision control for you but it’s using the old locking paradigm unless the most recent version is improved. If anyone can do a merge of excel it would be MS.

  63. @nht
    > Folks that use excel for business tend to have an bunch of macros, scripts and whatnot. Google sheets typically won’t handle those same use cases (yet).

    I guess there is an implication in your comment that you know this but… Google sheets does have an automation model similar to Excel where you can automate it with some Javascript functions. I’ve played around with it a little, and it is cool, though I don’t know how close it can match the extensive control possible with VBA. Plus is it probably not so end user friendly.

    > You can argue that it’s the wrong approach but it’s one that non-programmers can do to get fairly significant functionality.

    I agree, it is not very lovely for computer experts, but it is extremely empowering to users. They get themselves in all kinds of shit, but they fix it manually, and they are ok with that. I had a team of accountants that did various reconciliations in Excel (using reports dumped from the truly awful SAP) and they actually hired a temp 30 hours a week to go in and clean up the presentation. I spent about ten hours writing a little set of toolbar macros that did it for them in an instant. It was like magic to them. I was seriously concerned they were going to thank me, and then burn me as a witch.

    >although I wouldn’t want my business sensitive info in someone else’s cloud.

    Plus in the specific use case I was referring to the problem is that, despite what we might hear, the web isn’t everywhere.

    > Share point will do some rudimentary revision control

    MS’s SCCS TFS manages spreadsheet differences well. However, it has plenty of other compensating faults.

  64. @jessica agree, it is not very lovely for computer experts, but it is extremely empowering to users. They get themselves in all kinds of shit, but they fix it manually, and they are ok with that. I had a team of accountants that did various reconciliations in Excel (using reports dumped from the truly awful SAP) and they actually hired a temp 30 hours a week to go in and clean up the presentation. I spent about ten hours writing a little set of toolbar macros that did it for them in an instant. It was like magic to them. I was seriously concerned they were going to thank me, and then burn me as a witch.

    The temp might want to burn you as a witch. :)

    Part of the issue with Google as a replacement for Excel is that a lot of it IS black magic to the users but it works. They understand the macro/script/etc enough to use or tweak them but not replicate that functionality in Sheets. Also there’s a whole bunch of arcane stuff getting things out of their enterprise back ends into excel that would need to get replicated.

    You really need the CIO to pony up the money to do a full change out of their business process from MS Office to Google Docs. Something some small firms can easily and quickly do if they want to but anything complex will cost money and incur risk.

    My casual impression is that Google is only trying to capture low hanging fruit in business and maybe push harder in the edu market to grow market share in the future.

    The problem with Google is guessing if investing heavily into their infrastructure is all that wise. The net is still littered with dead links for Google Code projects that are zombies and you aren’t quite sure which fork on github is really the one you want to use. Microsoft has a lot of faults but CodePlex is still around which launched around the same time as Google Code.

    Google Docs isn’t critical to Google…it’s part of the “Other” revenue stream that is around 10% of the total and includes Chromecast, Play and other fairly big revenue generators.

    MS Office is critical to MS.

  65. @nht
    > The temp might want to burn you as a witch. :)

    Maybe, but can you imagine spending 30 hours a week just adjusting column widths, changing fonts and colors, changing the number of decimals, and adding total lines, over and over and over again to almost identical spreadsheets you didn’t really understand. All. Day. Every. Day.

    I like to think I free-ed her to live a better life :-)

  66. @jessica What you didn’t know was she had the same macros and was playing Candy Crush for 29 hours 45 min…

  67. @Parallel

    So yes, this is a larger pattern. There must be at least one person who viscerally understands both the user domain and the development process.

    Thanks. This seems related to dispersed knowledge and the “invisible hand” that turns individuals’ pursuit of self-interest into collective gain (except when such pursuit involves fraud or the initiation of force, naturally).

    @FooQuuxman

    On the subject of pending essays, is there any word on the HBD post?

    I was wondering the same. Upthread he used te verb “to grapple”, which reminded me of the announced “Grappling with HBD” post. I, too, look forward to it (and to the post on Austrian economics’ shortcomings you suggested, heh).

    Ah good! I was beginning to worry that you had been scared off.

    Thanks for worrying, but what do you mean?

  68. @jakub interesting that you got continuous delivery as the key takeaway out of that where I saw “Make Users Awesome” as the key take away. Followed by “Make Safety a Prerequisite”.

    I disagree with the above sentiment that code is what you deliver as opposed to a product that “makes users awesome” which has always been the core of agile. It is also what drives the success of major software ecosystems.

    Continuous delivery is like CM. Something your dev environment provides as a matter of course regardless of methodology. The technological CI/CD solutions support any process although ones that incorporate lots of unit and regression tests do better.

  69. @nht
    @jakub interesting that you got continuous delivery as the key takeaway

    FWIW, I really hated that article, I found the patronizing tone of “you guys are all doing it wrong” very offputting.

    Having said that, although obviously making users awesome is key, users don’t get to be awesome if they don’t get any code. And often you don’t know if they are going to be awesome until they have the code for a little while. So rapid delivery is, in my view, the most important thing (though I am as dubious about “continuous delivery” as I am about “continuous integration”, I think closure points are important and they should be measured in days not seconds.)

    In a sense putting “make users awesome” as your primary is to fall into the very trap that makes agile necessary, namely the idea that we, programmers, business analysts or whatever you call yourself, can somehow ahead of time either by revelation or by preliminary user testing determine what users want. Certainly those things help, but there is no user test so effective as users using working software, event if that software is somewhat weak.

    The key is frequent delivery and a methodology that allows for dramatic code adaption in response to frequent review, along with a clearly articulated and articulable set of outstanding tasks, equally flexible to adaption. Frequent delivery requires a frictionless delivery mechanism, and the ability to dramatically alter code requires that much neglected subject of comprehensive regression testing, since that is the basic prerequisite to refactoring.

    However, one thing I did agree with in the snooty article was that the MBA-ization of Agile that he talked about at the beginning, is a nightmare, and something Eric referred to in OP and comments — the illusion of control.

  70. Mine: everything ever written about software teaches you how to scale up. If you need to scale down, you are pretty much on your own.

    I.e. when you are doing the exact opposite of those million dollar projects, you are writing code two users will execute twice a month and save 5 hours of work each time with it, and thus you really should not invest more than 6-8 hours of work into making it, then nobody tells you what elements of good practice (from unit testing to having separate development, testing and live environments to documentation to version control) can you omit in order to save time. Surely you want less time spent on overhead vs. time spent on working code ratio than in a real big project, but nobody tells you how much less. The even trickier question is: what elements of good practice can be omitted if there is surely only one or two such mini projects on a site but you will wish you had not omitted them if you happen to end up with a hundred.

    To link to what Jessica wrote: making those SAP reports dump data in less awful and more reconcilable ways qualifies as a good example of these mini projects, each. Likely the customizations, if done (as Jessica writes about a case when it wasn’t done, the accountants did not demand the customization done, which shocks me) there are likely some good practices missing. Likely you don’t find any more documentation than a one-line comment, you never find out who ordered it and why, there are not test cases, no protocols, no sign-off, no design document, no nothing and if you are “lucky” it is done all right on the live system. If you are lucky someone finds you an old email that says “I want X” and then a reply “done” and that is your documentation. But then again, how big is big enough to worth it?

    Scaling up is a science, scaling down is an art.

  71. @nht @Yaroslav

    >>Massively late, massively over budget and then cancelled.

    >Please, do, by all means.

    Except for the “then cancelled” part, Wembley stadium, Berlin airport – the later becoming an absolute joke now. I suppose construction is harder to cancel, due to sunken costs – that is a fairly big difference.

    Construction _should_ be accurately plannable, after all a good quote – at least where I am from – lists every piece of work with how many man-hours that takes. And construction companies often pay hefty penalties for being late. Still, with those very large projects, like airports, they screw it up.

    I was taught PRINCE2 – mainly for software project management uses – by someone who used the very same methodology to study how construction projects like Heathrow Five got over budget and late and how to fix that.

  72. @Jakub

    >“There is a joke in science, “If we knew what we were doing, we wouldn’t call it ‘research’.” I have long wished we had an expression for a similar sentiment about programming. Because programming naturally, regularly, and just about always involves an encounter with the unknown, and having to reckon with what you didn’t reckon with.“

    I used to believe this – all those articles from Joel Spolsky about how software is not bricklaying and my experience matching it.

    Now I believe that in a certain sense, software and research is bricklaying. And that sense is the statistical. With enough experience, as in: decades of hands-on work (not managers), it is possible to have a clue about the average productivity of the average programmer or researcher, how many easy, medium and hard problems they can solve per month, and also to estimate how many easy, medium, and hard tasks a given project stage may contain. Estimates updated after every stage, agile style. OR one can compare to projects of the same kind in general. It is never accurate but with enough experience – NOT with software development in general but with the particular domain as well – and being a bit generous about adding contingency time, it is possible to rarely run out of it.

    I must emphasize it only works with domain experience. A guy who led payroll software projects in 10 jurisdictions can estimate the 11th. A researcher who led projects to create diagnosis methods for 10 kinds of very similar illnesses can estimate the 11th.

    It has to be routine before you can estimate. Once it becomes routine, it is almost like bricklaying. Including the part where it is boring. Including the part where sometimes people screw up the estimation of bricklaying, too.

    Being very vary of estimation is a sign of having an interesting career, doing many different things :) Focusing on a narrow and repetitive field, to become that big fish in the small pond (which does have certain advantages), even research can be routine. And estimable.

  73. > making those SAP reports dump data in less awful and more reconcilable ways

    I assume you have never worked with SAP then? The primary feature of SAP is its ability to generate consulting dollars. I have never seen a project, no matter how trivial, that costs less than 80 hours at offshore rates of $100, and onshore rates two or three times that.

    Me? I did the macros on my lunch break. (Yup, the accounting ladies even bought me bagels as a thank you. Carb filled, delicious, guilt inducing yumminess.)

    > the accountants did not demand the customization done, which shocks me

    Dude, have you worked in a corporate environment? How do they get budget approved to do that kind of thing?

  74. TheDividualist
    > I must emphasize it only works with domain experience.

    Ah, there’s your problem right there.

  75. The key is frequent delivery and a methodology that allows for dramatic code adaption in response to frequent review, along with a clearly articulated and articulable set of outstanding tasks, equally flexible to adaption.

    I think that there needs to be a desire to build a terrific product out of the gate and there’s a certain level of punting when you assume that you can fix it in the next release as part of your methodology. There should be a fundamental focus on UX and design that allows the product to “make the user awesome”.

    That’s a lot easier said than done. Even a company with a focus on UX like Apple gives us iTunes.

  76. TheDividualist
    “I.e. when you are doing the exact opposite of those million dollar projects, you are writing code two users will execute twice a month and save 5 hours of work each time with it, and thus you really should not invest more than 6-8 hours of work into making it, then nobody tells you what elements of good practice ”

    Very interesting problem. From my own experience, I see two important principles, my $0.02:

    1) Strictly limit development time to the “discounted” value of the saved time over the life-time of the application. Calculating the discounted value of the saved time is left as an excersize for the reader. Personally, I would spend upto 20 hours in your example, but my time is probably less expensive than yours.

    2) Go for absolute minimum LOC. That is, use they highest level of (macro) language you can use. This increases productivity and reduces debugging time. Run time and memory use are mostly irrelevant in such projects.

    Within these limits, you will have to fit design, coding, and testing yourself.

  77. @JessicaBoxer

    I did work with software packages similar to SAP (Oracle Financials, Microsoft Dynamics) and the smarter kind of corporations, not even very big ones, tended to hire in-house developers who would do the customization on a salary. Consultants will do it properly, because documentation etc. is also chargeable work. Salaried in-house developers are the type who are all too likely to cowboy it and add random hacks to the live system without documentation, testing etc. on the plus side, the temp never gets hired because the accountants are like “hey IT dept can you customize that report for me? not urgent , next week will do”.

    I don’t see what exactly is the problem with hiring people who have a decade long domain experience. In my view this tends to make things *smooth*. Why consider developers a fungible resource who can be hired to make anything from a medical software to a payroll software? Why not hire someone who is like “yes it is like my last 10 projects except A, B,C ?”

    It is like medicine. When I have problems with my spine, I go to a spine specialist. Yes, it is sort of boring to him or her that he deals with nothing but spines. But for me as the patient, it is a good thing? And he or she earns good money being the No. 1 spine specialist in the city.

    I get it, programmers typically want more variety. But that is a different sort of problem. Hopefully there will always be room for people who always want to explore new things, but the marker of an industry maturing is the existence of these really narrow kind of experts who can do only one trick but they get that one trick *right* including accurate estimation.

    After all, any medical diagnosis, the kind my doc does with my spine is, pretty literally, *research* and as such not that different from programming. It is an unknown every time, but with ten thousand cases behind ones back the patterns are rather clear.

  78. @Jessica

    Adding to my above comment: once a corporation – I don’t know the huge ones, I am talking in a 200-500 employees range – has in-house developers, in-house consultants the whole magic budget problem goes away. From that on the accountants feel *entitled* that the IT dept must provide them with a well customized system and in the worst cases they don’t even formulate the customization request as such but as a complaint or in the very worst case an accusation, as in, “the SAP software you provided me is not giving easy to use reports fix that ‘bug’ ASAP I do not consider it properly implemented for us yet” almost like a literal accusation of not doing your job properly. Granted it is my experience of corporations with a headcount of 200 in Mitteleuropa not 2000 in America but I think at some level this is universalizable: if only someone in-house is able to customize a software, not just external consultants, suddenly that person is personally responsible for it being extremely comfortable for everybody and any failure to that can be accused as a “bug” !

  79. TheDividualist wrote: I don’t see what exactly is the problem with hiring people who have a decade long domain experience.

    That’s easy. The problem is that they’re excellent, if you can find them. But what if you can’t? Cue the company who looks for months for a developer with N years of experience with technology X and K years of experience in domain Y. This is a well known horror story among business types; it’s why large corporations spend millions, and why small corporations can’t even do this.

    (It’s one of the problems my now former company had been trying to solve. Build a formal logical model of your domain, and a program that can turn slices of it into a database, a reasoning engine, and documentation for new hires. All they’d have to know is technology X.)

  80. @nht
    > I think that there needs to be a desire to build a terrific product out of the gate

    Yes, I’m with you 100% on that. My experience is that the belief we are building something awesome is the most powerful driver in developer motivation. It is why I always push to find time to “clean up the mess.” Programmers are simply much more productive in clean code with decent reliable patterns (from the UX down to the unit tests.) I don’t know if the broken windows philosophy works with policing but it definitely does with software development (though it is EXTREMELY rarely practices AFAICS.)

    It is also why those crucial closure points of biweekly product reviews with users are so important. They keep the developers on track to realize that real people use their stuff.

    And FWIW, I have to tell you that as a general rule I really dislike Apple products. I find their UX clunky, annoying and difficult. That is probably because I am not used to it, and I have a different set of usage patterns, but to me it certainly isn’t any more grandma friendly than Windows. The only Apple product I ever actually owned was an iPad mini that I won in a raffle. I found it extremely annoying to use, and was kind of happy when the power plug flaked out and I could justify buying an galaxy for all my Netflix needs.

  81. @TheDividualist
    > … tended to hire in-house developers who would do the customization on a salary Consultants will do it properly, because documentation etc. is also chargeable work.
    Salaried in-house developers are the type who are all too likely to cowboy it and add random hacks to the live system without documentation, testing etc.

    I think I disagree entirely with almost every word in this paragraph. Or, at least I would say that it is ENTIRELY the opposite of my experience.

    >“hey IT dept can you customize that report for me? not urgent , next week will do”.

    Except that everyone in the corporate world needs it NOW, and everyone in the corporate world knows that “next week” means never, unless your Vice President kicks up a fuss.

    You know it is funny, much as the great depression was not a failure of the American economy but a failure of the Fed and the fiat money system, so too budgeting and finance fails badly at this level in corporations. And so, just as during the great depression an alternative money system arose (I’ll trade you a chicken if you paint my house) so too in the corporate world there is an underground economy. Often focused on gifts of food, days off, favors and seating arrangements…. Those be the things the things that oil the wheels of a corporation.

  82. @jessica Love or hate Apple UX it wasn’t until Apple gave designers the same status as engineers that industry followed their success. Google UX was wandering the wilderness until Larry Page hired Duarte and gave him real power that could counter balance engineering and went to Material design (which you can also love or hate).

    And again, love or hate Apple UX they do spend a lot of time on polish which really makes a difference in the final feel of products. It’s glaring when they don’t.

  83. @Jessica

    Interestingly I haven’t seen this alternative economy. Since most employees focus on minimizing stress/conflicts, that is basically what they optimize for, perfecting their “nervous breakdown deflection shield”, when a request comes from another department, the question is always what is less stress for me, if I go on and do it, or if their boss talks to my boss and I have to explain what I am doing and why I don’t have capacity and it is still not sure I don’t have to do it. Especially if the other is a bit smart and evil and formulates the request not as a request but almost like an omission or mistake on my part that needs to be corrected ASAP – you probably seen that, almost anything people dislike can be called a “bug”. This is especially dangerous, due to the stolen coat mechanism.

    The stolen coat mechanism is a proverb around here. Once upon a time, a patron at a coffee house in Vienna had his expensive coat stolen. He made a lot of fuss, asked ever waiter, other patron, called the police who too interviewed witnesses etc. when he returned in later occasions he was remembered as “the guy whose coat was stolen” which through a game of telephone morphed into “the guy associated with that coat stealing scandal” then “the guy who had something to do with stealing a coat, maybe a thief, I dunno”. The moral of the proverb is that you don’t want bad sounding things associated with you even if you are innocent because human memory and gossip is a game of telephone. Similarly, if the other department is sufficiently smart and evil, and formulates every request as “fix this bug”, still explaining to your boss 10 times it is not a bug makes you “the guy / gal always accused of making mistakes”. Hence you rather not have your boss know it and just do the request anyway.

    An economic exchange is always based on the idea that refusing a deal has no cost, hence people need to offer something positive in return. If refusing has a cost, that is called a threat. In your experience, how does this work that people at corporate jobs can refuse important requests from other departments at no cost? There is no cost of their boss gossiping to your boss how unhelpful you are etc.?

    In my experience, refusing requests is something a very official, very bureaucratic, red tapey place can do, maybe really large ones, I don’t know, where job descriptions are very literally written down and meant seriously. In the more mid-range ones I know, it is far simpler, if Jane is the person hired because she can do X then she is the go to gal for X and nowhere it is written down who to help and whom not.

    The smart and evil method very frequently evolves into a war of attrition of the nerves – basically whoever has a less perfect nervous breakdown deflection shield loses. Accounting makes a mistake. If sufficiently smart and evil, accuses IT that it is their fault – it is a bug in SAP because why did it allow me to do that and fix it ASAP. They of course reject that. Emails go back and forth, passive aggressive, more and more bosses in CC. Who loses? Whoever has less tolerance for that nervous knot in the stomach, for that air full of tension, gives up, concedes defeat, and buys some period of peace at the price of extra work.

  84. @dividualist LOL, you’ve never pissed off an administrative assistant have you?

    There are many ways to say no without saying no. This is why “work to rule” strikes can be very effective.

    If your boss won’t give you top cover from an aggressive department, find a new boss. It’s not like good devs are unable to find jobs that don’t suck.

  85. > That’s a lot easier said than done. Even a company with a focus on UX like Apple gives us iTunes.

    I’m going to nominate “Easy using is cursed hard programming” as an aphorism.

  86. it’s that enterprise developers and sysadmins (or their managers) prefer the unifying model of “components with well-defined, possibly async IPC interfaces” over processes, sockets, and pipes.

    I agree with Jeff Read. His statement strongly matches my experience for several reasons.

    First, standard sockets and pipes do not work across machines. Much of today’s computing requires multi-machine capability, whether for reliability, performance, or whatever. The standard unix toolkit pretty much falls apart once you need to cross the machine boundary. You can (at best) pipe data over a network socket or ssh a command. But there’s no good standardized way to handle operations (eg. here’s the addresses of a set of videos I want you to recompress. Divide the workload across a dozen machines and let me know when you are done). Sure, you can write something that does this, but it isn’t straight-forward. And if you want to be multi-threaded you need to do more funky stuff.

    Second, there’s no standard way to represent messages or structured data. The closest might be Google’s Protocol Buffers, but these don’t handle all of the required cases, either.

    Third, standard sockets and pipes are mostly incompatible with data representations more recent than the 1970s. Relational databases generally hold more than 1 table. If you want to operate across this you need to start by collapsing the data down to a single table, usually with a SQL statement or something – it’s not straight-forward to do from the command line. There are finally a series of command-line tools that can be used to process complex XML documents and they’ve managed to shoehorn themselves into the shell reasonably well, but it clearly wasn’t easy.

    Fourth, tab-completion is awesome. Yet it isn’t available as a general feature in bash. You’d think that since just about everybody uses getopt for their command line arguments that we could be smart enough to handle complete syntax by now. Sadly, no. At least, not unless you are writing in PERL and add extra stuff to your bash profile.

    1. >First, standard sockets and pipes do not work across machines

      Um, Garrett, what are you smoking? I do inter-machine socket communication all the time.

      >Third, standard sockets and pipes are mostly incompatible with data representations more recent than the 1970s.

      It’s a plain byte stream, Garrett. That is compatible with everything – you can impose or simulate any crazy structure you like over it.

  87. Second, there’s no standard way to represent messages or structured data. The closest might be Google’s Protocol Buffers, but these don’t handle all of the required cases, either.

    How odd…I seem to recall that ASN.1 was the original layer 6 reference implementation in the OSI 7 layer model. I think that predates protocol buffers a wee bit.

    Third, standard sockets and pipes are mostly incompatible with data representations more recent than the 1970s.

    How odd then that even the latest protocols are layered above…sockets. We’re using Kafka which is reasonably leading edge in terms of big data messaging but NASA was doing this in 1983 and it wasn’t new then either. I know because I wrote a message bus protocol for the ASIST ground system on my first job…

    Good to know that sockets are incompatible to modern data representations…

  88. > It’s a plain byte stream, Garret. That is compatible with everything – you can impose or simulate any crazy structure you like over it.

    But there’s no standard for doing so. And nothing has emerged.

    1. >But there’s no standard for doing so. And nothing has emerged

      That’s because any one such standard would be optimal for some uses but pessimal for others. It’s a feature, not a bug.

      Go look up “end-to-end principle” and become enlightened.

  89. How odd then that even the latest protocols are layered above…sockets

    I think he meant using plain sockets and pipes for IPC. Those are like the assembly language of communication. As an application developer you almost never want to mess with it, you always want a higher level API that implements abstractions such as RPC, message queue, pub/sub, etc. And the data that gets passed should be structured and typed.

    Sockets and pipes are fine primitives, but they suck as an integration model. And plain-text streams are on the way out, because son of a gun, structured binary beats them every single time. HTTP/2 is seeing rapid uptake, and the kernel just got a binary interface to the stuff in /proc. At scale, the cost of parsing stuff out of text streams is too great to bear, and scale comes around a lot faster than you think.

  90. @Jeff Read
    > you always want a higher level API

    Always? Wow that a pretty bold claim. To me it is plain that the existence of a hammer does not make all problems nails, but that doesn’t mean there aren’t any nails needing hammering. Pipes are excellent as a quick, simple, easily debuggable solution that doesn’t require a complex infrastructure build.

    > And plain-text streams are on the way out, because son of a gun, structured binary beats them every single time.

    There are few that have been more of a critic of “text only” here than me. but that is plainly ridiculous. Plain text is used everywhere with great success. My message to you is coming over just such a stream.

    > At scale, the cost of parsing stuff out of text streams is too great to bear, and scale comes around a lot faster than you think.

    The computational cost of parsing is negligible, and irrespective of what you do you have to have some sort of way to convert a structure into a flat stream of bytes. The smart programmer in nearly every instance should choose simplicity for the brain of the person looking at the software over saving a few cycles (or a few millions of cycles) of CPU churn, Of course there are exceptions, but complexity and rat’s nest code comes even quicker than scale.

  91. The computational cost of parsing is negligible, and irrespective of what you do you have to have some sort of way to convert a structure into a flat stream of bytes. The smart programmer in nearly every instance should choose simplicity for the brain of the person looking at the software over saving a few cycles (or a few millions of cycles) of CPU churn, Of course there are exceptions, but complexity and rat’s nest code comes even quicker than scale.

    Worth noting is the Netflix approach, which I think we can agree qualifies as “at scale”. They’re running most of their individual components as microservices in completely separate VMs communicating via JSON over HTTP. A single public API request can get converted into dozens of HTTP calls, each with its own serialization and deserialization—and this is what many large organizations are converting to because of the inspectability and decoupling.

  92. > That’s because any one such standard would be optimal for some uses but pessimal for others. It’s a feature, not a bug.

    When I said nothing has emerged as a standard, I didn’t just mean on Linux. Powershell is useful for what it is, but it’s an isolated ecosystem. It’s best considered as a high-level object-oriented language that happens to use pipe syntax for streaming operations (such as map, filter, reduce) which allows some of the same mental machinery that works with traditional shell commands and pipes to be used for this.

    On the other hand, that is useful for what it is. The strength over Unix is that this allows you to sort by / filter by data that is not in a convenient form (or not present at all) in the eventual textual output. Can you get the list of all processes using above a certain amount of memory using only ps and grep? There’s no “where vsz > 50000” primitive, whereas the powershell equivalent of ps, presumably, returns an object that has a field that represents the memory usage.

    The best Unix can do is provide ad-hoc options in every single utility to handle sorting and filtering by (and displaying) an often limited set of fields and display formats that they imagine you may want to use, which seems like the opposite of the Unix philosophy. Much is made of how ridiculous it is how many options “ls” has, but how the heck else can you sort by date?

    For all that “anything can work on byte streams”, sort and grep only work on text lines, and a textual representation that would make them work well for the kinds of things people want to do with, say, files and process, would be ridiculously ugly to actually *read*. The ability to pick an arbitrary display format for files rather than the three or so that ls supports is locked away in ‘find’ (which is incapable of sorting)

  93. Always? Wow that a pretty bold claim. To me it is plain that the existence of a hammer does not make all problems nails, but that doesn’t mean there aren’t any nails needing hammering. Pipes are excellent as a quick, simple, easily debuggable solution that doesn’t require a complex infrastructure build.

    As an app developer — yes, always! Which is easier — to call an API that’s properly integrated into the host language’s type and exception system or to spawn off a new process, connect its standard input and output to file descriptors, feed it input, and handle all the errors? Even if the API is inplemented by piping into a separate process, a) that’s hacky as all get-out and b) the programmer is supposed to never need to see or know that.

    A component with an introspectable typed API can also be quick, simple, and easily debuggable. With frameworks like Twisted Python’s Perspective Broker, you need not even write any code to expose a class’s API over the wire; the framework introspects the class and generates the right handlers for incoming RPC calls. And I don’t think I need to tell you about what’s possible to rapidly build using VBA macros in Excel and COM automation. There’s a reason why this stuff won, and won big.

    As a sysadmin, piping may have some uses, but even there well-typed, auditable APIs have an advantage. You should know this — look at PowerShell. It has a piping operator, but that has a lot more in common with Ruby’s each method than it does Unix I/O pipes — it specifies a query or invocation to be applied across a collection of objects. Sysadmin tasks in PS are accomplished by calling into the system’s various exposed API endpoints — not by writing config files or modifying them with ‘sed’ or the like.

    The computational cost of parsing is negligible

    No, it’s not. There was a 10x speedup in performance when using the experimental ‘task_diag’ binary interface compared to using just /proc.

    Worth noting is the Netflix approach, which I think we can agree qualifies as “at scale”. They’re running most of their individual components as microservices in completely separate VMs communicating via JSON over HTTP.

    Yes, microservices are great — but there is considerable infrastructure and layers above sockets and HTTP that go into a microservice infrastructure in order to make it look like an API call to the application developers. Typically there will be a service registry and some sort of reverse proxy that routes calls to the appropriate service on the appropriate box. And Google’s services — micro or otherwise — communicate with each other using protobufs, not JSON, because — again — the parsing overhead is too much. Why do you think HTTP/2 is being taken up so rapidly? The overhead in parsing HTTP/1.1 (and difficulty in multiplexing many streams over what is fundamentally a text-based protocol) is a drag on performance.

    It wouldn’t surprise me at this point to see HTML6 be some sort of binary container format, akin to Microsoft Blackbird except developed by Google or Facebook or somebody, with a VM bytecode homomorphic to asm.js for scripting. The bandwidth savings in JavaScript junk downloaded to make a typical rich Web app run would be enough to justify the transition, eapecially among mobile users.

    Again, as long as the format is open, with open source tools to destructure and examine data in the format, the impact on developers compared to plain text is negligible, with huge savings on the machine side.

  94. @Jeff Read
    > No, it’s not. There was a 10x speedup in performance when using the experimental ‘task_diag’ binary interface

    And what was the impact on users of this super duper speed change? How did it change their lives for the better? How much do you expect to reduce hardware costs by implementing this? Who gains one iota from this change?

    Expending effort to save unused CPU cycles (and the vast majority of CPU cycles are unused) is a net negative.

  95. @nht: Yes. I recall ASN.1 notation (which few people use explicitly, for a reason). I also recall CORBA which has its own problems. I should have remembered standard C RPCs. Usable but certainly not elegant, but at least we’re moving up to the 80’s! Thanks for the reference to Kafka – I will need to read more.

    @esr “That’s because any one such standard would be optimal for some uses but pessimal for others. It’s a feature, not a bug.” True and irrelevant. The same argument applies to programming languages. And yet we’ve managed to standardize on less than a dozen for nearly a development. That there might be advantages or disadvantages to any particular implementation fails to explain why we would have have a lack of a small handful of solid, well-tested, supported, commonly-used APIs available. The end-to-end principle is separate from whether a user needs to re-implement everything from scratch every time.

    @esr, Jeff Read: “I think he meant using plain sockets and pipes for IPC”. Yes, I did. Thank you for catching that.

    @Jeff Read: “Sockets and pipes are fine primitives, but they suck as an integration model”. Absolutely!

    “And plain-text streams are on the way out, because son of a gun, structured binary beats them every single time.”

    I disagree with the qualifier “every”. Configuration files stored in text files can/could be great (I have a rant about this), mostly because they are typically read only on process startup and are designed to be a human-to-computer communications channel. Where it gets to be silly is when you are dealing with computer-to-computer communications. There the common case is humans not looking at the contents. What a good binary format needs is a solid set of debugging/GDB tools to allow the binary data to be examined when needed, and otherwise not deal with the overhead.

    @Christopher Smith: “separate VMs communicating via JSON over HTTP”. This has a lot of advantages. It does have down-sides. Most notably, no way to automatically validate messages (such as with an XML DTD), and processing overhead. The processing overhead may or may not matter, depending upon the latency requirements of the service. When you counting the microseconds to handle a complete request, that kind of parsing isn’t always practical.

    1. >That there might be advantages or disadvantages to any particular implementation fails to explain why we would have have a lack of a small handful of solid, well-tested, supported, commonly-used APIs available.

      I think, in the sense you now seem to mean, we actually do. See earlier remarks about new-school webapps being written as microservices passing around JSON over sockets. I know this works because GPSD uses the same design pattern.

  96. @Winter

    Meant to comment on this earlier; sorry for postponing it.

    A decade ago I got involved in the development of an educational application (accidentally, as someone misdialed a phone number).

    A favorable accident! Nice. :-)

    I only had to translate the old code into Javascript.

    Two questions:
    1. What language was the old code in?
    2. So you used JavaScript for server-side programming, perhaps with a framework such as Node.js?

  97. @Garrett: “Where it gets to be silly is when you are dealing with computer-to-computer communications.”

    I think the design pattern behind a lot of Unix tools is making the tool agnostic as to whether it’s communicating with a human or a computer. It might be running in a script, or it might be running interactively.

  98. Jon Brase,

    This can be achieved by building smarts into the shell that know how to destructure the structured binary output for human perusal when it is to be displayed on a human facing TTY.

    Hint: This is how PowerShell works.

  99. @Jorge Dujan
    “Two questions:
    1. What language was the old code in?”

    An idiosyncratic scripting language of a monolithic application.

    “2. So you used JavaScript for server-side programming, perhaps with a framework such as Node.js?”

    No, it was a full client side web app. It ran locally in the browser without needing any server interaction. Else I would not have used Js.

  100. @Jorge Dujan
    “Two questions:”

    Why these design choices:
    1) We needed sound recording and sophisticated signal analysis tools and something that could run on all three platforms. Including speech synthesis. And I was not going to code this myself. This monolithic application delivered all this and had a Turing complete scripting language. I could write the whole application once and it would run on Windows, Mac, and *nix. (And this actually worked “write once run everywhere”)

    2) But as time goes on, education moves to mobile. Akso, it became clear that people were becoming unwilling to install applications and platforms started to implement policies to discourage users from installing unsigned applications.

    I was not willing to write native apps for iOS or Android and the application would not run on mobile. Last year, WebRTC finally took off and html pages could record and play sound purely in javascript. With an OSS FFT library for frequency analysis in javascript and eSpeak speech synthesis available in javascript I could build the core functionality in pure Javascript.

    As this is an OSS educational application, I really, really do not want to deal with questions about the privacy of school children. Therefore, the application works strictly client side and will NOT communicate anything to a server.

    But the new javascript was something to get used to, indeed.

  101. Yes, microservices are great — but there is considerable infrastructure and layers above sockets and HTTP that go into a microservice infrastructure in order to make it look like an API call to the application developers. […] the parsing overhead is too much. Why do you think HTTP/2 is being taken up so rapidly?

    Pipelining, and because the interface to the developers is completely unchanged, meaning that CloudFlare enabled it for me and I don’t even know whether my container’s routing proxy speaks it or not.

    It does have down-sides. Most notably, no way to automatically validate messages (such as with an XML DTD), and processing overhead.

    DTDs and Schemas are a bear to keep up to date, especially when there’s usually an underlying data type that they’ll be parsed into anyway, in which case autogenerated XML or JSON schemas can be used—and Spring will autonegotiate the content type for me so that I don’t have to worry about whether the serialization is to XML or JSON. In practice, it doesn’t really matter, as the reaction to a malformed message is going to be to throw a 400 in any case.

    The processing overhead is usually worth the tradeoff. It’s a business decision, and an increasing number of companies are moving this direction.

    Regarding PowerShell generally: It’s basically a clone of the old BeanShell/groovysh model, and I’ve been an advocate for this sort of change for a long time. Even the vaunted Plan 9 “everything is a text file” handwaved the fact that the OS essentially implemented a type system that translated structured data into directories, and I have a hard time seeing any difference between that and using a dot operator.

  102. @Winter
    > Else I would not have used Js.

    Winter, I assume the implication from this is that you are not a fan of JavaScript. And rightly so. It truly is the worst language that is in common use. And FWIW, I am truly baffled by things like node.js where people want to move its nastiness into an environment where it isn’t needed. To me that is kind of like installing a snow maker in your front room. Sure, sometimes you need to deal with three inches of snow on the ground, but why would you choose to do so?
    However, my purpose in commenting is to say that you should try typescript. I have been working with it a lot lately and it is a transformative experience for me. It is so massively better than JavaScript that I can hardly recommend replacing all your JS work with it strongly enough.

    Having said that I am a person who is strongly in favor of static typing, I strongly dislike python and similar duck typing languages, in fact, I think the whole concept is rather silly. So ymmv. Some people actually LIKE the worst features of JavaScript (including its ridiculous “anything goes — we will check at run time so that your bugs remain in your code till the worst possible time” type system. But these people are obviously BAD people… lol.

  103. Here is an aphorism: it should be easy to do the right thing and hard to do the wrong thing. The above discussion of duck typing could be solved by this aphorism: the ideal is compile time type inference, where you still not waste time with boilerplate (easy to do the right thing) but mistakes are caught early (hard to do the wrong thing).

    1. >Here is an aphorism: it should be easy to do the right thing and hard to do the wrong thing.

      Yeah, the trouble with this plan is that – often enough to matter – making it hard to do the wrong thing also makes it hard to do any right thing that is outside the narrow range of the designer’s expectations.

      Yes, I’d like to use a language with unlimited-extent allocation, type inference, a rich type ontology, and the ability to express any programming style the job needs (that last point is why you Haskell fans can stifle). Until I get that, I’m OK with duck typing. Python has aged remarkably well. Now if they’d just fix the goddamn GIL…grrrr….

  104. @Jessica
    “Winter, I assume the implication from this is that you are not a fan of JavaScript.”

    No. Javascript is literally my second last choice, the last being Java. (I leave out a few pages of rants)

    @Jessica
    “However, my purpose in commenting is to say that you should try typescript.”

    Alas, my choices were limited. The whole point of the exercise was to get as plain vanilla a project as possible.

    My hope is to get others to take over the project. This did not work out in the previous incarnations. Nevertheless, I try to program to the largest common denominator. I use simple, straightforward javascript that should be understandable by anyone who ever wrote HTML5/Javascript. My target coders are advanced high-school students or early college level students. And the code should work in any standards compliant browser (I know, don’t get me started).

    These “requirements” exclude anything “fancy” that would make a programmer’s life more productive. But, as I started my programming career writing FORTRAN IV code after a course on Algol68, I have extremely low expectations of programming languages.

  105. @Winter

    But as time goes on, education moves to mobile.

    Yes, we’re producing a generation of “smart”-phone zombies. Sad indeed.

    As this is an OSS [emphasis added] educational application, I really, really do not want to deal with questions about the privacy of school children.

    Do you mean a closed-sourced educational application wouldn’t present such questions?

  106. Yes, we’re producing a generation of “smart”-phone zombies. Sad indeed.

    You mean a generation that has the world’s knowledge a google or Siri query away? Something out of science fiction?

    The same generation that starts doing robotics with First Lego League and programming with scratch or code.org or swift playground?

    What’s sad is this sort of glass half empty bs that every generation says about the next.

    Do you mean a closed-sourced educational application wouldn’t present such questions?

    Not ones from Apple.

  107. @nht
    I agree with your optimism, only thing I’d say is “not enough”. You say that schools are teaching this, from what I can see (from friends kids, I have yet to reproduce) is that they make a bit of an effort to get kids coding, but not much. Maybe computer club, or a couple of basic programming courses. But it certainly isn’t mainlined.

    I think the ability to code is on a par with second level subjects like history or music. No school would think to not offer history to every student or even question whether it should be a graduation requirement. But not so programming. Which is a huge mistake. Basic programming should be a high school graduation requirement in my view.

    My opinion? Every kid should be able to write code a little bit. You know some people graduate through a college science major and still can’t write a line of code. To me that is mind boggling, but it is true.

    I’m not saying everyone should be able to hack the Linux kernel or develop a major application. But I am certain that every kid should be able to hack a little utility program, or do some simple macro programming in Excel, or create a basic web page with a little JS.

    People think I am crazy for thinking everyone should code, but protest vehemently when the school doesn’t offer AP oboe lessons. I think we have our priority wrong there in education.

    (Obviously, all of the above is USA centric since I don’t know schools outside of my direct experience.)

  108. @nht

    You mean a generation that has the world’s knowledge a google or Siri query away?

    What good is that if most people don’t value knowledge? If they did, they’d read books. Sure, quick Web searches are very useful (I conduct them all the time); but they’re no substitute for the deeper learning one needs in order to think critically and use language correctly.

    Something out of science fiction?

    Which is not always a good thing. Some science fiction is dystopian.

    The same generation that starts doing robotics with First Lego League and programming with scratch or code.org or swift playground?

    On this we are agreed. I approve of efforts to teach children how to program (wish they’d taught me!). But that doesn’t constitute an argument for mobile devices, which are poorly suited for keyboard-heavy activities such as coding. Scratch may be an exception, since it appears to consist of blocks that are moved around; but they do teach more serious languages after that one, I hope. And if they do, real keyboards are needed.

    Not ones from Apple.

    Why not?

    This is the smartphone “zombie” generation:

    That’s nice, but I was thinking of this kind of situation. (Gotta love signs such as this one.)

  109. @Jorge Dujan
    > What good is that if most people don’t value knowledge?

    Everyone values knowledge. They often don’t value the knowledge that they are “supposed” to value. But they value knowledge that helps them acheive their goals whether it is tomorrow’s weather, or how to get to their destination, or how to get peanut butter out of the carpet.

    Or today, I had to find out how to write a regular expression to match alpha numeric identifiers in a way that works with unicode and foreign languages. Did I look it up in a book? What do you think. However, I had to get some deep understanding of how Unicode represents certain types of language. Unfortunately the local library was closed. The web however, was not.

    > If they did, they’d read books.

    Books are fabulous, but also deeply flawed. The rate at which you can acquire useful data from book is very low. Furthermore books are limited by their physical capacity. After all, nearly all books are between 100 and 1000 pages. Why? Because that is the optimal size for all knowledge? Of course not. It is because of the limitations of the medium. As a consequence many books are bursting full of fluff, and you’d be better off reading the Cliff notes or summary wikipedia.

    True knowledge really comes from the application of ideas not reading. The web offers a much more flexible medium in so many ways. It is readily size appropriate, it is readily available, it is hyperlinked, and it is dynamic, meaning that other stimulae can be used to convey information, and even to interact with the data, or put it in to practice directly.

    Want to learn geometry? For many people it is way easier to acquire it by using youtube lectures, and downloading some problem sheets. No book required.

    Books were amazing. A revolution in the history of humans. However, we live in a post book world. When was the last time you looked in a book to find out how to do something in your computer program? Ten years ago, right?

    > That’s nice, but I was thinking of this kind of situation. (Gotta love signs such as this one.)

    Just because you don’t approve of this method of communication doesn’t mean there is anything wrong with it. Tell me your phone number please. I want to text you to “Get off my lawn.”

  110. @ Jessica Boxer

    Basic programming should be a high school graduation requirement in my view.

    Given how promiment information technology has become, yes. Knowing the basics of programming would amount to a better understanding of the present world. So I basically agree with you and Nigel about that part. (Admittedly, I haven’t learned to program yet. Guess I should work through that Python tutorial Eric recommends in “How to Become a Hacker”.)

    However, isn’t it even more important to teach children how to think critically and use language correctly? Well, the teaching of programming might indirectly aid in both areas. I mean, doesn’t programming promote a practical and logic-oriented mindset? If so, schools should teach it by all means, and from an early age.

    Alas, we’d first need an education system that worked. When absurdities like saying “least favorite” to mean “least liked” are so common (to say nothing of youngsters who use “your” and “you’re” indistinctly), we can’t help suspecting most schools are failing to inculcate healthy thinking habits. Which brings me to why I object to the term “smartphone”, spelling it as “‘smart’ phone”: I perceive it as a marketing ploy to make people feel smart for using a supposedly smart object. I don’t believe – at least not firmly – that these devices are making people dumber, but they’re certainly not making them any smarter. Why not just call them “webphones”?

    or do some simple macro programming in Excel

    Well, they did teach me some of that in high school. I’ve forgotten everything, though; I’m not going to blame “the system” for it, since I could have retained that knowledge through regular practice. But if I haven’t been using Excel (or some equivalent), it’s because I haven’t needed it; and if I ever need it, I’ll relearn it on my own. Besides, my point above wasn’t that teachers should make sure no pupil ever forgets anything they’re taught, but that they should provide a solid, lasting foundation for sound thinking and articulate expression.

    Remember the Trivium: grammar, logic, and rhetoric. Elementary and high schools do impart the basics of grammar, but – as far as I know – not the other two.

    People think I am crazy for thinking everyone should code, but protest vehemently when the school doesn’t offer AP oboe lessons.

    What does “AP” stand for in this context?

    1. >However, isn’t it even more important to teach children how to think critically and use language correctly?

      Yes. I’m with Jorge on this one. Rather than trying to teach every child to code, I would much rather they get a course in methods of rational thinking based on some derivation of Eliezer Yudkowsky’s Sequences.

  111. Jessica Boxer:

    Everyone values knowledge. They often don’t value the knowledge that they are “supposed” to value. But they value knowledge that helps them acheive their goals [in mundane pursuits].

    Yes, but I fear that, by neglecting more profound subjects, we become more primitive. I also fear that poverty of grammar and vocabulary leads to poverty of thinking (I accept the Sapir-Whorf hypothesis), which in turn leads to evils such as poor choices in political leadership. (That such leadership exists is itself an evil, but some leaders are less harmful than others.)

    Did I look it up in a book? What do you think.

    I did acknowledge the usefulness of search engines; I was merely cautioning that they shouldn’t be the be-all and end-all of knowledge acquisition. For our purposes, in-depth online materials are no different from books; and we also have e-books proper, of course.

    many books are bursting full of fluff

    I’m not going to deny that. But, apart from those books our school or college forces us to read, we’re free to choose our reading material according to any criterion. Possible criteria include brevity and conciseness.

    Furthermore, the non-essential parts of a non-fiction book may possess literary value and thus make the experience more pleasant than a shorter-but-colder learning resource. Of course, the opposite scenario is also possible (i.e. an engaging non-book vs. a dry book); but I’m just saying that the book format has its place because of its potential for enjoyment. And when we’re dealing with fiction, who would prefer a plot summary over the real thing (provided one likes the premise of the work in question)?

    And just in case, let me clarify one thing: when I say “book”, I’m not referring specifically to paper books.

    The web offers a much more flexible medium in so many ways.

    You’re right. I wrote: “If [most people valued knowledge], they’d read books.” Instead, I should have said “…they’d engage in in-depth learning”, irrespective of the medium chosen for such learning. I narrowly focused on the book medium, but you’re right in that those other formats (such as YouTube lectures) are equally valid. I stand corrected.

    That said, I share Nicholas Carr’s concerns about the potential cognitive effects of chasing link after link. Nevertheless, it’s perfectly possible to exert self-discipline; and YouTube lectures are hardly problematic in that regard anyway.

    Just because you don’t approve of this method of communication doesn’t mean there is anything wrong with it.

    I meant to denounce its overuse, though I admit my choice of words was rather poor. But since you mention it: yes, I do feel there’s something inherently wrong with mobile phones, for they compromise security (they contain personal information about oneself and one’s contacts, and can be easily stolen), damage privacy through tracking, and encourage certain kinds of rude and/or reckless behavior.

    I want to text you to “Get off my lawn.”

    Heh. I’m confused: isn’t that phrase supposed to be what curmudgeons say? I, and I alone, am the curmudgeon in this story, the one who rejects the new ideas or practices in question (I’m twenty-seven, but old at heart). Or maybe I got it all wrong, due to cultural differences. Oops!

    Seriously, though: perhaps I could be accurately classified as a conservative, at least in cultural matters. Wasn’t it Burke’s contention that radical change in a society’s mores or institutions can have unforeseen and undesirable consequences?

  112. @Jorge Dujan
    “Yes, we’re producing a generation of “smart”-phone zombies. Sad indeed.”

    It is an app to practise language pronunciation. Also, mobile is becoming the dominant computer use platform.

    @Jorge Dujan
    “Do you mean a closed-sourced educational application wouldn’t present such questions?”

    Yes, but then you know whom to ask questions. An OSS app can be used by everyone, but the questions come back to the coder.

  113. @Jorge Dujan
    > Yes, but I fear that, by neglecting more profound subjects,

    You think that because people “don’t read books” that that somehow means they neglect profundity? I can’t imagine why. I am pretty sure in my adult life I have learned more from online interactions and study than I have from books. In fact the very process of interacting can provide a great deal more depth than the hollow interaction with paper. Look at where you and I are at right now. Lots of profound learning can be had here, that’s for sure.

    > I also fear that poverty of grammar and vocabulary leads to poverty of thinking (I accept the Sapir-Whorf hypothesis),

    If you accept SW, then presumably you will welcome the dramatic pace of change in the English language, perhaps the greatest speed of change and expansion in the language in its history, that we are experiencing? Changing, expanding language, by SW, would mean an expansion of the mind. A few gaffed spelling errors is the messy price we pay for broadening our minds, no?

    Let me give you one simple example, the emoticon. A genius addition to the written word that enables the clear conveyance of color and tone in language that had been extremely hard to capture in the past. It is a micro language within the context of our existing writing offering a broad range of communicative flavor. A million dropped apostrophes seems a small price to pay for such a useful tool.

    > which in turn leads to evils such as poor choices in political leadership.

    “Your”, “you’re” mix ups lead inevitably to Donald Trump? Is that your argument? ;-)

    (See what I did there?)

    Methinks you are extrapolating a little too much there.

    And FWIW, always on connectedness changes the political dynamic in powerful and useful ways. Thank god it allows us to bypass the dreadful American press and get access to a less groomed set of news. As for me, my French is kind of rough, but thanks to Google I regularly read Le Figaro and occasionally some German and Chinese newspapers (and frequently British newspapers) to get me out of my American parochialism. The web is absolutely transformative in broadening our global reach of political understanding if we want it.

    > I did acknowledge the usefulness of search engines; I was merely cautioning that they shouldn’t be the be-all and end-all of knowledge acquisition.

    Always on connectedness is a lot more than access to a search engine.

    > For our purposes, in-depth online materials are no different from books; and we also have e-books proper, of course.

    I disagree. In depth online materials (the better ones anyway) are vastly different than books. To use the idiom of the day, they are “Book 2.0.”

    > I’m not going to deny that. But, apart from those books our school or college forces us to read, we’re free to choose our reading material according to any criterion.

    I think you are missing my point. My point is that the fluff is an inevitable consequence of the paper medium. Of course there is plenty of fluff on the web, just as there are lots of very bad books, but the physical nature of books forces them to be of a size that often demands fluff (or on the other side of the equation, size constraints prevent sufficiently in depth treatments.)

    I was recently listening to a song that mentioned the island of St. Kilda. I wanted to know a little about its location. Clicky, clicky, tappy tappy and I found the 411. Without wikipedia I’d need a very general book like an encyclopedia or atlas. So to get my one paragraph information I need the ninety volumes of Encyclopedia Britannica? Now that is a lot of extraneous fluff. It is the nature of the physical material that makes it so.

    > Furthermore, the non-essential parts of a non-fiction book may possess literary value and thus make the experience more pleasant than a shorter-but-colder learning resource.

    You can’t think that books are, on average, more engaging that the dynamic content of a web site, or a video, or an interactive forum or one of the thousand other ways always on connectedness has changed the world?

    > And when we’re dealing with fiction, who would prefer a plot summary over the real thing

    I imagine many people would. It is why you can buy them. Certainly fiction is a useful form of recreation, but you can’t honestly believe that always on connectedness has decreased our recreational choices.

    > That said, I share Nicholas Carr’s concerns about the potential cognitive effects of chasing link after link.

    I don’t know what you are referring too, but I can tell you, from a recreational point of view, there are few things more enjoyable than random walk binge on wikipedia. That is one of my guilty pleasures. I like browsing around in libraries too, but wiki gluttony is much more intellectually calorific.

    > Heh. I’m confused:

    No, you aren’t. Curmudgeon is what I was going for. :-)

  114. “Your”, “you’re” mix ups lead inevitably to Donald Trump? Is that your argument? ;-)

    Carelessness in one area often translates to carelessness in others. Americans are notoriously apathetic in a cross-cutting way but I don’t think a causative relationship can be implied between apathy in one area and apathy in another. It’s just part of the American spirit.

    When the French discovered social media, they used it to engage each other in serious political and philosophical discussions. When the Americans discovered social media? Selfies and cat videos.

  115. @esr
    “Yes. I’m with Jorge on this one. Rather than trying to teach every child to code, I would much rather they get a course in methods of rational thinking based on some derivation of Eliezer Yudkowsky’s Sequences.”

    But that would be akin to math (actually, we know it IS math). For many children that would be too abstract. Hence the use of Logo.

    I remember ELAN, a derivative of Algol68 that was very instructive. You programmed by writing sentences that described what the program should do in ever greater detail. Only at the lowest level would you use language primitives. It kept you on the task of reasoning what should happen instead of worrying about syntax.

  116. @Jessica
    “> That said, I share Nicholas Carr’s concerns about the potential cognitive effects of chasing link after link.

    I don’t know what you are referring too”

    The Shallows: What the Internet Is Doing to Our Brains
    by Nicholas Carr (Author)
    https://www.amazon.com/Shallows-What-Internet

    A very shallow analysis of internet culture along the lines of:
    “The children now love luxury; they have bad manners, contempt for
    authority; they show disrespect for elders and love chatter in place
    of exercise. Children are now tyrants, not the servants of their
    households. They no longer rise when elders enter the room. They
    contradict their parents, chatter before company, gobble up dainties
    at the table, cross their legs, and tyrannize their teachers.”

    The above is from Plato, attributed to Socrates.

    The claim is the same as the MTV culture of yore. The Youth does not read linear texts anymore and that is considered a bad thing.

  117. @Jessica
    I am fond of reading books*, but I admit to do most of my reading on internet papers. The internet is more up to date, but a book is more engaging. The thesis of Carr is based on the false assumption that you should read a book cover to cover in one go. He also ignores the fact that a book is always part of a larger “discourse” with other publications. Eg, a Heinlein SF story is best read in context of contemporary SF stories and some US political ideas. You miss a lot when you read it in isolation.

    * Currently a veritable tome on economics on paper and Spinoza as eBook.

  118. Yes. I’m with Jorge on this one. Rather than trying to teach every child to code, I would much rather they get a course in methods of rational thinking based on some derivation of Eliezer Yudkowsky’s Sequences.

    I’m thinking that you are overestimating the abilities of the US primary and secondary educational system. Especially, say, in states like Texas…

    Getting school systems to teach coding is a lot simpler since STEM can be couched in a way to mitigate “they took ur jerbs!”

  119. @winter lol, yes. Humans have been documented bitching about the next generation sucks since 400 BC.

  120. Especially, say, in states like Texas…

    Ah, just when I was reconsidering having pegged you as a smug bigot.

  121. @Jeff Read
    > When the French discovered social media, they used it to engage each other in serious political and philosophical discussions. When the Americans discovered social media? Selfies and cat videos.

    What appalling bigotry. The aforementioned Donald Trump has pwned social media for political discourse. And I have a number of French friends whose social media is full of
    chats jouant du piano and shallow aphorisms. They just sound more classy and profound because they are, you know, in French.

    Americans are so shallow that, despite being 5% of the worlds population, they produce 50% of the world’s Nobel laureates (and far and away the best Olympic swimmers bitch…. :-)

  122. @Winter
    >the false assumption that you should read a book cover to cover in one go.

    Yes, indeed. Books well read are simply not read like that (except maybe fiction.) One of the new movements in English education is the practice of annotating. That is the kids have to annotate the book as they read it, marginal notes, highlighting key phrases and introductions etc.

    I think this is an excellent change in attitude from the “don’t fold the corner lest you damage the precious paper) attitude that has often prevailed with books.

    The annotation is a ersatz interaction between the reader and the author. Book 2.0, which we are doing right now is a much deeper and more profound discussion between reader and author.

  123. @ esr

    Yes. I’m with Jorge on this one.

    Thanks.

    Rather than trying to teach every child to code, I would much rather they get a course in methods of rational thinking based on some derivation of Eliezer Yudkowsky’s Sequences.

    That would be awesome, but maybe we should curb our enthusiasm. Winter has a point: wouldn’t that be too abstract for children? If Piaget was right, abstract thinking doesn’t become possible until the preteen phase. Could that be circumvented by means of early stimulation?

    @ Winter

    It is an app to practise language pronunciation.

    That is indeed a legitimate use for mobile devices; I wasn’t attacking your work (sorry if it sounded as though I was). I just don’t see why everything should go mobile, and doubt that it will suffice for fixing the shortcomings of any given school or school system.

    Also, mobile is becoming the dominant computer use platform.

    Yes, and I hate that. ’Sides, they’re gonna need real keyboards if they’re to write code, papers, mini-essays, poems, or (short) stories. And if children are no longer required to write any of that, I fear for the future of mankind.

    Likewise, I don’t want them to stop teaching handwriting or the algorithms for the basic operations of arithmetic. I don’t think we should become more and more dependent on electronic devices; and stimulating our brains in as many different ways as possible helps prevent neurodegenerative diseases, right?

  124. @ Jessica Boxer

    Part I

    You think that because people “don’t read books” that that [sic] somehow means they neglect profundity?

    What I think is that there’s a correlation between shallow mindsets and disdain for books.

    In fact the very process of interacting can provide a great deal more depth than the hollow interaction with paper.

    I’m all for interactive learning; as I child, I did use some pieces of interactive educational software. Maybe I expressed myself poorly in my previous comments, or maybe my reasoning itself was misguided. In any case, I thank you for bringing interactive learning to my attention. But I insist that I wasn’t talking specifically about paper.

    Changing, expanding language, by SW, would mean an expansion of the mind. A few gaffed spelling errors is the messy price we pay for broadening our minds, no?

    Having a bigger toolset is of little use if many of the tools are abandoned. If, in practice, people use fewer and fewer words and expressions, and – what’s worse – fewer tenses, I don’t see how language can be said to be expanding.

    You rightly praise the broadening of langue, but I don’t think it makes up for the impoverishment of parole.

    Let me give you one simple example, the emoticon.

    Yes, I do use them. :-)

    A million dropped apostrophes seems a small price to pay for such a useful tool.

    I don’t think so. If only we had emoticons and a more linguistically rigorous public…

    “Your”, “you’re” mix ups lead inevitably to Donald Trump? Is that your argument? ;-)

    No, but those two phenomena may have a common root: failure to perceive nuance. Ideologically, it manifests as a tendency to judge hastily and generalize; such a person will tend to choose political candidates who exhibit the same mindset, such as Trump.

    Methinks you are extrapolating a little too much there.

    If I am, then so was Hayakawa when he wrote that “everyone needs to have a habitually critical attitude towards language” and that “if the majority of our fellow citizens are more susceptible to the slogans of fear and race hatred than to those of peaceful accommodation and mutual respect among human beings, our political liberties remain at the mercy of any eloquent and unscrupulous demagogue.”

    My source is Wikipedia; I do appreciate its usefulness despite my reservations.

  125. @ Jessica Boxer

    Part II

    The web is absolutely transformative in broadening our global reach of political understanding if we want it.

    Yes, I agree. I just wish more people wanted it.

    Always[-]on connectedness is a lot more than access to a search engine.

    Again, how many people actually get deeper than that?

    I disagree. In[-]depth online materials (the better ones anyway) are vastly different than books.

    I meant they’re the same for the purpose of that part of the discussion. I was distinguishing between quick searches on the one hand and any kind of deeper learning on the other. (My wording was certainly poor in that case.)

    the physical nature of books forces them to be of a size that often demands fluff

    Not in the case of booklets. :-)

    (or on the other side of the equation, size constraints prevent sufficiently in[-]depth treatments.)

    Not in the case of multi-volume works, whose existence you acknowledge in your next paragraph.

    Without wikipedia I’d need a very general book like an encyclopedia or atlas.

    You’re right, and I have no “buts” this time. The ability to search with a computer – be it desktop, laptop, or palmtop – is irreplaceable sometimes.

    You can’t think that books are, on average, more engaging

    Not on average, presumably.

    I imagine many people would [prefer a plot summary]. It is why you can buy them.

    Yes, when your school or college requires you to read the book and you don’t want to. I was talking about reading fiction voluntarily and for pleasure.

    you can’t honestly believe that always[-]on connectedness has decreased our recreational choices.

    No, and I didn’t say that. By the way, I do play video games and watch YouTube videos. But reading still has its place – be it on paper or on a screen. I read on both, especially the latter; I admit I’m finding it somewhat difficult to read on paper. Nevertheless, I managed to read Citizen of the Galaxy, as promised. Didn’t like it; Eric will despise me for that, I’m afraid. But I digress.

    I don’t know what you are referring too [sic]

    Please read either the article in question or the shorter Wikipedia article about it.

    Curmudgeon is what I was going for. :-)

    But that’s my role. :-P


  126. Ah, just when I was reconsidering having pegged you as a smug bigot.

    Well, I suppose that Texas is improving by allowing teachers more say in text books (amazing concept) over very loud parents pushing intelligent design as science. The idea that a school of thought pushed by largely atheist proponents would play well in more religious states ignores reality.

    Sorry reality disagrees with your worldview.

  127. @Jorge Dujan
    > What I think is that there’s a correlation between shallow mindsets and disdain for books.

    Perhaps between shallow mindsets and a disdain for learning that you proxy with an attitude to books. However, my point is that the proxy is no longer as reliable as it was in the past. There are many smart people who never pick up a book but source their information in other ways.

    Oh and BTW, one of the developments of modern English, accelerated by quick online communication, is the verbification of nouns which I just did, twice. How delightful is that?

    > But I insist that I wasn’t talking specifically about paper.

    No your basic contention is that always connectedness makes us dumber. And the interactive nature of online communication would suggest quite the opposite.

    > Having a bigger toolset is of little use if many of the tools are abandoned.

    People unclutter their lives of outdated tools all the time. We programmers are constantly refreshing our toolbox. So I entirely disagree.

    > If, in practice, people use fewer and fewer words and expressions, and – what’s worse – fewer tenses

    But the opposite is true! Online communication has brought us many new words and expressions. And what tense do you fear we will loose? Past, present or future? Perhaps you are thinking of moods, aspects and so forth… and I’d ask, if English was to loose the subjunctive completely will we be all the poorer?

    (See what I did there? ;-)

    > I don’t think so. If only we had emoticons and a more linguistically rigorous public…

    Yeah, and if only a unicorn would bring me some zero fat muffins in a basket every morning for breakfast.

    > No, but those two phenomena may have a common root: failure to perceive nuance.

    Or their root could be the desire to move through thought quickly and not worry too much about the minutiae, focusing on the big picture. And that is often generally speaking a pretty good idea.

    > Ideologically, it manifests as a tendency to judge hastily and generalize; such a person will tend to choose political candidates who exhibit the same mindset, such as Trump.

    As opposed to candidates who befuddle with obfuscation and deceit such as Hillary Clinton? I’d suggest her husband was the master of this focus on distracting minutiae to the detriment of big picture truth. Depending on what the meaning of “is” is…

    Politics is and always has been an insult to language. It is the deliberate manipulation of language to mean what it doesn’t in an attempt to seduce by the emotions rather than logic. So I don’t know that political discourse adds much to your case.

    > If I am, then so was Hayakawa when he wrote that “everyone needs to have a habitually critical attitude towards language”

    Who he? And why should I care anymore about his assertions than yours.

    > My source is Wikipedia; I do appreciate its usefulness despite my reservations.

    You, you, you….pragmatist…

  128. @Jorge Dujan
    Omitting stuff to save rehashing.

    > I don’t know what you are referring too [sic]

    Now I thought this was interesting. Here you point out my little typo. I don’t take it personally by any means (esp. since it is plainly incorrect), but it is a perfect inline illustration of what we are talking about.

    The purpose of language is communication not grammatical correctness. Is it better to deal with the substance of the message rather than the form? Surely if the meaning is communicated it doesn’t matter too much how it happens, unless there are significant other effects (such as it is understandable, but takes more work to understand.) By no means do you dwell on this, so no foul no fault here. But I just thought it was an interesting online example.

    One of the things I am interested in is the subject of what makes a persuasive argument as opposed to the often orthogonal question of what makes a correct argument. One of the most powerful techniques for making a persuasive argument is to mock your interlocutor.

    We see it right now in big bright lights with the way the press are treating Donald Trump. Don’t get me wrong, he is often mock-worthy, but it is evident to anyone with eyes to see that it is a deliberate ploy by the media to drown out what he is saying with derisive laughter. Why? I think a lot of reasons, but certainly one is that they feel that the things that Trump say are dreadful but that they also resonate with the rubes in flyover country. Just for what it is worth, for those of you who care, it is why I think opinion polls are probably underreporting the Trump vote by about 5 pct points. The mockery has made people both resolute in their support of him and unwilling to admit it in public.

    Oh, and to be clear, by no means am I accusing you of such here… because if you did I’d mock you in response for your pretentious use of Latin….:-)

    > Please read either the article in question or the shorter Wikipedia article about it.

    Ah, the irony… (which I am sure was quite deliberate on your part.)

  129. @Jorge
    Belated: I was referring to “scared off” after the… um, well it didn’t even rise to the level of “spat”, so I’m not sure what you would call it with ESR.

    Wasn’t it Burke’s contention that radical change in a society’s mores or institutions can have unforeseen and undesirable consequences?

    Ah! But here lies a problem: Radical change imposed on people has obvious problems I don’t think I need to explain here, but radical change that evolves on it’s own is going to happen regardless of what anyone thinks of it. The only way you can slow it is to impose a crude imitation of “stability” on people. Much like “preserving [blah] culture” means forcing people to choose lives the preserver approves of. Which makes sense, because “preserving stability” is just a special case of “preserving [blah] culture”.

    The distinction between someone imposing a new or old culture, and culture evolving according to technological, economic, and previous cultural factors is one that many people seem to have an incredibly hard time wrapping their head around. For a good example see much of the alt-right, who correctly critique culture-change impositions and advocate not for the removal of those impositions, but different impositions that they would like.

    Which is just as crazy as subsidizing a company that is about to go bankrupt due to tax burdens.

  130. @jessica I’m not certain that I would bother delving too deeply into Carr. My recollection was that it was easy to find literature to show neuroplasticity but there was none that supported that the any cognitive changes from internet use was harmful (which was his main assertion).

    So at the end of the day it appears the data supports that the brain changes when you use it and not a whole lot else.

    One of the most memorable claims was about Nietzsche and his typewriter…which was actually broken and he used only for a very short period…while he was going blind. Which probably impacted his writings even more than the new medium he was using.

  131. I don’t buy the ‘Internet enstupidation’ theory. Technology is merely amplifying factors which were operating strongly long before the Internet was anything more than science fiction. As Frank Zappa put it, in 1981:

    Modern Americans behave as if intelligence were some sort of hideous deformity. To cosmeticize it, many otherwise normal citizens attempt a peculiar type of self-inflicted homemade mental nose-job (designed to lower the recipient’s socio-intellectual profile to the point where the ability to communicate on the most mongolian level provides the necessary certification to become ONE OF THE GUYS). Let’s face it . . . nobody wants to hang out with someone who is smarter than they are. This is not FUN.

    @FooQuuxman on 2016-08-27 at 20:29:27

    > . . . radical change that evolves on its own is going to happen regardless of what anyone thinks of it. . . .
    > The distinction between someone imposing a new or old culture, and culture evolving according to technological, economic, and previous cultural factors is one that many people seem to have an incredibly hard time wrapping their head around. For a good example see much of the alt-right, who correctly critique culture-change impositions and advocate not for the removal of those impositions, but different impositions that they would like.

    Deep insights, well said. Something else the alt-right (particularly the more religious strains) doesn’t get is that pressure on older cultural norms, both imposed and evolutionary, is a direct consequence of serious deficiencies in justification for those norms – many of which are in fact beneficial, but “cuz a Big Guy In The Sky sez so” ain’t gonna hack it anymore.

    Great avatar, BTW.

    1. >Modern Americans behave as if intelligence were some sort of hideous deformity.

      What planet do you live on?

      I can assure you this is not true even at my martial-arts school, let alone my Friday night gaming group.

  132. @ESR

    Might I suggest that this is not the most representative sample in the history of ever?

    @trying2b-amused

    You, or rather Frank Zappa, are suffering from Right End Of The Bellcurve Syndrome. Pay attention, I’m about to give a practical demonstration of it.

    Deep insights, well said.

    This is something that should be obvious to anyone with the cognitive capacity of a gopher. Unfortunately 99.9% of the population has an observable IQ of about 15, and is apparently proof of the longstanding meme that humans only use 20% (arguably 1%) of their brains. I sympathize more with the control-the-pesants-for-their-own-good fools than I used to, even as I still condemn them. High INT is not a pleasant abyss to stare into.

    I have strong evidence, though anecdotal, for being ~1SD above the mean, with the possibility of ~2SD. I don’t think I’m even close to 3SD, but I wouldn’t be in total shock if I were. Whatever the actual number, if I am smart, then we are completely screwed.

    Great avatar, BTW.

    Found here. I choose that after seeing 2 other people on G+ using other symbols from the set. One of them being Alistair Young, the writer of The Eldraeverse, which I highly recommend reading.

  133. @esr on 2016-08-28 at 04:54:17:

    >What planet do you live on?
    Unfortunately, this one.

    >I can assure you this is not true even at my martial-arts school, let alone my Friday night gaming group.
    No doubt. But the numbers involved, by their own inclination, versus mindless trend-worship (association with which I presume you would strongly – albeit possibly subconsciously – self-select against) in such activities are barely a rounding error compared to the foregoing.

  134. Jessica Boxer:

    There are many smart people who never pick up a book but source their information in other ways.

    OK. Maybe I was wrong.

    …the verbification of nouns which I just did, twice. How delightful is that?

    It’s not my cup of tea, so I generally avoid it.

    No your basic contention is that always connectedness makes us dumber.

    No, I merely denounced the isolation. It would be unfair of me to blame cell phones for a process of intellectual decadence that predates them. What I did say is that they don’t seem to be helping reverse such decline.

    People unclutter their lives of outdated tools all the time. We programmers are constantly refreshing our toolbox.

    I’m no programmer, but I suspect your analogy is flawed: having many tools for producing and expressing different thoughts is not like knowing how to write the same program in many different languages, is it?

    I’d ask, if English was to loose [sic] the subjunctive completely will [sic] we be all the poorer?

    I don’t know; but, if it matters, it would sadden me. Incidentally, since Interlingua is featured prominently in Citizen of the Galaxy, I decided to read the Wikipedia article about it. It was interesting until I saw this: “…there is no subjunctive.” I suddenly lost all interest in the language.

    Or their root could be the desire to move through thought quickly and not worry too much about the minutiae, focusing on the big picture. And that is often generally speaking a pretty good idea.

    Yes, but how can a big picture (complex) be understood by someone who doesn’t even understand the difference between “your” and “you’re” (simple)?

    As opposed to candidates who befuddle with obfuscation and deceit such as Hillary Clinton?

    As Jay Maynard would say: “Merciful $DEITY, no!”. I discussed Trump because that was your example, but please don’t think I approve of the Democratic Party. Especially after Eric described the outrageous effects of Obamacare on him, his wife, and presumably many other honest Americans.

    Politics is and always has been an insult to language. It is the deliberate manipulation of language to mean what it doesn’t in an attempt to seduce by the emotions rather than logic. So I don’t know that political discourse adds much to your case.

    There are gradations. Some politicians abuse language more than others. Nevertheless, it appears that neither of us wants the State to exist in the first place; can’t we agree that larger doses of rigorous thinking in our communities would make the advent of anarcho-capitalism more likely?

    Who [was Hayakawa]?

    A proponent and popularizer of General Semantics.

    And why should I care anymore about his assertions than yours.

    Because he was more intelligent and knowledgeable than me.

    (To be continued.)

  135. Jessica Boxer:

    Here you point out my little typo.

    I do that all the time; just ask Eric. I often correct my father as well; annoys the hell out of him. In my defense, I even correct myself sometimes. You, the rest of the A&D community, never correct me; I wish you did, for I know I make mistakes. In any case, I didn’t mean to offend you; if I did, I apologize.

    The purpose of language is communication not grammatical correctness.

    Communication requires a code, a set of rules all parties accept. The ignorance around us is not even a coordinated ignorance with just a poorer code; it’s a mess where each person follows or ignores certain rules regardless of what rules others follow or ignore. There is still a code (e.g. regarding the meaning of the most common words), but it’s getting fuzzier and fuzzier.

    it is understandable, but takes more work to understand

    I believe we must write in a way that eases the flow of reading, as a courtesy to the reader.

    One of the things I am interested in is the subject of what makes a persuasive argument as opposed to the often orthogonal question of what makes a correct argument.

    Yeah, rhetoric over logic. It sucks.

    “The truth of a proposition has nothing to do with its credibility. And vice versa.” — Lazarus Long

    Oh, and to be clear, by no means am I accusing you of such here…

    I certainly didn’t do that; again, I correct all the time. I assure you I wouldn’t purport to refute an argument just by pointing out a typo in it. More generally, having read Paul Graham’s “How to Disagree” helps me resist the temptation to debate dishonestly. (I know, I know: I need to diversify my sources. :-P)

    If you spot the slightest hint of dishonesty in my comments, please call me on it. There were moments when I didn’t have much of an argument and invoked personal preference, but I just do that as an admission of defeat. Some of my positions may be visceral, rather than rational, in origin. :-(

    because if you did I’d mock you in response for your pretentious use of Latin….:-)

    I’ll grant you that I’m pretentious and pedantic. But what’s wrong with “sic” or “Trivium”?

  136. Especially after Eric described the outrageous effects of Obamacare on him, his wife, and presumably many other honest Americans.

    Are Americans not normally honest that you need to specify that the ones impacted are particularly honest? Are the folks that benefited all dishonest?

    Let’s look at some stats:

    “In all, minorities gained more than whites, making up two-thirds of the increase in insured adults across the country, and 70 percent of the increase in private insurance. Minority men who work as groundskeepers and janitors saw substantial gains, rising to 59 percent insured, up from 51 percent in 2013. Hispanic male construction workers rose to 43 percent insured, from 36 percent in 2013.”

    Are these people not “honest Americans”?

    Who did Obamacare hurt?

    “But there are three subsets of people whose policies were canceled and who are likely to end up as losers under Obamacare — people who are self-employed, over 35, white, or some combination of all three. ”

    Are these the only “honest Americans”?

  137. “You mean a generation that has the world’s knowledge a google or Siri query away?”

    Yes,they have all the facts at their fingertips.

    One of those very smart Greeks from 500 BC observed that, “Learning facts does not train the mind.”

  138. @ Foo Quuxman

    I was referring to “scared off” after the… um, well it didn’t even rise to the level of “spat”, so I’m not sure what you would call it with ESR.

    What happened is that I treated him badly. He didn’t deserve that, especially after all the compassion he’d shown me (and has shown me since). I still feel bad about it.

    But I did comment on a later thread. We debated about Unity because I’m the only one here who likes it, remember? ;-)

    What I don’t understand, though, is why my leaving the community should worry anyone. You folks are very tolerant of me, and I appreciate that; but I know I’m out of my depth here and don’t contribute much, if anything. But I digress.

    …radical change that evolves on it’s own is going to happen regardless of what anyone thinks of it.

    Yes, but that doesn’t mean it cannot be adverse.

    @ trying2b-amused

    >What planet do you live on?
    Unfortunately, this one.

    Yes, cell phones promote inattentional blindness. Pokémon Go is just a particularly perverse manifestation of the phenomenon.

    @ nht

    Are Americans not normally honest that you need to specify that the ones impacted [by Obamacare] are particularly honest? Are the folks that benefited all dishonest?

    All I know for sure is that two Americans I admire are going through hardship because of that government program.

  139. (At one point, I wrote “blockquote” instead of “/blockquote”. That’s what ruined the formatting.)

  140. @LS So…because they have internet (and thereby need to memorize/learn facts even less than before) today’s kids have less time/opportunity/inclination to “train the mind”?

    Interesting logic.

    I was curious which “smart Greek” said this but googling the quote it appears to be attributed to Einstein.

    Hmmm…

    It would make sense that a society without the printing press and had to hand document everything (so fewer reference materials easily available) might place a higher educational value on memorization of facts than modern societies thus making this a more likely modern quote than an ancient one.

    On the other hand, rhetoric IS easier when folks can’t fact check you with their phones…

  141. All I know for sure is that two Americans I admire are going through hardship because of that government program.

    All I know for sure is that you avoided discussing why you choose the phrase “honest American”.

    Somewhat odd for someone who has hectored us that one must have a “habitually critical attitude towards language” to enable “larger doses of rigorous thinking”…on politics.

    Were you engaging in some uncritical application of language or perhaps just uncomfortable in exploring what you meant by using that phrasing?

  142. @nht
    > All I know for sure is that you avoided discussing why you choose the phrase “honest American”.

    I suggest you do what you are asking of him. His original statement was:

    >… the outrageous effects of Obamacare on him, … and presumably many other honest Americans.

    This statement claims only that some (or many) honest Americans were badly impacted by Obamacare. It does not claim that all honest people were so affected, neither does it claim that no honest people were not affected, nor that no dishonest ones were affected. It claims only that the intersection of the set of Americans who are honest and the set of Americans who were negatively impacted by Obamacare can be sized as “many.”

    Your implication that he is calling the people who benefited from Obamacare dishonest does not line up with the facts.

  143. @LS
    > One of those very smart Greeks from 500 BC observed that, “Learning facts does not train the mind.”

    But they also have amazing immersive, interactive, adaptive tools to train their mind too. And learning facts is also an excellent way to train the mind. Sorry, your Greek guy is way out of date.

  144. @Jorge
    > [verbification] is not my cup of tea, so I generally avoid it.

    Oh you should totally Sapir-Whorf it.

    > process of intellectual decadence

    Decadence? Really? It seems rather pejorative for a process of such delight. Maybe, but I think we should absolutely luxuriate in the new vistas of language tweeting in our direction. To quote the Simpsons, it embiggens us all.

    > I’m no programmer, but I suspect your analogy is flawed: having many tools for producing and expressing different thoughts

    You aren’t a programmer. We abandon the old tried and true ways all the time to embrace new ideas and discoveries. In fact that is at the root of the OP.

    > Yes, but how can a big picture (complex) be understood by someone who doesn’t even understand the difference between “your” and “you’re” (simple)?

    But people certainly do understand the difference. They just don’t spell it out (lol.) And let’s face it, this is, at its root, a flaw in English, namely that homophonic expressions can have such radically different meanings. Remember, Spanish schools don’t have spelling class for a reason.

    > can’t we agree that larger doses of rigorous thinking in our communities would make the advent of anarcho-capitalism more likely?

    I’m not an anarcho-capitalist, and I don’t really know the answer to your question. For sure, more rigorous thinking would make for a better body politic.

  145. @Jorge Dujan
    > Communication requires a code, a set of rules all parties accept.

    Sure, but the set of rules that are required for effective communication are smaller than the set required for formally correct communication. We programmers do this all the time: we design our languages to be promiscuous. In fact there is an aphorism, or perhaps a law, called Ma href=”https://en.wikipedia.org/wiki/Robustness_principle”>Postel’s law that many programmers abide by. (FWIW, I don’t actually agree with this principle in computer languages, but I do in human languages.)

    > The ignorance around us is not even a coordinated ignorance with just a poorer code;

    What you see as indiscipline I see as a human society pushing and stretching the boundaries and forms of communication to better suit their needs. Humans have been doing this since language was invented, and it is an extremely good thing. We are just doing it way faster than before, and that too, is a good thing.

    So lose the buzz kill, and plug in… lol, ttyl.

  146. @FooQuuxman on 2016-08-28 at 08:03:16:

    >I’m about to give a practical demonstration of it.
    Which struck me as actually a demonstration of the noble half (for lack of a more adroit phrasing) of the Dunning-Kruger effect.

    You might find this interesting. N.B. I haven’t read it.

    @Jorge Dujan on 2016-08-28 at 21:12:50:

    >Yes, cell phones promote inattentional blindness. Pokémon Go is just a particularly perverse manifestation of the phenomenon.
    Imprudent attempts at multitasking are one thing, blundering into traffic (or worse) chasing cartoon images is another thing entirely.

  147. What you see as indiscipline I see as a human society pushing and stretching the boundaries and forms of communication to better suit their needs.

    Human society also pushed and stretched the boundaries and forms of food supply to better suit their needs.

    Result: we all got fat.

  148. Your implication that he is calling the people who benefited from Obamacare dishonest does not line up with the facts.

    The “fact” is that saying that minority “welfare queens” profit off the backs of “honest Americans” has long been a staple of American conservative politics from even before Reagan help coin the term.

    Anyone that has a “critical attitude towards language” isn’t tossing in an unnecessary adjective without good reason. The sentence works perfectly well without the word “honest” and is more neutral with the presumption that most Americans are sufficiently honest that it isn’t worth remarking on. So, pray tell, what does it mean when added?

    I will thank you to not to piss on my boots and tell me it’s raining.

  149. @Jessica:
    >Sure, but the set of rules that are required for effective communication are smaller than the set required for formally correct communication.

    I’ll add that for human language, what constitutes “formal correctness” is quite often politically charged and can be completely contrary to traditions of usage going back centuries. The idea that a sentence should not end in a preposition comes from the imposition of Latin grammar on English. Sentence final prepositions are a common feature of all the Germanic languages, though some of the intricacies of the surrounding grammar have been lost in English, which means people have been talking that way for well over a millenium, possibly well over two.

  150. @nht:
    >The “fact” is that saying that minority “welfare queens” profit off the backs of “honest Americans” has long been a staple of American conservative politics from even before Reagan help coin the term.

    If Jorge were American, your judgement of what he meant by the term “honest Americans” would be much more likely to be correct. As it is, he is not necessarily aware of the political nuances carried by the term.

    So Jessica’s analysis, that he is stating that “the intersection of the set of Americans who are honest and the set of Americans who were negatively impacted by Obamacare can be sized as “many.”” is more likely the correct one.

  151. @Jorge:
    >Especially after Eric described the outrageous effects of Obamacare on him, his wife, and presumably many other honest Americans.

    Obamacare is horrible and the Democratic party is evil. But don’t take my word for it, or Eric’s. We both subscribe to schools of thought that are more likely than average to think those things. I happen to think that that’s the case because those schools of thought correspond to reality more than the school of thought that the average person subscribes to, and that the Democratic party and Obamacare are in fact Very Bad Things, but I could just be biased, or even stark raving mad. So investigate for yourself.

  152. Obamacare is horrible and the Democratic party is evil.

    As a republican I disagree. There are “honest Americans” on both side of the aisle and the Democratic Party is not “evil”. There appear to be a higher percentage of racist asshats in the GOP than before because a lot of moderates have been chased out the party.

    Given I was nominally in favor of Romneycare before it became Obamacare I may dislike the implementation but not the objective of making sure more Americans have health insurance. That Obamacare is broken is partially the fault of the Republican Party as we really wanted it to fail.

    But Romney is, of course, a RINO and so we are the party of Trump. No thanks. What is heartening is that a number of prominent republicans have said hell no to Trump. I’m sure they are all RINOs too.

    What is “evil” is the constant polarization of the electorate for higher radio and TV ratings. But the US political system has always been polarized so I guess it’s working more or less as intended as the media has always done this.

    What is apparent is that one party currently engages in using code words to fan the base to higher levels of racial and cultural intolerance.

    Someone castigating a generation on a lack of “critical attitude toward language” doesn’t get ANY slack on language use.

    Especially one that engages in name calling because they didn’t realize that Texas was one of the primary drivers of Intelligent Design appearing in Science textbooks nationwide and was too lazy to engage in any “critical thinking” as to why someone might not think the Texas school systems to be particularly enlightened.

  153. @nht
    > I will thank you to not to piss on my boots and tell me it’s raining.

    So let me see if I understand… yes you accept that the language he used does not strictly support the conclusion you derived from it, however, you believe that there is some secret conspiracy language from the evil Reagan people that was underlying his sentence…. He didn’t strictly say what you said he did, but clearly he implied it with secret “dog whistle” language?

    And you accuse him if being uncritical in his linguistic analysis!!

    Perhaps, after you change your boots, you might want to stretch your mind and see if you can find another function of “honest” in that sentence.

  154. @Jeff Read
    > Result: we all got fat.

    Perhaps I am being dumb, but I don’t understand the implication of your metaphor. Are you saying we all talk too much?


  155. So let me see if I understand…

    Nope.

    yes you accept that the language he used does not strictly support the conclusion you derived from it,

    Nope. I never agreed because while there are innocent uses if the phrase, anyone with a “critical attitude toward language” should be aware of the common non innocent uses of the phrase. Especially in the context of American politics.

    So either he was being lax with word selection or he knew what he was saying.

    Is that your stipulation? He did not have a critical attitude to language?

    Fine. I can agree to that.

    however, you believe that there is some secret conspiracy language from the evil Reagan people that was underlying his sentence…

    I am old enough that I was one of the evil Reagan people.

    He didn’t strictly say what you said he did, but clearly he implied it with secret “dog whistle” language?

    The phrase has been used that way in the past. By both sides but predominantly by conservatives to imply the opposition isn’t honest or American.

    Use on the left is often sarcastic although with the abandonment of the middle and the recent democratic convention looking like a republican one from years past they are using it more in the same way. It is “un-American” to be racist is a common meme…

    Dog whistle politics isn’t just done by the right but we do it a lot more often. To say it leaves even conservative minorities uncomfortable is an understatement. Which is why many largely vote democrat when their natural tendency is toward the conservative side of the political spectrum.

    You never know when you’ll fall out of favor and get thrown out of a trump rally for being a little too “swarthy” like that south Asian supporter.

    Are you going to tell me that you are unaware when such tactics and wording is employed against women? Or that dog whistle politics and code words are NOT part of the political process in both sides?

    Why make excuses for lack of critical thinking on what is a commonly employed political tactic not just in the US but all over the world.

  156. @nht
    “I was curious which “smart Greek” said this but googling the quote it appears to be attributed to Einstein.”

    Continued…

    Actually, Socrates (ie, Plato) claimed that you cannot learn to reason, it is just a recollection.

    http://philosophy.about.com/od/Philosophical-Theories-Ideas/fl/The-Slave-Boy-Experiment-in-Platos-Meno.htm

    A more modern view is that you need knowledge (ie, facts and memorisied examples) to construct new reasonings. That is because most reasoning is by analogy (this problem is like that problem except…). Just having a truckload of unsorted facts does not help much.

    An illustration is the abundance of nonsense and quack remedies on the internet when you search for anything medical. With some knowledge, it is fairly easy to get to useful entries. Without it, you end up with gems and infinite dilutions to cure diabetes.

  157. @nht
    “Who did Obamacare hurt?
    “But there are three subsets of people whose policies were canceled and who are likely to end up as losers under Obamacare — people who are self-employed, over 35, white, or some combination of all three. ”

    That seems to be the real pain point: Obamacare benefits poor people.

    A doctor bikes across the country to ask Americans about Obamacare. This is how he ended up feeling hopeful
    http://www.latimes.com/nation/la-na-obamacare-biking-doctor-20160816-snap-story.html

    Outside a convenience store in eastern South Dakota, another woman said — somewhat ashamedly — that everyone in town thought Obamacare and Obama were terrible. “He just gives all the taxpayers’ money away to poor people,” she said.

    And another article:
    http://azjewishpost.com/2016/bike-trekking-ua-doctor-seeks-views-on-obamacare/

  158. Btw, as a non-American I am amazed how the USA was able to mess up such a simple task. Literally EVERY OESO country, and a lot of developing countries, has universal insurance coverage but the USA was able to select the very worst parts of every system available.

    As if they wanted it to fail from the outset.

  159. @nht
    > So either he was being lax with word selection or he knew what he was saying.

    Or he doesn’t live in your butthurt fantasy world of language?

    I think it is funny that you lament the polarization of the American electorate and yet here you argue vehemently in support of one of the fundamental causes of that polarization of the electorate, namely the politicization of plain, clear language to have some secret conspiracy meaning.

  160. @Winter
    > but the USA was able to select the very worst parts of every system available.

    Yup. But before you preen your feathers too much I’d like to remind you that the vast cost of American health insurance subsidizes the health care costs of the rest of the world. It is our high costs that pay for the development of many of the drugs and procedures that bring about the massive improvements in healthcare worldwide.

    Your welcome….

    > As if they wanted it to fail from the outset.

    Maybe, certainly some have made this accusation. I think rather it is more meant as a step along the road to universal healthcare coverage. Why? Because the massively overspending, bankrupt American government definitely needs to add ANOTHER entitlement to our list of unfunded obligations.

  161. As if they wanted it to fail from the outset.

    There is an apocryphal story about how a power company worked in good faith with environmental organizations to site a nuclear power plant and finally selected a site based on their recommendations after many hearings and studies…on a major fault line.

    Which they then lobbied to have shut down due to earthquake risks.

    I say apocryphal because my recollection was I read it in some Pournelle book (in author comments or something) and while it probably was Diablo Canyon the significant fault line was found after approval…

    Makes for a good story anyway and it’s what comes to my mind when I look at Obamacare. How do you take a reasonable objective (moar health insurance for everyone) and end up with that?

  162. How do you take a reasonable objective (moar health insurance for everyone) and end up with that?

    Cue the story of the paperclip-making robot.

    Alternate cue: David Friedman’s favorite joke about incentive incompatibilities. (The one with Jose robbing a bank.)

  163. @Jessica
    “It is our high costs that pay for the development of many of the drugs and procedures that bring about the massive improvements in healthcare worldwide.”

    Yep, but I was not aware that it was necessary to keep so many Americans uninsured just to develop drugs for us. It is indeed the ultimate sacrifice to let your poor people die in agony just to keep our poor alive.

    I must admit that some other explanations for the plight of the uninsured keep cropping up in the back of my mind.


  164. I think it is funny that you lament the polarization of the American electorate and yet here you argue vehemently in support of one of the fundamental causes of that polarization of the electorate, namely the politicization of plain, clear language to have some secret conspiracy meaning.

    Yes, and conservative fascination with Clinton’s health and “physical stamina” has nothing to do with the fact that she’s female. They are simply concerned with her wellbeing.

    Saying that Clinton lacks the physical strength and stamina to take on Isis is because you really need to be able to bench at least 200lbs to be a good president.

    /s

  165. @Winter
    > to let your poor people die in agony just to keep our poor alive.

    That isn’t what happens. Every hospital is obliged to treat anyone who arrives at their door. Now it might reduce people to bankruptcy but that is a different question. However, you are certainly right, the healthcare industry is seriously messed up here. The solution is a free market in healthcare (and of course charity for those who can’t pay their way) but we haven’t done that for fifty years.

    The idea that we just need to set up government healthcare like the Dutch totally ignores the fact that Americans massively subsidize the world’s healthcare and so such isn’t possible without it costing massively more than it does the Dutch government.

    Again, you’re welcome.

  166. @nht
    > Yes, and conservative fascination with Clinton’s health and “physical stamina” has nothing to do with the fact that she’s female.

    You really do live in a different world than me. You say you are a Reaganite, then you will surely remember that there was great concern about his age and ability to carry out the job when he had that great line in the debates “I am not going to exploit, for political purposes, my opponent’s youth and inexperience”.

    So yeah, questions about the physical stamina are perfectly legitimate and not at all gender specific. If the talk is true that HRC can only go for a few days before taking the weekend off to recover, then that is a perfectly legitimate question about her ability to handle the most stressful job on the planet. It is nothing to do with her bench press.

    And FWIW, from what I can see the small number of people who want her to lose because they can’t imagine a woman president is much outweighed by the number of people who want her to win because they want to have a woman president. FWIW, I think it would be awesome, symbolically speaking, to have a female president. However, we have 150million women in American, and she is he best we can do?

    And for you to think that is sexism? You have to live in a world of micro aggression and safe spaces to believe that kind of nonsense.

  167. @Jessica
    “Every hospital is obliged to treat anyone who arrives at their door. ”

    The biggest burden for individual health and the budget by far are chronic conditions. Inadequate care for chronic conditions is what is the most often cause of “dying in agony”.

    @Jesdica
    “The solution is a free market in healthcare (and of course charity for those who can’t pay their way) but we haven’t done that for fifty years.”

    I have yet to see an example where this worked. If you have one, please share it with us. The costs of health care is around 9% of GDP for one of the cheapest systems, that of the UK.
    http://data.worldbank.org/indicator/SH.XPD.TOTL.ZS

    Given that amount I really doubt whether charity can cover the poor. If I was an American, I would demand solid proof before I would try your solution.

  168. @Winter
    > The biggest burden for individual health and the budget by far are chronic conditions.

    Yes, and most chronic conditions are extremely cheap to treat were it not for the massive costs imposed by the government.

    There is a big to-do over here because a company raised the cost of an automatic epinephrine injector to $500 from $100. But it is plain that the only reason it costs so much is there is effectively zero competition. Such a device should cost less than $5 at WalMart, but it doesn’t, it costs 100 times as much because of the massive regulatory, patent and rent seeking overload caused by the present system. The idea that the government taking over will make for a smooth efficient system and distribution of resources belies all the evidence about the US Federal government.

    But back to my main point, since Americans absorb most of the R&D cost, that means that those poor people, supposedly dying in the streets, is because they have to pay medical care at prices inflated by subsidizing rich Dutch guys like yourself.

  169. The solution is a free market in healthcare (and of course charity for those who can’t pay their way) but we haven’t done that for fifty years.

    Your proof?

    Thought so. Another market fundamentalist.

    Now, go back and reread Arrow 1963 until you grok why markets cannot solve the healthcare problem.

  170. @jessica

    It has not been true that she needs to take a weekend off after a few days of campaigning. She was evidently blitzing the fundraising circuit in August instead of holding rallies. Trump did well in July fundraising so it appear this was the response.

    Easily researched on the net. She did 5 fundraisers over the weekend of the 20th and then headed to California to do 7 fundraisers over 3 days last week.

    That was over Bill’s 70th birthday weekend. Which presumably meant another party to attend.

    This past weekend was more fundraisers in the Hamptons.

    So what made you believe so readily that she lacks stamina? Why do you think this is a “valid” concern when the Internet is available for you to see just what she was doing over the recent weekends for yourself?

    She’s probably rocking a schedule tougher than his. Doesn’t mean I’ll necessarily vote for her but I can dismiss misogynist BS when I see it.

    If I live in a different world than you do its one with Google so I don’t have to trust what either side says without doing at least a 30 second fact check.

  171. @Winter on 2016-08-29 at 11:20:17:

    >I have yet to see an example where this worked. If you have one, please share it with us.

    “When I was a general surgery resident at Roosevelt Hospital in New York in the 1960s, attending surgeons there would charge their wealthy Park Avenue patients relatively high fees while making themselves available to perform surgery free of charge on the hospital’s indigent patients.”

    The linked piece is rather lengthy, but highly recommended to anyone who doesn’t see any reason to expect socialized medicine to work any better than socialized anything else. It will provide many good arguments against those who believe otherwise.

  172. @Jessica
    “Yes, and most chronic conditions are extremely cheap to treat were it not for the massive costs imposed by the government.”

    No, the numbers do not support your view.

    Eighty-six percent of all health care spending in 2010 was for people with one or more chronic medical conditions
    http://www.cdc.gov/chronicdisease/overview/

    These numbers are comparable in other countries.
    http://www.euro.who.int/__data/assets/pdf_file/0006/96468/E91878.pdf (a book on the subject)

    http://www.commonwealthfund.org/publications/issue-briefs/2015/oct/us-health-care-from-a-global-perspective

    @trying2b-amused
    As the studies above show, the costs are not in the one time hospital care. We are talking about income transfer of ~5% of GDP to cover chronic illnesses in the poor. I have serious doubts that charity can cover such amounts.

  173. @Foo

    Belated thanks! If I understand it correctly: Agile works when the cost of breaking changes is low. That makes sense. One could argue that agile technical practices (as found in methods like XP) help drive down the cost of breaking changes, but there is probably a limitation to how much that cost can be minimized depending upon a host of factors (size of code base, size of user base, complexity of the feature set, type of software, etc.).

  174. @Jeff Read
    > Thought so. Another market fundamentalist.

    Wait, wait, wait… you didn’t give me a chance to answer!

    First, I am not a “market fundamentalist” I am a “freedom fundamentalist”. To me the government telling me I can’t provide for my healthcare in the way I choose is no different than them telling me I can’t marry the person I choose.

    As to the evidence? We are surrounded by it. Fundamentally the problem with healthcare is that it is too expensive. If it cost a family $50 a month for healthcare at full price, health insurance would cease to be relevant as a political issue. The EpiPen I just mentioned isn’t just over priced, it is two orders of magnitude more than what you would reasonably expect a free market to produce.

    Anyone who thinks that the way to reduce costs is by getting the government involved has been smoking too much medical marijuana. The only proven way to reduce costs while at the same time increasing quality is the free market. It certainly isn’t a perfect mechanism, but pretty much every market I know of, when left unfettered, gets cheaper and better quality inexorably.

    And before you go down the “all these foreigners control costs in government systems” lets just be clear: the Dutch (to pick on them) spend 1.2% of their GDP on their military, which the USA spends 3.3% . Is that an example of the prudent dutch and profligate Americans? No, of course not, it is because the American military protects the whole of western Europe — we are subsidizing the Dutch military. It is no different when it comes to medical care, and the Dutch, and all these other foreign governments had better hope we don’t withdraw that subsidy any time soon.

  175. @nht
    > It has not been true that she needs to take a weekend off

    But that doesn’t matter. The point of the discussion is the motivation for making this claim, not the veracity of the claim. You claim it is because she is a woman, I claim it is because they hate her for what she stands for not her chromosome configuration.

    By no means am I suggesting that their accusation is in good faith. However,you claimed that it is a dog whistle for her gender, but there are perfectly obvious and reasonable alternative expectations rather than the tin foil hat one.

    HRC is an absolutely appalling person, so there are plenty of reasons to hate her irrespective of her gender.

    And FWIW, as a female what I find rather galling about her presidential bid is that the only reason she is in the running for the presidency is because of who she is married to. So much for female empowerment.

  176. @Jessica:

    The idea that the government taking over will make for a smooth efficient system and distribution of resources belies all the evidence about the US Federal government.

    Or as I like to put it [with hyper-saccharine cheer and a cat-that-ate-the-canary grin], “Yes please! Nothing would make me feel more confident in good outcomes for everyone’s health than placing it under an organization which combines the speed of the DMV with the personal compassion of the IRS!”

  177. @Winter
    > No, the numbers do not support your view.

    But the numbers don’t reflect a free market pricing. Chronic care items are particularly subject to massive price inflation since they tend to be relatively small costs but cumulative.

    For example, diabetics are utterly screwed. The materials they need, when produced in a highly competitive market would be extremely inexpensive. Lancets, test strips, insulin cost marginally pennies whereas today diabetic supplies easily cost hundreds of dollars a month.

    Chronic drugs such as statins are massively overpriced because of rent seeking, patents and fat and ugly regulatory agencies. One of the best drugs to reduce the risk of heart attack is 81mg of aspirin. Which costs a few dollars for a years supply. Statins however are not free (as in speech), and are consequently hundreds of dollars a month.

    Similarly, for nursing care, massively over priced. Nobody can care over the cost of that ten dollar bandage that should have cost 10 cents, but in a competitive market they absolutely would care.

    I could go on. But you can’t look at static prices now and assume that the politburo’s negotiating team does as good a job as a hundred million coupon clippers.

    I’ll say it again. The US medical care industry is a mess. But everyone is focused on insurance and organization and who is in control and all these other peripheral issues, when the problem is much simpler. The problem is that the prices are MASSIVELY, like 10x to 100x inflated. The price is the problem, and free market competition is the only proven solution to price quality balance.

    If health insurance for a family cost $50 a month, nobody would even be talking about it.

    1. >But the numbers don’t reflect a free market pricing.

      Of course they don’t. But since when did evidence of government failure stop a statist from arguing for more government?

  178. @Jessica
    “But the numbers don’t reflect a free market pricing. Chronic care items are particularly subject to massive price inflation since they tend to be relatively small costs but cumulative.”

    The numbers are not much less in ANY country. In Europe, chronic diseases take up ~75% of health care costs. The absolute costs are far lower, but the relative burden is still very high. These high costs are not for drugs, but for care, that is, highly educated humans have to provide high quality medical care regularly over a very long period.

  179. @esr
    “But since when did evidence of government failure stop a statist from arguing for more government?”

    The quality of comprehensive medical care is an empirical question. So, all you need is to show some empirical evidence that the free market can deliver quality medical care to all. This include long term care for chronic conditions and excludes just so stories. There are several cases where states have delivered such care over long periods at low cost.

    This is literally, a matter of lif and death. So, just show us the empirical evidence free markets can deliver this too.

  180. @esr
    “Research medical tourism.”

    Which does not cover daily care for chronic ailments (over 75% of health care costs). Also, this does most emphatically does not deliver quality care to everyone.

  181. Free market advocates decry statists right up to the point where my shadow runners kidnap your top researcher and destroys your R&D lab in response to your developing products based on my trade secrets.

    Any sufficiently large corporation becomes a state able to enforce its will. @see East India Company

    I think patent protection is better than Apple having ninjas to go after Samsung.

    And as far as statins go…many are now generics. Given the length of time for FDA approval the period of protection for drugs is probably reasonable.

  182. Which does not cover daily care for chronic ailments (over 75% of health care costs).

    You keep repeating this. Your initial claim was that individuals with chronic ailments incurred most of the expenditures, which is not at all the same as saying that those ailments are what are expensive to treat.

  183. @nht
    >Free market advocates decry statists right up to the point where my shadow runners kidnap your top researcher

    I am not an anarcho capitalist and I think that one of the few legitimate functions of government force is to enforce property rights, including preventing or prosecuting the above.

    > Any sufficiently large corporation becomes a state able to enforce its will. @see East India Company

    You are aware that the East India Company was granted a monopoly on trade with India by the British government, and that the British Army and the Royal Navy was used to enforce that monopoly. These types of monopoly grant are exactly the type of thing I am railing against.

    > And as far as statins go…many are now generics. Given the length of time for FDA approval the period of protection for drugs is probably reasonable.

    Reasonable? Government force is causing the massive rise in the price of life saving drugs and you consider that reasonable? This government situation is like a guy stabbing you in the guts and then driving you to hospital, and then expecting to be paid double taxi rates and demanding your gratitude for saving your life.

  184. @Christopher Smith
    People with chronic illnesses get a lot of diverse health problems. Kidney failure and amputations for diabetes, suicide and general neglect in mental illnesses. In the end, chronic illnesses tend to wear down all organ systems. That is why you cannot separate out each symptom.

    Read the book I link to above.

  185. @Winter
    > Which does not cover daily care for chronic ailments (over 75% of health care costs). Also, this does most emphatically does not deliver quality care to everyone.

    But Winter your argument is basically this: governments the world over make it essentially illegal or impossible for a market in medical care to develop. Given this situation you say “Hah! look there are no examples of private medical care, therefore they don’t work.”

    In the USA it used to be that the US Postal Service had a legally enforced monopoly on delivery of letters and packages. Only after legal changes allowed competition did a thriving alternative business in delivery arrive, so thriving that USPS is a floundering blob now. Before the legislation one might well have argued “Look there are no private delivery services, so obviously we need the government to run the mail.” The same was true in the UK when phone service was provided by the Post Office until Thatcher sold off British Telecom and released the all powerful free market in phone services.

    In this situation where the government has locked up medical care (either directly as in much of Europe, or indirectly as in the corporatist, rent seeking, patent enmeshed, over regulated system we have here) all we can do is look on the margins at the parts of the system the government has not consumed into its gigantic machine.

    Medical tourism demonstrates that acute care can easily cost 1-5% of what it does in the USA with equal or better quality. Unpatented generic drugs or various other unregulated medical or close to medical services demonstrate the massively reduced costs of these, again typically 1-10% of what they cost in the government world. To give another example — I can get a flu shot at Walgreens for $9.99, but if I get a vaccination through my doctor it can easily be billed as $150 or $200. Or my doc recommended I get one of those MRIs to scan the thickness of your arteries. It cost $90, because it was a private business that didn’t take insurance. If it had been through insurance it would easily have been billed at $1000.

    So again, when you look at these unregulated parts on the margins and we can see what an medical system without the bully heavyweight of government tells us.

    You mentioned skilled medical staff, but what you ignore is that competition makes the cost of labor cheaper in several different ways: for example, more efficient practices mean the professional needs less time per patient, and less qualified people can be used for lower skilled parts of the job, or specialization means that people can do the job better because of more practice and better tools, and if people pay for the services they are less wasteful to name just a few of the most obvious cost reducing measures. This is not some brilliant idea I came up with, it is what happens every day, all the time when the power of competitive forces are brought to bear on a problem. And it is those powerful cost reducing, quality enhancing processes that are specifically stymied by the government’s sticky, messy paws being in the pot.

    Of course as soon as you start talking about “cost reducing” medical care everyone freaks out. “I want only the best medical care”. However, lets not forget that free markets don’t just reduce cost, they increase quality too. Not out of altruism, but out of greed. They want to attract more customers and so have to provide less for more.

  186. You are aware that the East India Company was granted a monopoly on trade with India by the British government,

    And you are aware that the monopoly was a rather weak one in the beginning considering that the Dutch were already operating in the region, more established and quickly formed the Dutch East India Company in response.

    Portugal was also there but a declining power at that point and England was not yet a colonial power.

    It wasn’t really until the Dutch got smushed by the Royal Navy for running guns to a bunch of breakaway English colonies that the VOC took significant losses from British forces.

    At the height of its power the VOC had 40 warships and 10,000 troops.

    And there was the Danish East India Company, French East India Company, the Ostend Company and Swedish East India Company as well.

    In the absence of any other controls the ships of the various companies attacked each other to protect trade interests and markets…

  187. @esr: “But since when did evidence of government failure stop a statist from arguing for more government?”

    …and then they turn around and accuse us Christians of irrational belief in the absence of evidence. *sigh*

  188. @nht
    > And you are aware that the monopoly was a rather weak one in the beginning…

    Can we try to stay on topic please?

    The point you were claiming to make is that the East India Company was an example of what we would see in an unfettered free market. But, as you correctly say, it was a game of national interests played out by proxy corporations. Consequently it has absolutely no bearing on the original matter you used it to illustrate. Even though the East India Company was a “corporation” it was not a corporation anything like what we mean by corporation today.

    It certainly took a while to establish its monopoly by violence through the support of their private army and the backing of the British Army and the Royal Navy, but that is the point, isn’t it? It was a monopoly enforced by violence (explicit violence rather than the more hidden violence that governments use today to enforce monopolies) and so has absolutely no bearing whatsoever on how a free market of voluntary exchanges takes place.

    It is valuable also to observe that their violence, even with support of the government was not effective and the company eventually collapsed and their violent monopolistic ways were taken over by the British Government. The Indians only freed themselves of the overwhelmingly powerful coercive forces of the British government when it had been so badly weakened by the second world war that it could not longer keep them natives in check. Ah, the white man’s burden can be so overwhelming.

  189. @Jessica – “…Or my doc recommended I get one of those MRIs…”

    Yup, similar story here. Insurance Co. was claiming it would be ~$3000 for an abdominal image, yet there was a private business only a 1 hour drive away that did it for about $250 and sent the results straight to the doc.

    That was an eye opener for me.

    Since then, shopping around for medical deals has become my default practice…saving me a fortune. The cartel-esque pricing we get spoonfed is criminal.

  190. @Jessica
    “But Winter your argument is basically this: governments the world over make it essentially illegal or impossible for a market in medical care to develop.”

    Sorry, but most people in this world are not protected by the likes of the FDA. Literally Billions of people live in countries like China, India, Nigeria, or Russia, where almost anything goes. So pointing out that things are not allowed in North America and Europe does not cut it.

    There were times and places where growth hormones were marketed as a cure for malnutrition in children and formula to replace breastfeeding in places without clean water. Surely, if the free market would have given people quality care for all, it would have been in these places.

    I see this argument as bad excuses. In all the places where there is or has been no limits on the market of health care, surely one should have come out with your quality care for all?

  191. @Jessica
    ” It was a monopoly enforced by violence ”

    More accurately, these were buccaneer/piracy operations. Plain robbery.

  192. @Winter – “In all the places where there is or has been no limits on the market of health care, surely one should have come out with your quality care for all?”

    Well, it isn’t going to magically spring from the earth…there has to be economic incentive and related expertise to generate those kinds of healthcare resources. Waving your hands around and claiming that a free market has failed to ‘provide’ something is cargo cult nonsense.

  193. @Dan
    “Well, it isn’t going to magically spring from the earth…there has to be economic incentive and related expertise to generate those kinds of healthcare resources.”

    I really do not understand what you mean. There is a demand, there is supply, there is knowledge, there is money to invest, there are few or no rules. What else is needed?

  194. Depending on the example countries you have in mind, there will probably *not* be the economic incentive for investment simply because the nation/population isn’t wealthy enough. For those places, there are charitable/altruistic orgs like Doctors Without Borders, Red Cross etc that try to tackle the top life threatening issues facing the native people….but there simply isn’t enough wealth to justify investing in healthcare infrastructure like we do in the western world. They can’t afford it.

    As much as you may find it lamentable, there are obvious reasons why these ‘free’ areas don’t enjoy the same standards of healthcare as we do…cargo-cult expectations of ‘free markets’ aside.

  195. @Jessica:
    >Unpatented generic drugs or various other unregulated medical or close to medical services demonstrate the massively reduced costs of these, again typically 1-10% of what they cost in the government world.

    I’ll note that the bill my parents paid for my mom’s hospital stay when I was born in the 80’s was about 10x what my grandparents paid when my aunt was born in the 40’s (including a cesarean in my aunts case!) *after* adjusting for inflation.

    > Even though the East India Company was a “corporation” it was not a corporation anything like what we mean by corporation today.

    That’s becoming less true with every passing year , though. Something like mercantilism (called “capitalism” by both its beneficiaries and by its opponents in the left, for different reasons) is making a resurgence, and capturing markets left and right, often with government aid.

  196. It certainly took a while to establish its monopoly by violence through the support of their private army and the backing of the British Army and the Royal Navy, but that is the point, isn’t it? It was a monopoly enforced by violence (explicit violence rather than the more hidden violence that governments use today to enforce monopolies) and so has absolutely no bearing whatsoever on how a free market of voluntary exchanges takes place.

    In the absence of control by “hidden violence” from governments this is what you end up with. Without this there is no “free market of voluntary exchanges”.

    And yes the HEIC failed…after 100 years of rule…

  197. @jessica

    In the USA it used to be that the US Postal Service had a legally enforced monopoly on delivery of letters and packages. Only after legal changes allowed competition did a thriving alternative business in delivery arrive, so thriving that USPS is a floundering blob now.

    “For FedEx alone, the post office delivers an average of 2.2 million packages a day, or about 30% of the express-mail company’s total U.S. ground segment.

    UPS won’t specify how much of its shipments go through the post office, but a regulatory filing indicates those types of lightweight shipments accounted for 40%—or about 37 million packages—of its total increase in ground shipments in 2012.”

    http://www.wsj.com/articles/u-s-mail-does-the-trick-for-fedex-ups-1407182247

    So these successful thriving alternative businesses are successful in part because they rely on the USPS for the last mile delivery part…

    It is true that competition is good. It does not appear to be true that unfettered competition is good.

  198. @nht
    > In the absence of control by “hidden violence” from governments this is what you end up with.

    If you don’t understand the difference between the enforcement of property rights and capricious arbitrary violence of might makes right, then I think we are done here. I will say again, I am not an advocate of anarcho capitalism.

    In regards to the USPS, express delivery services as you know the USPS is heavily subsidized by the government and has a monopoly on your mailbox, so I’m not sure what exactly you expect. Competition transformed mail delivery, but, as usual, the government stopped short of releasing the full power of it. They just had to hang on to one last piece of control.

    The best thing that could happen to the USPS is to eliminate the universal mandate, eliminate their monopoly and sell the organization on the public markets. That is the only opportunity that organization has to thrive and grow. And we’d all be better off for it.

  199. What you would end up with is mail service ending for areas that are unprofitable to service.

    What I understand is that enforcement of property rights currently includes intellectual property rights which leads to “monopolies” and “rent seeking”.

    Something you advocate against.

    As an aside, the violence was not “capricious” at all but done to maximize profits for the shareholders in what was then a free market environment without controls.

    Removing annoyances like the FDA, patent protections and other forms of governmental controls that provide the environment that pharmaceutical companies operate under that provide with mostly safe drugs and relatively cheap drugs after a decade leads to what?

    Where do governmental regulations end in your scheme? Ignore violence for the moment.

    What keeps major pharmaceutical companies from employing monopolistic tactics like collusion, dumping or tying? And why those limits and not IP protection?

    Monopolies are often the natural result of free markets because of economies of scale and the ability to crater the revenue stream of market entrants until they run out of money and go away.

  200. @nht – “…What you would end up with is mail service ending for areas that are unprofitable to service….”

    Which would probably give rise to profitable businesses providing PO boxes, collection stations etc for those remote people.

    Solutions abound….given fresh air to breathe.

  201. What you would end up with is mail service ending for areas that are unprofitable to service.

    1. Duh.

    2. Um, DUH

    3. This is as it should be. That it is unprofitable (and no one is proposing a market failure here) is in itself evidence that it should not be happening. Period.

  202. @nht
    > What I understand is that enforcement of property rights currently includes intellectual property rights which leads to “monopolies” and “rent seeking”.

    Patents, monopolies and rent seeking are all different things. For me it is plain that patents are an absolutely dreadful infringement on liberty, but I will grant you that many people disagree with me. One thing is for sure, they massively raise the cost of drugs and medical care.

    And FWIW, the few studies that have been done on the subject indicated that patents decrease innovation in drugs, which rather takes away even the one shred of decency on which such force is based. But I have written about a million comments on that here before, so let’s not rehash.

    > [unloading all the govenrnent crap] leads to what?

    A free market in drugs and medical devices where you take your advice on safety and efficacy from your trained professional doctor. Not sure why that is hard to understand.

    I know nada about maintaining my car, can barely put gas in the tank. But I know this guy Frank who is knows all about that stuff and I give him money and he gives me excellent advice. I used to have another guy who did the same, but he sucked, so I stopped giving him money and started giving it to Frank instead. My car never ran better.

    > Where do governmental regulations end in your scheme?

    Why do you have such a small vision of the world that you think that only the government can regulate things like medical care? There are perfectly viable alternatives that allow me to choose. I know only a little about how bodies and disease work, but I know his guy, Dr. Glenn, who knows all about that stuff. He went to college for decades. He tells me what to do. I promise you, I trust Dr. Glenn with my healthcare a lot more than I trust some political appointee of Hilary’s medical politburo (or, come to that Trump’s politburo.)

    > What keeps major pharmaceutical companies from employing monopolistic tactics like collusion, dumping or tying?

    Competition and independent, professional advisers. It really isn’t complicated.

  203. @nht
    > What you would end up with is mail service ending for areas that are unprofitable to service.

    I mean really, what do you think would happen if the people in unprofitable areas couldn’t get mail? Do you think they’d try to innovate a solution, or do you think some innovator would do it for them? OR do you think they’d all sit around in their houses never getting letters anymore.

    I mean honestly, think about it. If the government couldn’t solve this problem for whatever reason, what do you think would actually happen?

    Do you really think the government (or pseudo government in the case of USPS) are the only organization that can fix things?

  204. A free market in drugs and medical devices where you take your advice on safety and efficacy from your trained professional doctor. Not sure why that is hard to understand.

    The doctor is going to tell you to use the drugs they gets a bonus to prescribe? Even assuming that they are operating in good faith what doctor has enough time to actually conduct the safety research to insure that the drug DOES do what the marketing says it does and doesn’t have some terrible side effect that is being hidden?

    To assume that your local doctor can judge the efficacy and safety of a new drug is simply stupid. EVEN if they had the data available to them, which they will not, they don’t have the time to personally verify the claims of all the drugs, equipment, etc that undergoes FDA testing.

    I know nada about maintaining my car, can barely put gas in the tank. But I know this guy Frank who is knows all about that stuff and I give him money and he gives me excellent advice. I used to have another guy who did the same, but he sucked, so I stopped giving him money and started giving it to Frank instead. My car never ran better.

    And Frank has done the crash testing to tell you which cars survive better in accidents? He also knows that specific airbags are defective and should be recalled? Or the power to force a manufacturer to recall unsafe cars and fix them?

    Why do you have such a small vision of the world that you think that only the government can regulate things like medical care? There are perfectly viable alternatives that allow me to choose. I know only a little about how bodies and disease work, but I know his guy, Dr. Glenn, who knows all about that stuff. He went to college for decades. He tells me what to do. I promise you, I trust Dr. Glenn with my healthcare a lot more than I trust some political appointee of Hilary’s medical politburo (or, come to that Trump’s politburo.)

    How do you know Dr. Glenn went to college for decades? Or that he studied medicine? What are his qualifications? Do you know which drug companies he works for?

  205. Speaking for Pharmaceutical Companies

    Many doctors earn money from pharmaceutical companies for speaking formally about a specific drug, either to a group of patients or to other doctors. Device makers may also pay doctors, often surgeons, to speak about a device such as a hip implant or a pacemaker.

    A common speaking fee is $1,000 per talk, though many make more than that. If the talk is not near the doctor’s home location, then airfare, food and lodging are often provided in addition to the base fee. Most doctors speak on more than one occasion for the same drug or device.

    This kind of work can earn a doctor who keeps at it a lot of money. ProPublica, a nonprofit social justice newsroom, has been tracking payments made to doctors from pharmaceutical companies since 2010. According to their publicly searchable database, some doctors made more than $100,000 in a single year for speaking engagements.

    The Centers for Medicare and Medicaid Services also publishes all payments over $10 made from medical companies to doctors and hospitals under the Sunshine Act, part of the Affordable Care Act. According to that database, the highest payment made in the category of “compensation other than consulting,” which includes speaking fees, was $4.25 million in 2013. That category includes other educational services, so it may not necessarily be a speaking engagement.

    So without the Sunshine Act, how do you know the very genial Dr. Glenn that you trust SOOOO much isn’t getting a $100K to push a drug or procedure you don’t need and might be dangerous?

  206. @nht
    > The doctor is going to tell you to use the drugs they gets a bonus to prescribe?

    This is a bug in the cannon of ethics for doctors. No doctor should be getting kickbacks like that. INAD so I don’t know how that works, but I am very dubious about this claim. Don’t worry though, I’ll ask Dr. Glenn.

    > Even assuming that they are operating in good faith what doctor has enough time to actually conduct the safety research

    LOL. I can’t believe your thinking is so trapped within that tiny box. Obviously Dr. Glenn isn’t going to conduct the research, what he is going to do is use his professional expertise to judge which sources of data on drug efficacy are worth depending on (or which sources of data on sources of data on drug efficacy are worth depending on.) And then he is going to use that to advise me. He might, for example, use a text book approved by his professional organization, or subscribe to a database that has a reputation within the medical community for reliability. It is, after all, what he does already.

    However, I have a question for you…. that guy Tim Cook is a pretty hard worker, but how the hell does he find time to solder the CPUs onto all those iPhones?

    > How do you know Dr. Glenn went to college for decades? Or that he studied medicine?

    I went to a trusted source of doctors who validated him medical education and verified that a professional organization like the AMA had certified him as a qualified doctor.

    Again, why do you think only the government can do stuff like this? This is really not complicated.

  207. @nht
    > So without the Sunshine Act, how do you know

    Why can only the government do this? And when, pray tell, did you last consult this database of information, and how exactly did you judge the information? If you can’t use it what use is it to you?

    Wouldn’t you rather have an expert judge it for you? Does that expert need to be a civil servant? Can only a civil servant ensure this information is produced or are there possibly other sanctions that can be applied privately to make it happen?

    Jump out of the box dude. There is a big happy colorful world out here without the dull, gray, grinding oppression of the government machine.

  208. @nht:

    In the absence of control by “hidden violence” from governments this is what you end up with. Without this there is no “free market of voluntary exchanges”.

    As a free-market anarchist, off the top of my head I can think of multiple fields of human activity which are so dominated by a “free market of voluntary exchanges” that the universal phrasing used in this statement is absolutely preposterous. I am going to limit myself to only the three most illustrative examples:

    First, eBay, Etsy, RedBubble, et.al. Because no government claims the exclusive force to prevent vendors and purchasers alike from masking or changing identities, forging ratings, charging back or otherwise withholding remittance, overcharging, or selling across real-world jurisdictional boundaries, online person-to-person sales must be fraught with hazards, even if not direct violence. (Admittedly, all of the parties involved cannot directly exert force on each other to block this malfeasance, but even without the online element an offline rummage sale should suffer from many of the same flaws.) Does this sound correct? [I mean both ‘correct application of your thesis’ and ‘conclusion validly matches real world experience’.]

    Second, we have personal relationships, e.g. dating. [Your thesis is not phrased in a way to limit itself to only the economic sphere. Correspondingly, I have applied it to all markets, even those with limited or fringe financial connections.] Since Western, democratic governments exert no influence over this realm of life, by your statement we must instead see overt violence by individuals as the predominant factor. Rather than courtship, the primary means of determining romantic relationships would be people clubbing one another over potential mates or overt subjugation of said mates: there are no means to establish a romantic relationship which does not expose oneself to imminent, unavoidable violence. All long term relationships, romantic or otherwise, are primarily defined in terms of dominance rather than compassion or mutual respect. Does this sound correct?

    Finally, religion. Again, most Western government have withdrawn from this influencing this sphere of their citizens’ lives, so your thesis would demand overt violence must predominate. All arguments over whether creationism should be taught in American schools or whether religious monuments may be erected on public grounds were settled through guns & private armies, not lawsuits. Denouncing the faith of another or changing ones own faith practice carries such a life-or-death risk that religious organizations must hire bodyguards for preachers and new converts alike. Does this sound correct?

  209. @ nht

    I didn’t know the term “honest Americans” had racist connotations. I apologize.

    As for what I did mean: Jessica called it right. I was referring to “the intersection of the set of Americans who are honest and the set of Americans who were negatively impacted by Obamacare”. I included the adjective “honest” to emphasize that honest people don’t deserve to be wronged by politicians (or by anyone, for that matter).

    Now, you say that Obamacare has benefitted some honest persons. Does that atone for the damage it’s done to other honest persons?

    Someone castigating a generation on a lack of “critical attitude toward language” doesn’t get ANY slack on language use.

    You’re right. I’ll be more careful.

    Especially one that engages in name calling because they didn’t realize that Texas was one of the primary drivers of Intelligent Design appearing in Science textbooks nationwide and was too lazy to engage in any “critical thinking” as to why someone might not think the Texas school systems to be particularly enlightened.

    That wasn’t me.

    @ Jon Brase

    Obamacare is horrible and the Democratic party is evil. But don’t take my word for it, or Eric’s. […] So investigate for yourself.

    I think that’s generally the correct approach, but Eric’s explanation of the problem makes perfect sense. I don’t feel a need to conduct my own research in this case.

    @ esr

    …since when did evidence of government failure stop a statist from arguing for more government?

    “I exploit you – still you love me. I tell you one and one makes three. I’m the cult of personality.” Ever heard that song? If not, you might enjoy it – three-chord rock has its place, right? :-)

    1. >I didn’t know the term “honest Americans” had racist connotations.

      It doesn’t, except in the minds of those ideologically predisposed to see racism everywhere.

  210. @ Jessica Boxer

    Thanks for clarifying my point, and sorry for postponing the resumption of our debate. Where were we? Let’s see:

    You aren’t a programmer.

    I did admit that.

    We abandon the old tried and true ways all the time to embrace new ideas and discoveries.

    What I meant is that discarding knowledge, and even structures for thinking and communicating (such as tenses), is not like discarding programming tools. The latter makes perfect sense, but the former – in my humble opinion – impoverishes our understanding of the world.

    Remember, Spanish schools don’t have spelling class for a reason.

    Alas, Spanish is susceptible to misspellings. They’re increasingly frequent, which saddens and irritates me.

    I’m not an anarcho-capitalist, and I don’t really know the answer to your question.

    My bad. But, from what I gather, you do want less State and more market. So we largely agree regarding political philosophy. :-)

    I am not an anarcho capitalist and I think that one of the few legitimate functions of government force is to enforce property rights…

    This isn’t actually part of our debate, but I’m curious: what would be wrong with a plurality of freely-competing providers of legal enforcement? Why do we need a monopoly on that particular service?

    In the USA it used to be that the US Postal Service had a legally enforced monopoly on delivery of letters and packages.

    I hear you. What they did to Lysander Spooner was utterly unjust.

  211. @Jorge Dujan
    > What I meant is that discarding knowledge, and even structures for thinking and communicating (such as tenses), is not like discarding programming tools.

    I’m not sure what you are referring to here. What has been lost, aside from the basically useless subjunctive.You might also consider that ebonics has a verb structure that has additional capabilities beyond modern English verb structures. Something to consider.

    You want an expressive language, yet you eschew verbification. That seems a little contradictory to me.

    > This isn’t actually part of our debate, but I’m curious:

    I am disinclined to go down that rabbit hole. Another time perhaps.

  212. LOL. I can’t believe your thinking is so trapped within that tiny box. Obviously Dr. Glenn isn’t going to conduct the research, what he is going to do is use his professional expertise to judge which sources of data on drug efficacy are worth depending on (or which sources of data on sources of data on drug efficacy are worth depending on.)

    I find that when I start out with an insult that I’m usually on my back foot.

    The reason that I said this is because the sources of reliable data is not likely to exist.

    Let’s see if we have the scenario correct:

    * We have removed the regulations requiring expensive drug trials over a long period to bring drugs to market.

    * We have removed patent protection from new drug formulations. The assertion is that the advantage for innovation is the ability to be first to market.

    * In order to carry out the drug testing protocols to insure new drugs are, at the least, not dangerous and hopefully efficacious the inventor must delay bringing the product to market.

    * By performing any testing they are letting the drug out of their lab and into the testing community, increasing the possibility that a rival will acquire a sample and reverse engineer it. The more testing over longer periods increases this likelihood.

    If the rival does so and releases their generic version they can:

    * beat the original inventor to market because there is no patent protection.
    * sell at a lower price because they did not have to pay for the research and development costs
    * sell at a lower price because they did not have to pay for drug testing

    This set of circumstances means it is detrimental for the inventor to carry out the same rigorous testing over several years that is currently required before selling the product. Especially since any such trials may indicate that the drug doesn’t work or has some nasty side effect to some part of the population. They would most likely do quick and dirty testing and quietly settle any injury cases with hush money for side effects or death in order to preserve their first to market advantage for as long as possible.

    Ah, but the market will demand such testing!

    Given the number of less than credible “nutritional supplements” that the FDA has to quash I don’t believe that the individual consumers will drive significant desire for long term safety testing. Even if some they do, my belief is that sufficiently many will select a cheaper generic that did not undergo testing over the “brand name” that did undergo trials thereby driving the testing costs even higher.

    Doctors might want more testing but even today pharmaceutical companies pay many doctors significant sums to market drugs to other doctors and the public. It would take a very stalwart Dr. Glenn to not be influenced by well respected (and well compensated) peers who say a particular drug is safe while at a convention in Hawaii subsidized by the pharmaceutical industry.

    Even if there was a private source doing drug testing, access to that expensively collected data will be behind an equally expensive paywall. Are you willing to pay extra for Dr. Glenn to have an account to access the data? At what price point will you stop doing so and go to Dr. Lee instead? She sources generic drugs from overseas for a fraction of the price as the pharmacy that Dr. Glenn recommends. Never mind that 10mg of Rosuvastain is really only 3mg of some sort of medicine and 7mg of additional filler.

    Plus Dr. Lee has paid the AMA extra to give her top billing on their lists of recommendations (@see Angie’s List), has a very nice office with many smiling nurses and all the latest medical gear (funded by her numerous speaking engagements).

    Finally, with no regulatory oversight into testing procedures there is also no certainty that the company doing the testing isn’t engaged in pay to play as well.

    The consumer can’t demand that drug companies do the testing required if they elect to collude and not provide such testing. If a company attempts to buck the oligopoly by providing more rigorous testing they can crater that company’s revenue by selling at a loss until it stops.

    The consumer can elect to not buy drugs (and suffer) or pay whatever the oligopoly demands and hope they aren’t in the small fraction of the population with negative reactions to the drugs. At least their heirs might be well compensated.

    This isn’t theoretical. This is the way it was before these annoying regulations were instituted.

    tl;dr:

    There isn’t a business case to provide for the same level of drug testing required by current regulations. There ARE significant reasons not to perform such testing and for the drug industry to collude to sabotage testing regimes.

  213. Why can only the government do this? And when, pray tell, did you last consult this database of information, and how exactly did you judge the information? If you can’t use it what use is it to you?

    Who can compel doctors and companies to report this negative information? Will an AMA run by doctors for the benefit of doctors require such reporting? It’s not as if YOU really are the customer of an AMA that primarily makes money by selling certifications.

    When did I last consult this database? Never. Haven’t needed to directly (because I’m on cheap generics and not on an expensive set of exotic cocktails) do so but the fact that such information is available keeps things in check and should a malpractice suit occur it could be entered as useful evidence.

    If my monthly outlay in medicine was in the hundreds then I surely would take a look to see if there was something going on. For kicks, I looked for my GP and nope, nothing. Then I looked for my ENT and he has been given $120…probably a conference or trip.

    https://openpaymentsdata.cms.gov

    Go look up Dr. Glenn. If there is a large payment then wouldn’t that be a point of discussion on your next visit?

    Jump out of the box dude. There is a big happy colorful world out here without the dull, gray, grinding oppression of the government machine.

    I am glass half full on a lot of things…but that includes on the ability for charming human predators to cause death and injury for their own profit.

    https://en.wikipedia.org/wiki/Radium_Girls
    https://en.wikipedia.org/wiki/William_J._A._Bailey
    https://en.wikipedia.org/wiki/Jim_(horse)

    That happy colorful world may be particularly glowing because the paint contains radium painted by workers that didn’t know how dangerous it was.

    We are here today not because of a whimsical desire for more regulation but because there were real world problems from not having this kind of protection the free market did not provide.

    I don’t trust any drugs in China* because their government regulators are so corrupt that many high party officials source their baby formulas from overseas. No amount of money, back channel networking or private efforts is sufficiently reliable to want to trust domestic sources. Sure, you might have the power to shoot the plant owner the next day but that’s a cold comfort if your child was harmed.

    It works here because as flawed as the FDA may be, for the most part it does keep us safe from the issues seen in countries without effective governmental protection.

    * Drugs made in China and sold here is likely safe because of our regulatory practices.

  214. @nht
    > Who can compel doctors and companies to report this negative information? Will an AMA

    How about their competition “Doctors for America”. “Only buy medical services from ‘Doctors For America’ certified organizations. You can be confident, unlike those jokers at the AMA, that our doctors aren’t taking kickbacks.”

    > I am glass half full on a lot of things…but that includes on the ability for charming human predators to cause death and injury for their own profit.

    Do you need me to list some of the terrible events that have happened under this supposedly amazing regulatory control? Bad stuff happens all the time. Did you consider that this blessed FDA that you seem to love so much is one of the slowest regulatory approval agencies in the world, and that life saving medications are often unavailable to Americans for decades when they are readily available in other parts of the world. The numbers don’t lie. Literally millions of people have lost their lives because of this delayed approval process. However, you are seduced by the big lights, the headlines of some terrible things that happen while quietly ignoring the dreadful things that could have been prevented were we not choking in the sclerotic grip of the death dealing monsters in the FDA.

    > I don’t trust any drugs in China* because their government regulators are so corrupt

    What have corrupt regulators in a fascist state got to do with our discussion here?

    > It works here because as flawed as the FDA may be, for the most part it does keep us safe from the issues seen in countries without effective governmental protection.

    Does it keep us safe from coral snake bites? They are deadly, however, no antivenin is currently available in the USA because stockpiles have expired and nobody can deal with the huge cost of the regulatory burden of manufacturing more. So if you get bitten by one, you’d better hope it is in backward Mexico, because that is the only place you can get treatment.

  215. How about their competition “Doctors for America”. “Only buy medical services from ‘Doctors For America’ certified organizations. You can be confident, unlike those jokers at the AMA, that our doctors aren’t taking kickbacks.”

    And if DFA doesn’t care either? You cannot assume that professions will self regulate because they haven’t in the past. You can HOPE they self regulate but why do you believe it’s a sure thing?

    Neither DFA or AMA works for you unless you are the primary source of revenue. How much are you paying the DFA to safeguard your interests more than the interests of the doctors?

    Who would the DFA or AMA court? You or the doctors that provide fees for certification?

    Do you need me to list some of the terrible events that have happened under this supposedly amazing regulatory control? Bad stuff happens all the time.

    The difference is that the bad stuff today is more of the anomaly in comparison to the days when snake oil was far more prevalent.

    Literally millions of people have lost their lives because of this delayed approval process.

    Citation needed. And how many were saved because of the delay? If literally millions of people have lost lives due to delay than was saved then this should be easy to document.

    Does it keep us safe from coral snake bites? They are deadly, however, no antivenin is currently available in the USA because stockpiles have expired and nobody can deal with the huge cost of the regulatory burden of manufacturing more. So if you get bitten by one, you’d better hope it is in backward Mexico, because that is the only place you can get treatment.

    “Case reports of patients in which antivenom was delayed show that symptoms can progress to paralysis over a 12-18 hours period. In the last 40 years, there has been one fatality reported when a person failed to seek medical treatment after a bite.

    http://www.poisoncentertampa.org/poison-topics/coral-snake-antivenom/antivenom-faq/

    I’m thinking my odds are pretty good.

    In any case:

    FDA.gov:

    If Licensed Coral Snake Antivenom is Not Locally Available or Not Available in a Timely Fashion
    You may consider using an investigational (non-licensed) imported Coral Snake Antivenom. These are available in some regions, under an Investigational New Drug (IND) protocol with informed consent. The Poison Control Centers may assist you in locating IND supplies of antivenom.
    Use of IND products requires informed consent under the existing IND if your patient is eligible under that IND protocol.
    Alternatively, you may request use of the IND product as an emergency IND for single patient use by calling FDA:
    Business hours – 800-835-4709 or 301-827-1800
    Nights/holidays/weekends FDA emergency operations at 301-796-8240.

    If you get bit in Florida you can get access to INA2013 by going to the Tampa General Hospital:

    http://www.poisoncentertampa.org/poison-topics/coral-snake-antivenom/

    If you are in Texas…they have some Wyeth antivenom left (at least in June 2015) and:

    “The Wyeth antivenin was made using only Eastern coral snake venom, and research performed by Elda Sanchez, of the National Natural Toxins Research Center in Kingsville, suggests it may be not be as effective in counteracting Texas coral snake venom.

    When FDA officials could find no manufacturer to take up antivenin production, they turned to Boyer, who had just developed a new antivenin for scorpion bites. With funding from the National Institutes of Health, Boyer’s lab created a new coral snake antivenin from a mixture of Eastern and Texas coral snake venom, with additional processing designed to minimize allergic reactions.”

    http://www.houstonchronicle.com/news/health/article/Coral-snake-antidote-shortage-poses-ethical-6352084.php

    So while there is a distinct shortage there was still some supply left and a new version is likely more effective for western coral snakes and lower likelihood of allergic reaction.

    So, I’m curious…was it too hard to google to confirm there was no coral snake antivenin in the US before making an easily refutable assertion? And if you were wrong about this, what else are you wrong about that you base your opinion on?

  216. @nht
    OK, thanks for the discussion, that last one is tl;dr. But I thank you for your perspective.

  217. @jessica I’ll make it shorter for you:

    Your opinions are based on supporting facts. Presumably the coral snake antivenin one is one of those since you brought it up although you didn’t want to read that it was wrong. But if you were wrong about that, what else are you wrong about that you base your opinion on?

    I’m not trying to be mean but the hatred of the FDA is based on what? That millions more have died than were saved because the actions of the FDA?

    If you can actually prove that with solid evidence, I’ll change my opinion.

  218. How we went from aphorisms on software development methodology to discussion about medicine, health system and free market?

    Anyway, free market is proven to work if certain conditions are met, and one of them is symmetry of information.

  219. Anyway, free market is proven to work if certain conditions are met, and one of them is symmetry of information.

    And benign actors. Throw in Vanderbilt, Bush, Rockefeller into the mix and you end up with Standard Oil.

    So not “proven to work” in the real world.

  220. And benign actors. Throw in Vanderbilt, Bush, Rockefeller into the mix and you end up with Standard Oil.

    Well if there was any doubt that nht has no clue what he is talking about, he just stomped on the ignorance tripwire. Y’know, I had wondered where the Idiot Ball went when Jeff Read wasn’t using it.

    nht, you are full of shit.

  221. @nht:

    So not “proven to work” in the real world.

    I had a response held in the moderation queue overnight that would address this. Kindly scan back in the discussion and tell me how I’m wrong.

  222. Well if there was any doubt that nht has no clue what he is talking about, he just stomped on the ignorance tripwire. Y’know, I had wondered where the Idiot Ball went when Jeff Read wasn’t using it.

    nht, you are full of shit.

    It’s a lot easier to insult than actually discuss. Because actually providing discussion may lead to refutation. Will you run away when specific examples are disproven?

  223. I had a response held in the moderation queue overnight that would address this. Kindly scan back in the discussion and tell me how I’m wrong.

    Sure. But I’m going to guess it’s going to be tl;dr…

    First, eBay, Etsy, RedBubble, et.al. Because no government claims the exclusive force to prevent vendors and purchasers alike from masking or changing identities, forging ratings, charging back or otherwise withholding remittance, overcharging, or selling across real-world jurisdictional boundaries, online person-to-person sales must be fraught with hazards, even if not direct violence. (Admittedly, all of the parties involved cannot directly exert force on each other to block this malfeasance, but even without the online element an offline rummage sale should suffer from many of the same flaws.) Does this sound correct? [I mean both ‘correct application of your thesis’ and ‘conclusion validly matches real world experience’.]

    Lets take eBay. It is regulated because banking is involved and while fraud does occur the sales are mediated via PayPal, credit cards, etc and thus the hazards limited through the way our banking and credit systems operate (i.e. under some regulatory controls).

    Once we’re at the stage that money exchanges hands the anonymity drops away. My disagreement with Jessica that this all works without either overt or “hidden” (aka governmental) violence keeping bad actors from breaking the market.

    Why do you believe the eBay is currently an example that works without governmental controls? When you move to paying with bitcoin/cash/other payment processors the risks increase tremendously that it is a small fraction of the transactions (with the most incidence of fraud).

    Second, we have personal relationships, e.g. dating. [Your thesis is not phrased in a way to limit itself to only the economic sphere. Correspondingly, I have applied it to all markets, even those with limited or fringe financial connections.] Since Western, democratic governments exert no influence over this realm of life, by your statement we must instead see overt violence by individuals as the predominant factor. Rather than courtship, the primary means of determining romantic relationships would be people clubbing one another over potential mates or overt subjugation of said mates: there are no means to establish a romantic relationship which does not expose oneself to imminent, unavoidable violence. All long term relationships, romantic or otherwise, are primarily defined in terms of dominance rather than compassion or mutual respect. Does this sound correct?

    Excuse me what? Democratic governments very much influences this aspect of life so you CAN’T just buy/sell/steal/trade a mate. Without these controls @see ISIS controlled territories.

    You can argue that these not need be “governmental” controls but if it’s not from a stable and strong government then you better hope your clan has more guns and people than the clan that wants to take your women and shoot your men.

    Finally, religion. Again, most Western government have withdrawn from this influencing this sphere of their citizens’ lives, so your thesis would demand overt violence must predominate. All arguments over whether creationism should be taught in American schools or whether religious monuments may be erected on public grounds were settled through guns & private armies, not lawsuits. Denouncing the faith of another or changing ones own faith practice carries such a life-or-death risk that religious organizations must hire bodyguards for preachers and new converts alike. Does this sound correct?

    So the common denominator among these three examples is that free market works in the benign environment of western democracy BUT we’re going to ignore that and assume none of the benefits of western democracies need to exist and claim these are examples where free markets work in the absence of the “hidden” violence of government.

    Western governments have not withdrawn from this sphere and in fact enforce freedom of religion via “hidden violence” which I claim is a lot better than the alternative…which DOES require bodyguards. For a recent example of where western governments intervene see French laws regarding head coverings and the recent court cases on the legality of burkini bans.

    The “Vicar of Baghdad” had 35 bodyguards before the Anglican church pulled him out because he’d have been killed eventually. Just too high a visibility target to survive and do any ministry at all. The Archbishop didn’t want a martyr for the cause.

    It’s obviously not perfect even under western democracies as even here someone shot a Imam in NYC a week ago…possibly because of religion. But that sort of activity is expressly prohibited by western democracies.

  224. Time zone differences make me late to answer.

    @Dan
    “…but there simply isn’t enough wealth to justify investing in healthcare infrastructure like we do in the western world. They can’t afford it.”

    It seems you just admitted that free markets cannot supply quality health care to poor people. Anyhow, no one volunteerd empirical evidence that free unregulated markets can indeed deliver quality health care to all people. Maybe there is other, non-empirical evidence, but I have not seen it. I did see many Just So stories, but I would not bet my health on Just So stories being correct.

    @Jakub Narebski
    “Anyway, free market is proven to work if certain conditions are met, and one of them is symmetry of information.”

    That is indeed crucial. Almost by definition there is an enourmous asymmetry of information in health care. So, indeed, it seems extremely unlikely that a free market of health care would work.

  225. @nht
    > But if you were wrong about that, what else are you wrong about that you base your opinion on?

    Dude, I’ve been schooling you on this stuff for what, two weeks now? How much longer do you expect a free education? If you were fun to debate it’d be a different thing, but you aren’t. I often like debating red flag waving socialists like Winter, who is fun and I learn things from him. From you? Not so much.

    I recommend you read a book outside your echo chamber. A great starting point would be Harry Browne’s excellent “Why Government Doesn’t Work” which, if I recall correctly gives some very specific examples of how delayed drugs caused the deaths of literally millions of people. If I had my copy handy I’d check out the specifics, but I don’t, so I won’t.

    JB out.

  226. @ nht

    Oh shit, my mistake. Many apologies.

    No problem.
    @ Jessica Boxer

    What has been lost, aside from the basically useless subjunctive.

    Basically useless? If I didn’t know better, I’d agree with you… and say: “If I wouldn’t know better…”. I’m glad I can discern between the subjunctive and conditional moods and would rather retain the distinction, thank you very much. In fact, I’m so fond of distinctions in language that I’m even intrigued by how nouns in Latin vary according to grammatical case. I’d like to learn Latin… but I digress.

    You might also consider that ebonics has a verb structure that has additional capabilities…

    Well, this does suggest a remarkable degree of versatility. Thanks for bringing it to my attention.

    You want an expressive language, yet you eschew verbification. That seems a little contradictory to me.

    You’re right; that was just a gut reaction (I probably thought of the disgusting sentence, “Did you wikipedia it?”). I see nothing wrong with, for instance, “I wanna rock-and-roll all nite and party every day”. And I do like that song, though not as much as “Shout It Out Loud” or “Sure Know Something”. But I’m digressing yet again. ;-P

    If there’s such thing as winning a debate, you’ve won this one. As a reward, let me show you this cat video. I know what you’re thinking, but this one is relatively clever: the cat’s piano playing is accompanied by an orchestra. (No, the players in the orchestra are not cats – though that would be cool.) The orchestral arrangement seems to have been written ad hoc to match the cat’s piano part. The result is not earth-shattering, but nonetheless worth checking out (especially if you like cats).

  227. >nht, you are full of shit.

    That’s unnecessarily rude. He is, however, profoundly ignorant of the principle of what is seen, and what is not seen:

    “. . . As in other areas, with medical regulations, unintended consequences can be hard if not impossible to see. Legislators and voters see the deaths resulting from complications with a drug, for instance, even if there’s “only” one death for every one million people who take the drug. So, in response to a few tragic deaths, legislators may push through new regulations that make it much more difficult to bring life-saving new drugs to market. . . .”

  228. @Winter:

    Almost by definition there is an enourmous asymmetry of information in health care. So, indeed, it seems extremely unlikely that a free market of health care would work.

    The same is true of any industry where the product being sold is expert knowledge rather than some physical good: medicine, law, software engineering, plumbing, auto repair, &c…. Anyone who can accurately assay the quality of service provided, does not need the service to be provided.

    Odd then that we don’t see a push for software, plumbing, or auto repair to be provided by the government…

  229. @jessica

    You know one sign of an echo chamber is everyone repeating “facts” that simply aren’t true but everyone believes anyway.

    So you cannot provide data to support millions of deaths from the FDA…a claim which would put the FDA on the same level as Mao, Stalin and Pol Pot. What’s the saying? Extraordinary claims require extraordinary evidence?

    I prefer: “A Wise man proportions his belief to the evidence.” – David Hume

    A parting Google search for you:

    https://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=2&ved=0ahUKEwjBmMfn_uvOAhUEJCYKHVcQB84QFggnMAE&url=http%3A%2F%2Fjrbooksonline.com%2FPDF_Books%2Fwhy_govt_doesnt_work.pdf&usg=AFQjCNG1OdI7zL76hr120xNSpdq_CXWH2A&sig2=nGQU3pfOKUTmGy3KTnE36Q

    The book is copyrighted so that may be a pirate copy unless he has since put it into CC or public domain. He would right? If he did, you now have a copy anywhere you have internet.

    I’ll note that this version does not provide data to support “millions”. It has only one example (propranolol) where the delay was 5 years and may have resulted in 10K deaths per year. 50K if you accept that without any supporting evidence (other sources critical of the FDA says 30K deaths). There’s is no way to determine the veracity of the 10K death assertion or method in which it was determined.

    You can argue that the FDA does this every year so after a few decades you can get to millions but where are the other examples? If there are many drugs that are delayed that cost 10,000 lives a year where’s the list? Why do the sites that discuss the FDA being evil because of drug delays keep citing propranolol and reference the same original source or each other?

    Is that an echo I hear?

    Yeah, I can imagine it’s no fun debating someone that constantly asks for data and fact checks using the internet.

    Safe journeys until next time.

  230. @trying

    “That’s unnecessarily rude. He is, however, profoundly ignorant of the principle of what is seen, and what is not seen:

    “. . . As in other areas, with medical regulations, unintended consequences can be hard if not impossible to see. Legislators and voters see the deaths resulting from complications with a drug, for instance, even if there’s “only” one death for every one million people who take the drug. So, in response to a few tragic deaths, legislators may push through new regulations that make it much more difficult to bring life-saving new drugs to market. . . .””

    This cuts both ways in this discussion and it’s a point of discussion we had not yet reached before it ended.

    Comparisons regarding the impact to drug delays is in comparison to the rate in which drugs appear in other regulated markets (Europe, Canada) as opposed to unregulated or effectively unregulated markets.

    I will agree that deaths have occurred because the FDA is sometimes too slow to approve new drugs and the deaths are “hidden” when this occurs. Likewise the deaths that would have occurred in unregulated markets are likewise “hidden”.

    You cannot compare drug mortality rates between two regulated markets and draw conclusions on what the death rate in an unregulated market would look like. The faster regulatory agency (which oddly may be the FDA*) still eliminates the majority of injurious drugs and drug interactions.

    You need to compare the death rates between unregulated drug markets to that of the US or Europe (taking into account the other differences) to be able to show that FDA regulatory control is harmful.

    Where such limited comparisons are even vaguely possible it’s usually unfair because the unregulated market is unregulated because of rampant corruption (so you don’t REALLY know what drug you are getting) or war/anarchy. Or historical death rates from before drug regulation which has it’s own issues for comparison.

    But tell me…in what way is a private agency that does drug approvals not suffer from the same pressures? Competitors and watchdogs will highlight when they wrongly stated a drug was safe and deaths occurred and the harm when they delayed a drug to avoid a bad headline is equally hard to see.

    How many headlines of “the new drug approved by Private Drug Agency has resulted in a death of a 4 month infant in Scotland from drug complications…” before the PDA starts adding more tests and delaying approval?

    I guess you can argue the FDA isn’t allowed to pay off victims when they are wrong so the private agency can better hide approval mistakes. If the PDA is quick enough it can quash the story by paying off enough people and quietly withdraw approval. Of course, then decisions becomes a cost benefit analysis and not a health and safety analysis.

    * http://www.cancerworld.org/Articles/Issues/58/January-February-2014/Cutting-Edge/637/Approval-rating-how-do-the-EMA-and-FDA-compare.html

  231. @Alex K
    “Odd then that we don’t see a push for software, plumbing, or auto repair to be provided by the government…”

    The asymmetries are mostly less profound in these areas. A medical specialist studies until s/he is in her thirties before s/he can practice. Most customers are utterly ignorant about health care (quote the smokers who are anxious about the health risks of tap water).

    And oddly enough, when lives are at stake, like with automobiles, governmental regulations are stict. With health care it is more like fire fighters, a huge information asymmetry and often urgent needs for help.

  232. Jessica Boxer on 2016-08-20 at 13:54:03 said:
    > William O. B’Livion
    > > The problem is that Agile doesn’t have any real meaning any more.

    > Yes, William I agree with that broadly speaking. However, even if the plan is laid out to the
    > end, there is a process for adapting at least. Which is a step in the right direction.

    When one of the metrics for determining success (meaning whether your consulting company gets paid) is whether you get your work done inside the sprint it was assigned then there goes *that* idea.

    https://giphy.com/gifs/train-wreck-oVM4SGyjdVlv2

  233. The problem is that Hammers don’t have any real meaning anymore.

    When one of the metrics for determining success is how many screws are applied then it just sucks.

    I don’t see how the misapplication of Agile is a problem for the methodology itself because any tool can be incorrectly used. That Agile IMHO doesn’t scale all that well to larger projects is an issue. That some companies force contractors use agile like earned value* waterfall it is not. They’d have screwed you somehow regardless of what development methodology you choose.

    * folks insist you can use EVM on Agile. Maybe but it doesn’t feel natural.

  234. @nht:
    [When I scanned through the discussion the first time this didn’t seem to appear. Since it seems something I said may be triggering the spam filters, I’m snipping slightly aggressively.]

    My disagreement with Jessica that this all works without either overt or “hidden” (aka governmental) violence keeping bad actors from breaking the market.

    Excuse me what? Democratic governments very much influences this aspect of life so you CAN’T just buy/sell/steal/trade a mate. Without these controls @see ISIS controlled territories.

    OK, I see the miscommunication now. You were defining “hidden violence” to be all governmental violence, whereas I was using a far more restrictive definition. [Because to me, calling the sort of violence which inspired the BLM movement or the warfare which supported the HEIC “hidden” is a stretch too far — by definition, that which is hidden is supposed to go unnoticed.]

    However, then that turns the question of what makes for a “free market” — ie. one not influenced by government violence — into quite a different question. Under your definition, I cannot think of a single example of a “free market” anywhere in the world. (Your counterexample of ISIS does not work, because following it immediately by the phrase “controlled territories” implies they are a government. The people and land they organize may be contested, but that is not sufficient to violate my definition of what comprises government. )

    In fact, your explicit call-out that “Democratic governments very much influence…” is of itself rather notable. Are you implying that (most, if not all) non-democratic governments such as ISIS or North Korea don’t have laws against rape or slavery, or that only democratic governments are worth including in economic and social critique?

  235. @Aldx K
    “Under your definition, I cannot think of a single example of a “free market” anywhere in the world.”

    Indeed, when I look at history, I cannot think of a single market that was not protected by an armed force. Historically, market rights were granted by the local lord, who also protected the market, or were held in walled cities. But also more abstract markets are placed inside a protected (abstract) space.

    There is a good case to be made that any place where profits are made and value amassed attracts violence. Disputes have to be settled and ownership enforced. The historical solution has been a final arbiter with power to enforce. An arbiter that is paid by taxes. That what is called “government” in modern terms.

    If someone knows counter examples, please share them.

  236. In fact, your explicit call-out that “Democratic governments very much influence…” is of itself rather notable. Are you implying that (most, if not all) non-democratic governments such as ISIS or North Korea don’t have laws against rape or slavery, or that only democratic governments are worth including in economic and social critique?

    I call out democratic governments because you called out western democratic governments:

    Since Western, democratic governments exert no influence over this realm of life

    No other reason so it is not notable.

    My definition of “hidden violence” is based on what I thought Jessica to meant by “hidden violence”…which I took as regulations and police to enforce such regulations as opposed to the obvious violence inherence in HEIC oppression of native peoples under their domain.

    At the end of the day, any law/regulation/etc must be enforced by the potential threat of violence or many folks will simply ignore them. Some folks ignore them anyway but fewer…and we try to throw those folks in jail when we can catch them.

    However, then that turns the question of what makes for a “free market” — ie. one not influenced by government violence — into quite a different question. Under your definition, I cannot think of a single example of a “free market” anywhere in the world.

    You said it, not me, but I agree. In what context do you feel free-market anarchy works without governmental violence?

  237. > USPS is a floundering blob now.

    It’s a little more complex than that. The US Postal Service makes a ton of money. Congress takes it all, dumps it into the General Fund, and the USPS has to beg for appropriations to cover its operating expenses. A varying percentage of your postage is another hidden tax, like ITAR registration fees.

    They’re also saddled with the American Postal Workers Union, with the usual union featherbedding and sandbagging. Every postal worker I ever talked with loathed the APWU, but the APWU’s hooks are set tight since they’re Federally recognized.

  238. >Sorry, but most people in this world are not protected by the likes of the FDA. Literally Billions of people live in countries like China, India, Nigeria, or Russia, where almost anything goes.

    “Protected” is an interesting way to put it. As in “Protection Racket”.

    It is because of the FDA systematically denying alternatives to EpiPen that they have their effective monopoly. Note that epinephrine’s patent is expired, so it’s only the patent on the delivery mechanism that is in play. And the sole alternative delivery system to be approved can’t be provided as a generic equivalent to EpiPen because delivery systems are not subject to the generic-substitution rules for drugs. Only if one can persuade the MD to write the Rx in a very specific way that would allow the alternative to be used is it legal to dispense it.

  239. “Government violence” cleverly lumps together punishing those who transgress against the person or property of others with those guilty only of trying to compete against a favored player. To paraphrase Jefferson, that government governs best which employs just enough force against those transgressors to act as disincentive such that the extent to which force governs human relationships is minimized. Minarchism is tough, because it’s easy to overshoot that sweet spot in either direction.

    The outright anarchist thinks “the government which governs least” is the one that doesn’t govern at all, but a government that fails utterly like Iraq/Syria leads inexorably to an ISIS to fill the vacuum. The statist thinks that if a little bit of government force is better than none, it just stands to reason that more government force is even better. The totalitarian simply turns that up to eleven, (and perhaps slaps grätüïtöüs ümläüts* all over while he’s at it).


    *Not to be confused with any metal band actually named Grätüïtöüs Ümläüt)

  240. It is because of the FDA systematically denying alternatives to EpiPen that they have their effective monopoly. Note that epinephrine’s patent is expired, so it’s only the patent on the delivery mechanism that is in play. And the sole alternative delivery system to be approved can’t be provided as a generic equivalent to EpiPen because delivery systems are not subject to the generic-substitution rules for drugs.

    The FDA blocked filling EpiPen prescriptions with Adrenaclick, Twinject or the generic version of Adrenaclick because attempting to use the other auto injectors the same way you use EpiPen may result in no drug being administered. So they aren’t interchangeable.

    They want the doctor and user to be aware that the mechanisms are different.

    Teva and Sanofi injectors have problems…enough that Sanofi had to recall all their Auvi-Q injectors and pull them off the market. There’s a reason the FDA wants to be sure alternative injectors reliably inject the correct dose under stressful situations.

    Adrenaclick’s problem isn’t the FDA but that insurers haven’t wanted to cover it.

    The other problem isn’t regulatory at all…drug companies are paying off generic drug companies to delay manufacturing of generics. Something that would also be possible in a deregulated “free” market.

  241. @nht:
    Point conceded re: democracy callout. I had mis-read this as having been highlighted previously in your discussion.

    My definition of “hidden violence” is based on what I thought Jessica to meant by “hidden violence”…which I took as regulations and police to enforce such regulations as opposed to the obvious violence inherence in HEIC oppression…

    In what context do you feel free-market anarchy works without governmental violence?

    … and @Winter:

    Disputes have to be settled and ownership enforced. The historical solution has been a final arbiter with power to enforce. An arbiter that is paid by taxes. That what is called “government” in modern terms.

    I feel that both of you are approaching this topic from a similar misapprehension. If you define an “free market anarchy” as containing literally zero government and follow Winter’s definition of government as dispute arbiter, then you have defined a square circle that cannot exist. Every well-defined marked will always have at least one arbiter. Yes, historically this process has required creating a singular arbiter endowed with the power to use violence.

    However, that does not suffice to show that (a.) this centralization must inevitably be the case [an argument from authority, namely history] or (b.) that government structures need monopoly-like plenipotent control over their market [this implies there would one sole website governing online person-to-person sales, not multiple popular examples]. This line of argument cannot even suffice to show violence is required— nht’s earlier claim that “because banking is regulated, and eBay uses banks, therefore eBay is regulated” would imply that even the black market could be regulated, an outright contradiction.

    If centralization is a mere historical artifact, blocking more efficient structures through application of violence moves markets away from their theoretical minimum levels, and the governing system as it exists today (plenipotent, centralized, geographically-based monopolies) therefore produces additional violence above the minimum level required. While I doubt that the theoretical lower limit on violence is actually zero, we cannot state with any certainly what that limit is, or even that markets today are close to or approaching that limit.

  242. This line of argument cannot even suffice to show violence is required— nht’s earlier claim that “because banking is regulated, and eBay uses banks, therefore eBay is regulated” would imply that even the black market could be regulated, an outright contradiction.

    Black markets tend to be cash or cash equivalent markets that require quite a bit of laundering. They are not regulated as much as constrained to the fringes.

    eBay is different in that it leverages the banking infrastructure for transactions and that provides significantly more protection.

    If you define an “free market anarchy” as containing literally zero government and follow Winter’s definition of government as dispute arbiter, then you have defined a square circle that cannot exist. Every well-defined marked will always have at least one arbiter. Yes, historically this process has required creating a singular arbiter endowed with the power to use violence.

    And the reason this has not changed is because human nature changes more slowly than the rate of our technological progress.

    While I doubt that the theoretical lower limit on violence is actually zero, we cannot state with any certainly what that limit is, or even that markets today are close to or approaching that limit.

    It can be stated with reasonable certainty that the amount of violence in the governance of western democracies is low in comparison to historical alternatives.

  243. @Alex K
    “(a.) this centralization must inevitably be the case [an argument from authority, namely history] or”

    No, this solution is the only one observed, and for which there is empirical evidence. That is not an argument from authority. It simply means all other solutions are hypothetical without any empirical evidence they could actually work.

    Show us evidence it could actually work.

  244. @Alex K
    “(b.) that government structures need monopoly-like plenipotent control over their market [this implies there would one sole website governing online person-to-person sales, not multiple popular examples]. ”

    Who claimed that? Governments developed for many reasons. Protection of ownership, trade, and markets were just one aspect.

    @Alex K
    “This line of argument cannot even suffice to show violence is required”

    I know pacifist have a line of reasoning in that direction. But, again, history teaches us that wealth attracts people who will resort to extreme measures of violence to take their unrightful part. Disputes also tend to escalate occassionally into violent feuds. Every market must deal with such people. I do not know of any examples that did not include some level of violence and threats of violence.

  245. @Winter:

    That is not an argument from authority. It simply means all other solutions are hypothetical without any empirical evidence they could actually work. Show us evidence it could actually work.

    First, I agree that pointing to an actual, real world system is evidence it could actually work. However, your statement that “all other solutions are hypothetical” seems to imply this is the only means to demonstrate viability. Did I misunderstand, or can you give me an example of evidence a system is viable when that system does not yet exist in the real world?

  246. ” Every well-defined market will always have at least one arbiter. Yes, historically this process has required creating a singular arbiter endowed with the power to use violence.”

    No. Just a short while ago, esr blogged here on “silent trade”. That worked as a way of avoiding violence. ( It worked for centuries, as a matter of fact.)

  247. @Alex K
    “However, your statement that “all other solutions are hypothetical” seems to imply this is the only means to demonstrate viability.”

    No, I did not try to say that. I said there is no empirical evidence that these other solutions work. And I invited readers to share any other, non-empirical, evidence they know of. I have not seen such evidence.

    @LS
    Indeed, there are more forms of trade than “free markets”. The silent trade as blogged is a trade negotiation process between two parties. A market is where a multitude of sellers and buyers meet to trade. I know of no examples where this was done the “silent” way.

  248. @esr
    A question that’s – hopefully – relevant to the original topic: do you approve of the proposal to apply Taleb’s ideas to software development?

    1. >do you approve of the proposal to apply Taleb’s ideas to software development?

      I think the general idea of mimicking the adaptive strategies of antifragile systems is sound. I don’t see that the author has succeeded in extracting much actionable advice from it, not after you subtract the things we already know about process transparency and rapid iteration from open-source development.

      I’d like to see more made of this idea, if possible. Right now I don’t see enough novelty to justify the weight of rhetoric being deployed. Maybe that will change.

  249. @nht
    Probably Elsevier. They have ruffled many feathers. Scientific publishing is an oligopoly and Rand Elsevier is one of the biggest. They do what oligopolists are best at, making lots of money at very, very high margins. Universities balck at having to pay for the research, then to get it published (page charges), their employees do the peer review and the editor jobs unpaid, then they have to pay to read the papers they published, and again if they use their own writings in class.

    And Elsevier then locks up the research in ScienceDirect where the universities have to pay yet again to search the literature.

    I think people should just switch to open access publishing. Publishing costs should be considered part of doing research and paid out of the research funding.

  250. @winter
    You’re dead right! That’s my real complaint. The ideas are definitely good ones, but here’s the thing: it’s not worth optimizing corner cases unless there’s an obvious optimization. Too easy to introduce bugs.

  251. @EMF
    “it’s not worth optimizing corner cases unless there’s an obvious optimization. ”

    Not sure what you intend to say here. But Open Access publishing is partly based on the realization that the universities and grant agencies pay all the publishing cost anyway as they are both the primary producer of scientific articles as the consumers who pay for using the papers.

    Essentially, with electronic publishing, there is little that the comercial publishers contribute as middle men. But they rake in a lot of profits for it. Profits that the researcher would rather spend themselves. So why not cut out the middle men?

    Hence, the economic case for Open Access publishing.

  252. Elsevier are literally the scum of the earth. They are the mafia whose protection racket you have to pay in order to science. Not only do they overcharge for access to real science, they’ll gladly take cash to pass off fake science as the real thing. They were the ones who published the vanity journal Merck commissioned that contained marketing material for their products (such as Vioxx) disguised as scientific research.

    Open access is the way to go for scientists and the beneficiaries of science (us). The problem with your economic case, Winter, is that it is not Pareto-efficient: it makes scientists, universities, and ordinary citizens better off while making Big Publishing, Big Pharma, and the like significantly worse off. That’s why it won’t be adopted. The biggest chunk of the benefit from the status quo will simply disappear.

    This is why I giggle whenever I hear libertards speak of “efficient markets”.

  253. @Jeff Read
    “The problem with your economic case, Winter, is that it is not Pareto-efficient: it makes scientists, universities, and ordinary citizens better off while making Big Publishing, Big Pharma, and the like significantly worse off. ”

    Thaf is why the grant agencies in the EU are requiring that recipients of grant money make any publucations open access (currently with a grace period). Universities are following. Who pays the piper calls the song.

  254. Ab got it. Only published to IEEE conferences and a lot of the other stuff is under ONR or DARPA so not much publishing there.

    Looks like the White House pushed for open access on research papers last year and agencies like the NSF implementing new policy in that regard.

    All in all a good thing.

    What is funny (and I kind of killed the discussion thread on License-Discuss with the observation) is the new federal OSS policy is based on a non-OSI approved license: CC0. A license approval that should have been a slam dunk and turned into a circus until CC withdrew it as more drama than necessary since the FSF was already recommending it.

  255. @nht
    CC0 has legal problems. In some jurisdictions, you cannot disown all copyrights. So the license would not be binding. It is even suggested that it might not hold in the USA.

    But I could not say, you would have to ask a specialist.

  256. CC0 has a backup permissive license. The sticking point was CC0 originally for open data and that community wanted to freely publish findings and data without giving up potential patent rights.

  257. @esr

    I think the general idea … is sound. […] I’d like to see more made of this idea, if possible.

    Thanks for your thoughts.
    A lighthearted question about software development (not methodology): do you ever include Easter eggs in the software you write?

    1. >do you ever include Easter eggs in the software you write?

      I don’t think I’ve ever actually done that. Well, not unless you count snarky remarks and jokes in source-code comments.

  258. >It is even suggested that it might not hold in the USA.

    I find that very hard to believe. The entire point of CC0 was that just declaring a work to be in the public domain would work in the US, but might not elsewhere.

  259. Eric, would clever and hard-to-grok references in Wesnoth count? How about jokes that are only shown upon meeting ceetain, specific, and somewhat obscure game conditions?

  260. “Extraordinary claims require extraordinary evidence?” – and there’s the problem, and there’s why people with different beliefs can never agree with each other. The anti-FDA people are enthusiastic to believe that the FDA is a bunch of murderers, and so in their view it is not an extraordinary claim, and that they are not is one (in addition to having to prove a negative).

  261. There is “differing beliefs” and there is claiming the FDA is willfully responsible for the same number of (if not greater) deaths as Pol Pot.

    Trying to claim that isn’t an “extraordinary claim” because the believers are “enthusiastic” is nonsense. Folks are “enthusiastic” about aliens and Area 51 too.

  262. @Jon Brase
    “The entire point of CC0 was that just declaring a work to be in the public domain would work in the US, but might not elsewhere.”

    I think you are right. But, as I understood, whatever you throw at a group of lawyers, they will find a way to say it is legal and a way to say it is illegal. I vaguely remembered hearing someone claiming there were problems in the USA, but this might have been FUD.

  263. “Extraordinary claims require extraordinary evidence” is an unhelpful phrase that I advise you to phase out. It appears to operate by bait-and-switch on the meaning of “extraordinary”. Consider the following interpretations of the latter part:

    1) claims far outside my paradigm of belief require evidence correspondingly far outside my paradigm of knowledge, or, “A little bird told me/I had a vision” is trivial to assert but convinces practically nobody
    2) claims far outside my paradigm of belief require evidence correspondingly far inside my paradigm of knowledge, or “Show me that Marx is wrong using only recent citations from my sociology department” is a ludicrous standard I have no interest in trying to meet
    3) claims far outside my paradigm of belief require evidence that is extraordinary as measured on the platinum-iridum bar of objective extraordinariness in the metric system vault in France, which doesn’t exist. And probably can’t exist by hypothesis: it’s precisely when the parties to an argument don’t agree on a standard of “extraordinary” that they’re likely to start throwing around “extraordinary claims” rhetoric in the first place!

  264. How about castigating the person that believes that the FDA has generated the same pile of bones that a nasty communist dictator has without showing said pile of bones?

    You guys really like avoiding the main point. a) she presented a supporting case that was debunked in a 30 second Google search b) when pressed for data she ran away claiming that she “schooled me”.

    There is no solid evidence that the FDA has killed MILLIONS. There is a debatable assertion that they have killed thousands because one drug was delayed based on arguments of authority (a former yada yada says so) but not how that number was derived or the data shown.

    A small number tragic delays can be mistakes or “differing opinion” on the timidity of the FDA in making mistakes.

    Killing millions requires systematic and constant delays over decades that it should be easy to generate a sizable list of drugs (given that the most oft quoted case resulted in a claim of tens of thousands there will need to be many such cases) approved overseas and differing death rates that tallies ups to “millions” of victims as a result of those delays. Drug approvals and when they occurred is a matter of public record.

    That list of drugs and a methodology for determine the delta in deaths vs Europe is not fucking “extraordinary” but what anyone that thinks they engage in “critical thinking” to demand in order to believe that thousands of employees never fucking noticed they were killing people on a fucking massive scale.

    Never mind that reading about the FDA via a vis Europe and Canada will result in articles from the other side of the pond decrying how THEIR agency is slow in comparison to the “speed” of approval by the FDA. Which leads me to believe that it’s a wash for the speed of regulatory approval.

    Show me that list and a reasonable approach in computing the differing death rates and I’ll change my opinion because then there will be data to support that assertion as opposed to handwaving and debunked cases.

  265. @winter when the FSF says it’s legally okay then it probably is for the US.

    They make debatable assertions on what is or isn’t a derivative work but otherwise tend toward being conservative on what is open source and what isn’t.

    I dont recall the status of CC0 with dfsg.

  266. >Killing millions requires systematic and constant delays over decades that it should be easy to generate a sizable list of drugs

    It is not possible to list the drugs that were never even created because the cost of adhering to the various regulatory regimes was too high a hurdle. How many times did a researcher get the word that there just aren’t enough people with ${DISEASE} to generate sufficient revenue to pay back all of those costs?

    This is not a hypothetical with me: The Bride of Monster has Myasthenia Gravis, which damn near killed her when she was nine years old, and is likely to be what eventually does kill her. There isn’t much happening with new drugs to treat it because it only affects 50~200 out of every million people. So the upper bound is ~66K patients in the US, and a more reasonable central estimate would be more like half that. That also limits the number of people who can be killed by the lack of drugs to the same number, but when you multiply tens of thousands of people per disease by thousands of “Orphan” diseases, you get to “millions” very quickly.

    You can’t list the things that never happened. “Seen v. unseen”, pace Bastiat.

  267. How many times did a researcher get the word that there just aren’t enough people with ${DISEASE} to generate sufficient revenue to pay back all of those costs?

    This is not a hypothetical with me: The Bride of Monster has Myasthenia Gravis, which damn near killed her when she was nine years old, and is likely to be what eventually does kill her. There isn’t much happening with new drugs to treat it because it only affects 50~200 out of every million people. So the upper bound is ~66K patients in the US, and a more reasonable central estimate would be more like half that.

    Well, my opinion is that we should try to reduce the cost of meeting regulatory requirements and provide federal grants to help bring needed drugs to market that may not make economic sense to do so because the number of patients are too low. Regulatory costs are not the only costs for development and clinical trials are still needed.

    You would have to ask FooQuuxman if this statement still applies:

    “3. This is as it should be. That it is unprofitable (and no one is proposing a market failure here) is in itself evidence that it should not be happening. Period.”

    I disagree with that sentiment whether it is mail service or drugs.

    With respect to counting useful drugs that don’t make it to market it is equally hard to count the number of harmful drugs avoided except by comparing to the number of historical deaths due to the selling of harmful snake oil medicines. Given that medicine (and reporting) has changed since that period that’s a hard estimate to make.

    The positive aspect of the evils of intellectual property rent seeking is that if my kids were suffering from something were a potential cure existed in Europe I could afford to bankrupt myself once in getting them that treatment. Persons of greater wealth could just afford to do it. Persons of lessor wealth probably could not afford to do it even once.

    But that’s with the assurance that it passed regulatory checks in Europe…not no regulatory checks at all for safety and efficacy and trusting the free market that my kid isn’t getting the modern day equivalent of radium water as a cure. I don’t think you would want that for The Bride of Monster either.

    I’ve had cancer…its not entirely hypothetical for me either. If folks can show me hard data that the 10 year survival rate as a whole (as in across all illnesses) is higher without the FDA than with the FDA then I’m all for helping change the rules by voting for whomever is willing to try to do so.

  268. > . If folks can show me hard data that the 10 year survival rate as a whole (as in across all illnesses) is higher without the FDA than with the FDA then I’m all for helping change the rules by voting for whomever is willing to try to do so.

    If it’s higher with the FDA than without, then there’s no reason for FDA compliance to be enforced by Men With Badges And Guns. People will want effective and safe medicine that’s been vetted by independent labs, and will trust in the FDA Seal of Approval just as they do Good Housekeeping or Underwriter’s Laboratories. When someone offers an unapproved treatment, they’ll have to be in a situation where no approved treatment is available or people won’t touch it with a 10-foot pole.

    The Fundamental Contradiction of the Democratic Nanny State is the supposition that people are incompetent to make such decisions for themselves, but somehow when they step into a voting booth become competent to make the decisions for us all, either directly via referendum, or indirectly by electing Really Smart People Who Know Better. You can’t decide whether to light up a joint to fight the chemo-induced nausea, but you can vote for legislators to decide for everyone. You can’t weigh the risks of vaping vs. smoking cigarettes, but you can elect Your Betters to make rules that outlaw both.

    If you’re competent to make these decisions for us all, you’re certainly competent to make them for yourself. As Jonathan Edward put it: “He can’t even run his own life; be damned if he’ll run mine, Sunshine.” (Anyone who can’t be trusted to run his own life can’t be trusted with a vote over mine.)

  269. So you are saying you don’t have the data either…

    AND you’re going to ignore that “unprofitable drugs shouldn’t happen” in your “free market” and complain about “nanny state” instead. The drugs to treat forgotten illnesses will not exist except in a so called nanny state because it will almost always be more cost effective to let those people die.

    The answer to the supposed “Fundamental Contradiction” is simple. There are bad actors that require “men with badges and guns” to constrain and not having this has historically lead to bad outcomes.

    Granted sometimes the “men with badges and guns” have been the problem but having western civilization is better than not having western civilization and that has been based on rule of law and centralized government.

  270. Re: healthcare and similar things, especially to Winter. Time machine back to 1960. Healthcare both in the US and Europe was free market based, almost unregulated, hence, cheap and simple. Europeans practically nationalized most of it while it was cheap and simple. Americans proceeded to regulate the ever living heck out of it while keeping it nominally private. Costs skyrocketed. As a result now it is too expensive and convoluted to nationalize.

    Lesson: if you ever want to nationalize shit, do it while it is cheap and simple. Don’t proceed regulating it until it becomes expensive and convoluted. I bet that it would simply collapse if nationalized, because only great minds can make that mess stay up, and great minds work for corporations, not the government.

    And to pull a Bernie and basically make the taxpayer pay the profits of private corporations would be literal worst way to solve it. I don’t even understand who supported Sanders. Did he really say the taxpayers should pay education provided by for-profit universities and his supporters did not go “wat?”

    And that is a bit like a checkmate.

    Interestingly, the part of European healthcare that did not get nationalized stayed mostly unregulated and that demonstrates how cheap modern free market healthcare can be. I can show my knee to a professor who teaches knee stuff at the university for €140 an hour. That insane cheap, because that part stayed a truly competitive free market, unregulated, basically no costs beyond his time. This is a pretty decent thing, because we have the option open to smoothly re-privatize it, if there is a political will. If the law says my knee doc must put in 50 hours a month serving public patients (state insurance) and he can sell the rest on the free market for cash like he did to me, which is very roughly how European healthcare tends to work, we could reduce that quota in the law, it is a system that can be smoothly changed without big bumps.

    But I don’t know how overregulated messes can be smoothly changed.

    Likely the smart thing would be in America is to define a new class of people who are in practice doctors but are not called so. Call them health consultants. As a new class it is unregulated. Keep it so. Observe the results. Basically starve the healthcare mess gradually from patients, channel them into something new.

  271. > There are bad actors that require “men with badges and guns” to constrain and not having this has historically lead to bad outcomes.

    Yes, having Men With Badges And Guns to constrain murderers, robbers, rapists, perpetrators of fraud, etc. is a good thing. Having them to pre-emptively constrain people from allegedly self-destructive behavior is not. If you can’t grasp the difference between these two things, there really is no hope for a reasonable discussion.

    If we accept that government may routinely restrain people from self-destructive behavior, then what people do in their bedrooms (potentially transmitting STDs) suddenly becomes everyone’s business, empowering the worst of the Puritans who would ban anything that they consider sinful. What should be personal decisions about what and how much to eat become public policy matters, so shut up and eat your Michelle Obama-approved tasteless gruel, and like it, peon.

  272. To whom it may concern:

    At the risk of further contributing to the chaos in this thread, and in shameful knowledge that it’s me who caused such chaos in the first place, I want to go back to S. I. Hayakawa.

    I just read a paper of his, and it would appear that he shared Jessica’s position that “The purpose of language is communication[,] not grammatical correctness.” Thus, I may have unintentionally misrepresented Hayakawa’s thought.

    In my defense, I was merely echoing his sentiment that “everyone needs to have a habitually critical attitude towards language”, regardless of what might actually constitute such attitude. Besides, I presume he would have shared my concern about the confusion between “your” and “you’re”, for the distinction it blurs is not merely grammatical but semantical: “you’re” is the contraction of “you are”, a conjugation of the verb “to be”; by contrast, “your” has nothing to do with said verb.

    At any rate, I recommend the paper. It’s interesting (in my opinion), short, and not hosted at ScienceDirect. ;-)

  273. How about this aphorism: programmers are often toolmakers, and bad tradesmen blame their tools.

    It is not about bad programmers. Rather it is the idea that the bad accountant will blame the accounting software, which can expose the programmer to surprising accusations, and having to defend and argue (relevant) buglessness. This is something to prepare for or avoid.

  274. Jorge

    It never ceases to amaze me how people who know they must be precise in writing computer code can have a sloppy attitude toward writing in human languages.

    I don’t know which is worse:
    1) Being so stupid as to genuinely not know the differences between “your”, “you’re”, and “yore”; or
    2) Knowing and being so careless as to use the wrong one.
    I lean toward thinking 2 is actually worse, because the first sort of person has the excuse that they don’t know any better (and, so long as they have a healthy attitude about it, can learn. Those who don’t know nor care to learn combine the worst of both 1 and 2).

    For some reason, caring about these things is considered “pedantic” and a lot of other pejoratives, but in my mind sloppy writing that mixes up these homophones indicates a lack of respect for the readers, forcing them to waste brain cycles trying to suss out what the hell the writer really meant, instead of just saying it correctly in the first place.

    But the really important thing to recognize is that communication isn’t even the primary function of language (although it may be considered its primary purpose). Without ever communicating anything to anyone, we use language as a tool of cognition far more often than we use it to communicate, and even then we must first formulate the idea before we may attempt to communicate it.

    Any conceptual abstraction is associated with some symbol we use to organize our knowledge. The choice of those symbols isn’t particularly important (Spanish has different symbols from the ones English uses, but they both work) so long as they can be integrated into a self-consistent system of some kind. Sloppy language doesn’t just impede communication; it leads to sloppy thinking, which in turn produces sub-optimal decision-making.

  275. @nht:

    There are bad actors that require “men with badges and guns” to constrain and not having this has historically lead to bad outcomes.
    Granted sometimes the “men with badges and guns” have been the problem but having western civilization is better than not…

    I read these two lines like a syllogism [technically, two]:

    1) Premise: the only means to constrain bad actors in society is external use of force, aka. “men with badges and guns” (MWB&G).
    2) Premise: sometimes, the MWB&G will become bad actors.
    3) Conclusion: The only way a particular group of MWB&G can be constrained from becoming bad actors would be to create another group of MWB&G with the authority to constrain the original group.
    4) Premise: not constraining all possible bad actors will eventually lead to a “bad outcome” for society in which the order and/or prosperity of Western civilization breaks down.
    5) Conclusion: This second group of MWB&G will also need a constraining force: either by defining new groups of MWB&G ad infinitum, or eventually assigning the role of constraining force to a previously defined group of MWB&G.

    Now, your post is not explicit regarding what you would consider to be “historically bad outcomes”, but the conclusion reach in #5 above (society must have two or more groups, each armed with deadly force and legal authority over the others) sounds to me an awful lot like the sort of civil war or failed state which you seem to imply would be a bad outcome.

    So, either I am completely misinterpreting (one or more portions of) the argument you were trying to make and one of the three leading premises is thereby flawed, or your argument does not actually show that men with badges and guns will reduce the likelihood of a “historically bad outcome”.

  276. Yes, having Men With Badges And Guns to constrain murderers, robbers, rapists, perpetrators of fraud, etc. is a good thing. Having them to pre-emptively constrain people from allegedly self-destructive behavior is not. If you can’t grasp the difference between these two things, there really is no hope for a reasonable discussion.

    Constraining “perpetrators of fraud” is what the FDA does. The FDA got more regulatory power after “radium water” in terms of preventing sales of “cures” that not only didn’t work but was deadly to their users.

    Perhaps, it’s not me that can’t grasp the difference.

  277. TheDividualist
    “Time machine back to 1960. Healthcare both in the US and Europe was free market based, almost unregulated, hence, cheap and simple. ”

    Maybe, I’m too young to have experienced it. I do know the incentives are to over-treat with sub-standard or non-standard treatments.

    Case in point, routine tonsillectomies. That generation had their tonsils removed routinely. Parents didn’t know that it was not helpful so doctors would remove the tonsil in every child.

    Second case, overprescription of antibiotics. Patients ask for antibiotics for each and every ailment, whether they would work or not. Result, rising resistance. Dutch authorities and doctors have restricted use of antibiotics. Resistance to antibiotics is very rare in the Netherlands.

    Worst scandals ever: Thalidomide and DES scandals. Pharmaceutical companies subverted regulatory authorities and caused huge damage.

    The whole regulatory increase in medical treatments was caused by scandals like the Thalidomide case.

    @TheDividualist
    “If the law says my knee doc must put in 50 hours a month serving public patients (state insurance) and he can sell the rest on the free market for cash like he did to me, which is very roughly how European healthcare tends to work, we could reduce that quota in the law, it is a system that can be smoothly changed without big bumps. ”

    That is not how it works in the Netherlands.

    There seems to be a Rusdian proverb: Stupid people learn from their mistakes, smart people learn from other people’s mistakes. The USA have botched up public health big time, while almost everyone else have build better and cheaper systems. Just get over it and learn from other peoples successes and mistakes.

    Instead, you are blinded by ideology and want to reinvent a square wheel in healthcare.

  278. So, either I am completely misinterpreting (one or more portions of) the argument you were trying to make and one of the three leading premises is thereby flawed, or your argument does not actually show that men with badges and guns will reduce the likelihood of a “historically bad outcome”.

    Yes, you are willfully misinterpreting the statement. This isn’t a logic problem but a history problem. No society survives against predators without sheep dogs. When the sheep dogs become predators that society is in big trouble. Show me a historical counter example where armed sheep dogs are not required to protect from external threats of violence (i.e. conquest, raids, etc) or internal threats of violence (murder, rape, etc). The fact that predatory sheep dogs has been the cause of the fall of some societies does not mean that sheep dogs are not a necessary component of a successful civilization.

    In some scenarios where external threat of invasion or raiding was nonexistent (i.e. no large packs of predators) then that society may have gotten away without military sheep dogs. Usually only until western colonialists showed up with guns. But even fairly remote islands often had to contend with raiding from other tribes on the same island.

  279. > Constraining “perpetrators of fraud” is what the FDA does.

    No, it is not. If it confined itself to that, I’d have no objection to it at all. Punishing force and fraud are the only legitimate reasons for a government to exist.

    But it goes far further than just stopping fraud. Even if you tell a patient “This treatment has not been approved by the FDA, which considers it to not only be helpful, but dangerous”, then go on to provide the FDA’s findings verbatim, if the patient says “I understand all of this, but I want your radium water anyway” it is illegal to provide it to him. There is no fraud whatsoever in this scenario. The FDA doesn’t just want the power to prevent lying to people; it insists that it is the sole judge of what costs people may risk incurring in exchange for a potential benefit. You may think it’s OK to risk potential cancer decades from now to cure a disease threatening to kill you tomorrow, but they say it’s just too dangerous, so no, you can’t do it.

    And that’s one of the biggest flaws in The Fundamental Contradiction: There is no clear, objective standard by which those costs and benefits can be evaluated. Each of us places different values on all of the criteria that go into decisions like this. That inequality is what allows for economic transactions that benefit all parties. Each party values what he’s getting more highly than what he’s giving.

    The Statist impulse justifies itself from the ASS|U|ME-tion that a particular valuation scheme is privileged above all others. That inexorably leads to viewing all transactions as zero-sum, and that one party MVST be taking advantage of the other unless the One True Fair Trade Price has been paid, in which case at least no one loses.

    So the do-gooders shut down “sweatshops” because the working conditions are horrible by their standards and the fact that the alternatives they thereby force upon the former workers are worse by the workers’ standards is irrelevant. After all, those people (who actually live there and know the neighborhood) can’t possibly know what’s good for them like we* do. That this constitutes imposing our elites’ cultural norms upon those Persons of Color Whose Pure Cultures Are Untainted By The Original Sin of Western Civilization somehow escapes them, despite normally being the biggest fans of Multiculturalism around.

    * We learned all about that particular third-world country that one semester in college. Or was it that other country with a similar-sounding name on the opposite side of the continent? I forget. But we’re still more cosmopolitan than those Vo-Tech-graduate Flyover morons who don’t even speak a language other than English (and poorly at that), so we’re pretty sure we know what’s best for everyone everywhere at all times. Including the people in whatever country that sweatshop is in, you bet.

  280. > When the sheep dogs become predators that society is in big trouble.

    When the sheep dogs say “bugger this for a game of soldiers, I’m pissing off” society is in big trouble.

    Which is happening now.

    What is going to *start* happening in places like Chicago and New York (probably not so much in STL) is that groups like the Nation of Islam or the New Black Panthers will start “patrolling” areas that “The White Man’s Police have abandoned” (or rather driven out by protests funded by billionaire progressives and organized by old-school communist agitators).

    Patrolled by people who have no notion of “Rule of Law” or any sort of limitation on the force they meet out other than the physical limitations of their bodies and what they can carry.

    It’s almost like someone *wants* it that way. But that would be tinfoil hat territory, right?

  281. The Monster on 2016-09-08 at 12:23:57 said:
    > I don’t know which is worse:
    > 1) Being so stupid as to genuinely not know the differences between “your”, “you’re”, and “yore”; or
    > 2) Knowing and being so careless as to use the wrong one.

    > I lean toward thinking 2 is actually worse, because the first sort of person has the
    > excuse that they don’t know any better (and, so long as they have a healthy attitude about it,

    I think the channel must be taken into account. I’m a lot less, um, pedantic on a SMS device where autocorrect can kick in, it’s hard to type etc. than I am anywhere where there is a keyboard.

    As far as I’m concerned if the typist has English as their primary language, or has passed the eighth grade the only reasonable excuse is “I was drunk/on drugs/at the end of a 36 hour shift”

    Number one means you flat out failed grade school English. So yeah, if you’re that dumb and have made NO effort to get better, piss on you.

    > For some reason, caring about these things is considered “pedantic” and a lot of other
    > pejoratives,

    Pedantic is a pejorative?

    Oh.

    Uh…

  282. @monster:

    FDA, however, has a policy explaining that it typically does not object to personal imports of drugs that FDA has not approved under certain circumstances, including the following situation:

    * The drug is for use for a serious condition for which effective treatment is not available in the United States;
    * There is no commercialization or promotion of the drug to U.S. residents;
    * The drug is considered not to represent an unreasonable risk;
    * The individual importing the drug verifies in writing that it is for his or her own use, and provides contact information for the doctor providing treatment or shows the product is for the continuation of treatment begun in a foreign country; and
    * Generally, not more than a 3-month supply of the drug is imported.

    Radium water would likely not pass this as it is an unreasonable risk to others since it is actually radioactive.

    It’s the “Even if you tell a patient” part that you object to that isn’t allowed. As an individual you can find a way to import the drug you want (as long as it’s no illegal for other reasons).

    As a medical provider, pushing unapproved drugs for sale as a medical cure, no. There is fraud involved.

    You CAN sell it as a herbal supplement and many quack cures are and they follow all the rules and end up on the market. You are free to do your own due diligence and side step the evils of the FDA and take whatever quackery floats your boat.

    http://www.medicaldaily.com/herbal-supplements-consumer-reports-safety-15-ingredients-dangerous-392920

    http://www.consumerreports.org/vitamins-supplements/15-supplement-ingredients-to-always-avoid/

  283. I’ve been busy on a big project, but it was bugging me that I didn’t document my claim about millions of deaths caused by FDA delays. So I thought I’d drop back in on it.

    There is an excellent paper on this written in 1985. Of course things are much worse today than they were in 1985, and the FDA has spilled a lot of blood since then, but it is a good starting point. It documents a fairly wide range, extremely conservative estimate of 20k-120k lives lost due to the delay in submitted drugs per decade, so, taking an average number 380k since 1962. However, this specifically excludes lives lost due to the considerable delays in various cancer drugs (cancer being the second leading cause of death) because the author of the paper did not feel able to pick between the massive reductions in cancer mortality due to new treatment protocols and specifically chemotherapeutic agents themselves. However, the number is undoubtedly extremely large, perhaps 10k per year would be a reasonable number, or 500k since 1962,

    Already we have nearly hit the one million mark, but of course that only accounts for delayed drugs. What it does not count is drugs that were never developed in the first place because it is too damned expensive.

    According to Pfizer it costs about $1.2billion dollars to create a new drug. Of course by no means is all of that the cost of approval, but a significant amount is. This means that some drugs just don’t get developed at all, and the lives that would have been saved don’t get accounted for on the FDA’s wall of shame. How many? Who the hell knows, but the difference here is that, unlike delayed drugs which are approximately linear (lives lost x number of years delayed), this death toll grows exponentially, because these drugs don’t, eventually, get approved.

    This is especially compounded because American medical consumers essentially bear most of these costs, and so the fact these drugs aren’t developed due to the high cost of FDA approval kills people all over the world, not just in the USA. All those socialized medical systems? They don’t pay the NRE costs of drugs, just the marginal costs. So because we are not subsidizing them, their people are dying from the FDA delays too. So exponential growth, plus extrapolated from 300 million Americans to 2 billion people in the first world. How many is that? Hell if I know. But it ain’t forty.

    And of course, none of this takes into account morbidity. The damage done to patients short of death that the delays and lack of drugs brings on them. Easily multiply the death rate by ten to get the cost in lives ruined from serious medical conditions not averted.

    Now what is the flip side? In 2007 the head of the State Food and Drug Administration in China Zheng Xiaoyu was executed by the government for corruption and graft. He had been taking bribes from the largest drug companies in China to approve their medicines, This is about as bad as it gets. It is perhaps the scenario that the anti-freedom advocates would imagine in a world without the FDA. It was so bad that China entirely shut down the agency and created a new one from scratch, the CFDA.

    Under his watch he approved 150,000 medicines, mostly from the drug companies who had bribed him. How many people died from this? Of course it is hard to say, but there are forty documented deaths from a poison in a cough syrup. Forty. Not forty thousand. Forty. Of course that number is probably low, the real number based on the experts I have read is probably in the low thousands. But it sure as hell isn’t 50,000 or 100,000 every decade. However, who can argue with the legitimacy of the sentence of this corrupt man who stole the lives of fathers, mothers, sons and daughters simply for his own financial advancement? He deserved that bullet through his skull, the murderous bastard.

    So how many deaths did the FDA doctors of death deal? 380,000 from beta blockers, antibiotics and similar medicines that can be documented, plus cancer deaths that can’t easily but probably in the same order of magnitude, plus exponential growth worldwide of drugs not developed that is even harder to document, though no doubt the number is huge and growing. It doesn’t take much to see the millions of dead bodies at the hands of the FDA, never mind the lives ruined by delayed or undeveloped medical procedures that, although not life saving, were quality of life saving. And what do we trade the blood of these innocents for? Preventing the deaths of a few thousand people, visible, politically inflammatory, obvious, and so, from a politicians point of view, much more important.

    Of course it doesn’t make you feel any better when you gave your coughing kid some medicine to let them sleep only to wake to their cold dead body. But there is no perfect system, and if the government is going to rob us from something as fundamental as the right to take whatever damn medicine we want, in consultation with expert advice, they had better have a damn good case to make.

    The case they make is one of a small number of visible deaths against a gigantic number of hidden deaths, and the politicians made the evil choice they always make. I won’t advocate they suffer the same fate as Zheng. But they are just as murderously guilty as he, and they surely don’t deserve a fat and happy retirement on the public purse or a big fat seat on some drug company’s board.

    Here are some references to details I cite above:

    http://object.cato.org/sites/cato.org/files/serials/files/cato-journal/1985/5/cj5n1-10.pdf
    http://www.pfizer.com/research/clinical_trials/phases_of_development
    http://www.chinadaily.com.cn/china/2007-05/29/content_882475.htm

  284. “Trying to claim that isn’t an “extraordinary claim” because the believers are “enthusiastic” is nonsense. Folks are “enthusiastic” about aliens and Area 51 too.”

    I wasn’t taking their side (was my phrasing not cynical enough for you?). What I said was that the “extraordinary-ness” of a claim is fundamentally subjective and meaningless, so there’s no real value in platitudes like “extraordinary claims require extraordinary evidence”, no matter who they come from.

  285. “It is not possible to list the drugs that were never even created because the cost of adhering to the various regulatory regimes was too high a hurdle.”

    To be fair, it’s also not possible to list all the poisonous snake oil that didn’t get sold and all the people those things didn’t kill, so can we call it a wash and stick to things that can be listed?

  286. To be fair, it’s also not possible to list all the poisonous snake oil that didn’t get sold and all the people those things didn’t kill

    Funny, Jessica just provided an example of the actual harm caused in the regulatory fan’s worst nightmare.

  287. @Jessica
    Regarding drugs against cancer.

    Chemotherapy agents are extremely poisonous. More often than not, they reduces death rates with mere single digit percentages. E.g., cis-platin can increase 5 year survival rates in some cancers from 55% to 61%. This at the cost of very serious morbidities in all patients (very serious). Also, there is a very real risk an early improvement is followed by an even bigger worsening later.

    All in all, most “promising” cancer drug turn out to do more harm than good.

    With the small effects come very large patient samples to prove efficacy. That increases the costs. A real cure for cancer does not cost $1.2B. It can be done for a few hundred million.

    At the other end are the scandals of drug companies hiding dangerous side effects from the FDA. There are some every few years.

    So, I believe the millions of deaths caused by the FDA if the investigation really accounts for all the deaths prevented by not releasing drugs that were dropped late in the process because they were found to kill more patients than they saved.

    And I also want a scheme that does try to prevent a new thalidomide scandal. And new scandals like selling growth hormone as a cure for child malnourishment, or formula to mothers without clean water, or even poisoned formula.

    The backlash of such scandals are a seriius harm themselved. A billion chinese do not trust baby formula anymore. After the formula poisoning, Chinese people are still trying to buy so much Dutch formula that shops over here have to limit the number of items per customer to prevent an acute shortage in the Netherlands. (They are expanding production, but that only pushes the shortages down the supply chain)

    All in all I do not believe your claims the FDA killed a net 1% of the US population. I have seen too many studies where new promising cures turned out to be worse than what existed. Not properly accounting for that makes accusations against the FDA empty.

  288. I’m going to be boring and actually read the citations. The primary source is a paper from the Cato Journal which is a libertarian think tank. The assumption is that they will provide the most aggressive estimates of FDA deaths. Ignoring methodology (which we can debate in detail if desired) their numbers are:

    In sum, Table 1 shows estimated mortality reductions attributable to the use of new drugs of 50—102 per 100,000 from 1950 to 1977 (or 43—88 per 100,000 if tuberculosis is excluded). At this rate, it follows that a one-year delay in new drug benefits would cost 37,000 to 76,000 lives per decade in the U.S. population (32,000 to 65,000 excluding tuberculosis). By comparison, FDA delays in approving new drugs have often been estimated at two years or more.

    So using these worst case numbers, published by a libertarian think tank, the estimates range between 370K (37K * 2 years * 5 decades) and 760K (76K * 2 years * 5 decades). I think these numbers are inflated given the method in which they were calculated but even so they still don’t end up being “millions”.

    What is interesting is the admission:

    it is unclear, however, how much of the apparent slowdown in new drug development can be attributed to FDA regulation since it began some months before the 1962 amendments took effect, and a similar decline occurred in foreign countries.

    Until recently, the problem of FDA overregulation received little public attention. Throughout the late l960s and 1970s critics complained that the political pressures on the FDA were such as to systematically favor overregulation. This complaint was explained by the fact that new drug accidents, such as with thalidomide, were more widely publicized by the press than casualties from the drug lag, whose victims were largely anonymous.” However, criticism of the drug-lag problem finally mounted to the point where the FDA was forced to respond. In 1979 Congress held hearings on the drug- lag problem,’ prompting the FDA to establish a new “fast-track” approval system for important new drugs. Reforms to streamline new drug regulation were further pressed by the Reagan administration. These efforts succeeded in reducing approval times for important new drugs by as much as 40 percent, to an average of 19 months, while new drug approvals reached their highest level since the early 1960s

    So rather than the problem being invisible to congressional oversight it was noticed and the FDA did streamline approval times and greatly improved NDA (new drug approval) rates.

    I do reject cato’s assertion that simply providing warnings on the drugs will be sufficient.

    A casual google survey on prescription drug warnings results in pretty much every warning containing the term “…even death may occur” after the usual warnings that the drug may cause “nausea” etc. The warning that a drug can cause death or have serious side effects is so common as to be useless without study of the results of the clinical trials. Which even I don’t do…at most I look at the PDR and skim what it says and trust that the system works.

    Note that in order to provide such drug warnings so the consumer could make an informed decision on drug safety (and efficacy) the same clinical trials have to occur. If not then I wouldn’t have been able to have been warned by my pharmacist that two of the three anti-cholesterol drugs prescribed had interacted badly in those expensive clinical trials outlined in the Pfizer citation. Left to itself Pfizer won’t conduct those trials.

    As far as deaths attributable to an unregulated drug market isn’t “40 deaths from poisoned cough syrup”:

    Less than 1 percent of medicines in developed markets and as much as 30 percent or more in developing countries are fake, the Geneva-based WHO estimates. China and India are probably the biggest suppliers of the counterfeits, which cause 700,000 deaths a year among malaria and tuberculosis sufferers alone, according to the Washington-based International Policy Network.”

    http://www.bloomberg.com/news/articles/2012-05-17/stopping-fake-drugs-from-pakistan-is-too-late-for-victims

    Take that with the same large grain of salt as the numbers provided by the Cato Journal but it’s not “40 deaths”.

    How is the FDA and drug regulations helping?

    http://www.newsweek.com/2015/09/25/fake-drug-industry-exploding-and-we-cant-do-anything-about-it-373088.html

    They made a new instrument to help detect fakes AND they can ban generic drug manufacturers or even entire countries from the US drug market if their quality control is allowing fakes to appear in the US.

    Nevertheless, one of the reasons the U.S. hasn’t taken stronger international action is that most citizens don’t know the problem is so pervasive. Americans enjoy some of the strongest regulatory and enforcement systems in the world. Despite the occasional problematic batch here and there, what consumers buy at their local pharmacy in the States is probably genuine, which means Americans are less likely to push our leaders to make changes.

    “Oh but that’s fraud prevention and not drug approvals” is going to be the likely objection but the poisoned cough syrup was the same kind of example provided used to imply how this is some trivial problem that only leads to “40 deaths” or “low thousands” in comparison to the “millions of dead” from FDA drug approval delays.

    These issues are blocked by the “FDA doctors of death” and US drug regulation and enforcement. The fake drugs didn’t just appear in shady side-street pharmacies but in hospitals that didn’t know they were getting fakes from what they thought were safe channels.

    Also, when you look at the state of medicine prior to the FDA and other western regulatory agencies you see the same pattern of fake cures that were at best truly fake with no harmful ingredients and at worst poison. In unregulated markets today, it’s easier to fake a known brand/drug and not have to do your own snake oil advertising to drum up a market since the large pharmaceutical companies are doing all that marketing for you. All the predators have to do today is fake their labels and packaging.

  289. I wasn’t taking their side (was my phrasing not cynical enough for you?). What I said was that the “extraordinary-ness” of a claim is fundamentally subjective and meaningless, so there’s no real value in platitudes like “extraordinary claims require extraordinary evidence”, no matter who they come from.

    Noted. Next time I’ll settle for the less colorful “show me the data” or “citation needed”.

  290. Funny, Jessica just provided an example of the actual harm caused in the regulatory fan’s worst nightmare.

    No, she didn’t. The actual “worst nightmare” is the 700,000 deaths per year from fake malaria and tuberculosis medicine.

    Odds are that actual number of deaths (as opposed to the ones listed by both sides) that could be attributed to fake medicine in unregulated markets is far higher than the deaths due to drug delays in the heavily regulated markets. Given they faked aspirin and ended up with needing to withdraw a large lot of contaminated aspirin from the Pakistani market (which couldn’t have been THAT profitable to fake) means even the OTC drug market would be significantly negatively impacted by removing FDA regulatory oversight.

  291. I thought “that’s really dumb” and then remembered it’s a conference for server side javascript.

  292. @nht
    > I thought “that’s really dumb” and then remembered it’s a conference for server side javascript.

    Ah, on that point nht I will agree wholeheartedly. Server side JavaScript is perhaps one of the dumbest ideas I have ever heard. (Though Typescript makes JavaScript very, very much better.)

  293. @nht
    > . The actual “worst nightmare” is the 700,000 deaths per year from fake malaria and tuberculosis medicine.

    What a strange example to choose. You are aware that the only reason people still die of malaria in large numbers is because of regulatory overreach, specifically the banning of DDT? Rachel Carson is one of the biggest mass murderers in history.

    Read the reports on these fake malaria medications what you will find is that the patients can’t afford regular medications. For example this:

    Artemisinin-based combination therapy (ACT) is an important tool in the fight against malaria, but a sizable number of patients are unable to afford to this first-line treatment. Consequently, patients tend to procure cheaper anti-malarials, which may be fake or substandard.

    So it isn’t as if they are being duped into buying the wrong thing, it is the only choice they have. Why are these legitimate medications so expensive? Well, I think we have been talking about that at great length, haven’t we? So this is an example of the second order consequences of regulatory overreach, not more reason to bolster the power of these death dealing agencies.

    You have to have some compassion for these people. First they take away the only simple cheap means to rid themselves of the disease vector, then they make it impossible to get the real medicine and so they are left with no option but to go get it in the back alley with all the concomitant risks.

    Kind of like when you make abortion illegal or prohibitively difficult women get people to stick metal coat hangers in their cervixes. Or when you make marijuana illegal, cancer and glaucoma patients have to risk poisonous street drugs and violent street gangs to get the medicine they need.

    > Odds are …

    ?!? :-)

    > to fake medicine in unregulated markets

    Who is talking about unregulated markets? I’m certainly not. Let’s not limit our imagination to think only governments can regulate things. Winter gave us an excellent example above. Chinese company makes poor baby formula. Company’s reputation goes in the toilet and Chinese people buy their baby formula from the ever dependable Dutch. Dutch makers rewarded by more sales and ramp up their manufacturing and profits. Bad Chinese company goes out of business.

    By no means am I suggesting that that is the only appropriate regulatory mechanism, but the blood stained hands of the FDA need to be kept well away from this business. Life and death is far too important to be put in the hands of the politically appointed medical politburo of Hillary Clinton or Donald Trump.

    These slimy political appointees have the deaths of millions to their account,. When is enough enough?

    Anyway, I have been doing a netflix binge on Greys’ Anatomy, so watch out, I now know ALL about medicine…. :-)

  294. > Funny, Jessica just provided an example of the actual harm caused in the regulatory fan’s worst nightmare.

    Funny how the large corporations that have the money to bribe someone and the connections to know that they’re bribable still have incentives to protect their reputation that a fly-by-night snake oil salesman in a landscape where there’s truly no regulation won’t have.

  295. @Jessica
    DDT was fast becoming ineffective due to resistance in the target pests.
    https://en.m.wikipedia.org/wiki/DDT#Mosquito_resistance

    On the other hand, DDT was starting to appear in penguin eggs and breast milk and rising. It killed off most of the birds of prey in many countries.

    And, obviously, it was not banned for disease vector control, only for agricultural use. There is more to disease control than just spraying the world with broad spectrum poisons.

  296. Node.js is what all the cool kids are using…no wait, it’s Scala…or something.

    Whatever gets you up fast and then scales I guess.

  297. winter on 2016-09-09 at 18:01:32 said:
    > You do realize that Dutch formula is world class due to very strict regulation?

    Don’t be so coy Winter. I know you dutch are a very discerning and demanding people. Only the best for adorable little dutch babies. Except for those wooden shoes. That can’t be comfortable.

  298. Jessica Boxer:

    Except for those wooden shoes. That can’t be comfortable.

    I’m more curious as to whether the average Dutch is comfortable with having an Argentinean queen. :-P

  299. @Jessica
    “Except for those wooden shoes. That can’t be comfortable.”

    They are not comfortable (but you stuff them with hay). But they are a boon in the mud. And if a cow steps on your foot, you wish you were wearing wooden shoes.

  300. Chinese company makes poor baby formula. Company’s reputation goes in the toilet and Chinese people buy their baby formula from the ever dependable Dutch.

    No. Affluent Chinese parents source Dutch formula from overseas. Poor Chinese parents pay a premium on the street for Dutch formula that probably is fake or buys the Chinese version and hopes for the best.

    The poorer urban working moms that can’t nurse and can’t afford to pump are the ones that are forced to take that risk with their children.

    In comparison anyone in the US can go to Walmart or supermarket and buy the store brand formula and expect any food or drug to be safe and not a fake.

    Kings and nobles always got the best foods and medical care available anyway. The regulatory environment was created over time to protect commoners who didn’t have their own private force of guys with badges and guns or lawyers like the nobles.

    Everywhere you have an unregulated market andor a weak/corrupt central government unable or unwilling to enforce these regulations you see this outcome.

  301. > I thought “that’s really dumb” and then remembered it’s a conference for server side javascript.

    Nowadays it is less server side JavaScript, and more idempotent JavaScript (the same script for client side one-page app, the same script for server side pre-rendering).

  302. In comparison anyone in the US can go to Walmart or supermarket and buy the store brand formula and expect any food or drug to be safe and not a fake.

    Same with electrical products, because of the UL sticker on them.

  303. Same with electrical products, because of the UL sticker on them.

    Because it’s so hard to fake a UL sticker?

    Also UL is one of several Nationally Recognized Testing Laboratories authorized by OSHA. While there are no federal requirements for NRTL testing of all electric equipment (just those in the workplace) many local jurisdictions require a NRTL mark to pass local inspection.

    UL works because of the rest of the US regulatory, judicial and enforcement system backs them up with “guys with badges and guns”.

    That includes the dreaded rent seeking IP laws that protects the UL trademark.

  304. That includes the dreaded rent seeking IP laws that protects the UL trademark.

    It never ceases to amuse me when people bring up trademarks as evidence in favor of IP. To be fair, trademarks are the least objectionable and most useful part of the hydra that is IP….. and also the one in least need of special legal structures to handle it.

    Trademarks can be handled under fraud, and do not involve rent seeking.

    As usual, you demonstrate that you need to read a kindergarten level economics book. But credit where it is due; Jeff Read took his Idiot Ball back, so you aren’t rising to Full Of Shit as you were earlier.

  305. @FooQuuxman Lol…full of insightful comments as usual.

    Do you care to explain to Monster how the lack of research and development into uneconomic drugs like those that might treat Myasthenia Gravis falls under:

    “3. This is as it should be. That it is unprofitable (and no one is proposing a market failure here) is in itself evidence that it should not be happening. Period.”

    Even if you drive regulatory/testing costs to near zero (which wouldn’t happen in any viable scenario) it is still uneconomic to spend significant research dollars into new treatment regimes for illnesses that affect so few in the richer nations and the ROI is negative.

    What no brilliant comments on that?

  306. @nht:

    This isn’t a logic problem but a history problem.

    When we consider any idea, new or old, we must guard against carelessly accepting a false conclusion through error or cognitive bias. No mind and no field is immune to such ills, and therefore no idea is above scrutiny. Logic — as a formalized, systematic method of rational scrutiny — therefore applies to all arguments, whether phrased as a formal problem or not.

  307. >To be fair, it’s also not possible to list all the poisonous snake oil that didn’t get sold and all the people those things didn’t kill,

    And yet that’s exactly what the pro-regulation folks do. They list the snake oil that they allowed to be sold for a while, the number of people killed before they stopped allowing it, divide by the number of years they allowed it to be sold, then assure us they continue to save that number of people every year in perpetuity. Never do they consider the idea that when people found out what the regulators found out, they’d decide on their own to stop using the stuff that was killing people.

    > Even if you drive regulatory/testing costs to near zero (which wouldn’t happen in any viable scenario) it is still uneconomic to spend significant research dollars into new treatment regimes for illnesses that affect so few in the richer nations and the ROI is negative.

    But MG isn’t one of those. It had enough of a market to justify the R&D on pyridostigmine bromide decades ago, but that was when the compliance costs were much lower than they are now. I don’t think it could be developed today, however.

    >It never ceases to amuse me when people bring up trademarks as evidence in favor of IP.

    “IP” is an anti-concept. It subsumes patents, trademarks, copyrights, and trade secrets, each of which is a different thing. Grouping them together is an attempt to give some of them credit for attributes of others, as dealing with them individually weakens their support.

    >Also UL is one of several Nationally Recognized Testing Laboratories authorized by OSHA.
    UL earned its reputation decades before OSHA even existed. Had OSHA not recognized the value of UL testing, even its staunchest defenders would have trouble defending its decision to do so.

  308. @The Monster
    “Never do they consider the idea that when people found out what the regulators found out, they’d decide on their own to stop using the stuff that was killing people.”

    Any evidence of that? People still spend large amounts of money on Homeopatic drugs although it has been proven in every conceivable way that they do not work.

    On the other hand many people refuse to vaccinate their children even though it has been proven conclusively that not vaccinating endangers the life of your children.

    Even now “believers” are able to dominate online and social media to the point that all science is discarded.

    And you claim that after the retraction of people will suddenly get informed and shun dangerous and ineffective drugs? I simply do not believe you.

    What would happen would be that people would be drowned in conflicting information that would make rational decisions all but impossible.

  309. > “IP” is an anti-concept.

    Copyright itself is also such an “anti-concept”, grouping ownership of character and setting ideas with ownership of text.

  310. Any evidence of that? People still spend large amounts of money on Homeopatic drugs although it has been proven in every conceivable way that they do not work.

    Evidently historical evidence is not required but merely the assumption that humans would behave in a logical fashion. Poorly formed logic at that that.

  311. @alex K

    “When we consider any idea, new or old, we must guard against carelessly accepting a false conclusion through error or cognitive bias.”

    We should guard against carelessly formed logic as well. You jumped from premise to conclusion that was not supported by either logic or the constraints of the premise. Moreover it was a conclusion that ignores reality. As in we don’t exist in a state of civil war even though we have MWB&Gs at different levels of authority in the US and some prove to be bad actors.

    Reality trumps logic every time.

    The flaw in your logic is that while MWB&Gs are listed as a requirement by me, it’s not the only requirement and was never stated as such. The statement “MWB&G is required to constrain Bad Actors” does not mean that ONLY MWB&Gs are required. Especially since the next sentence talks about Rule of Law. (And centralized government. Easier to make the case against centralized governments. If you would like to go down that path we can discuss the military efficacy of imperial armies vs individual warriors…especially from a logistics perspective)

    To disprove my statement logically you would have to show that having MWB&Gs is NOT a requirement for western civilization through logical proof. Good luck with that. It would be easier to do proof by counterexample.

  312. @nht
    “Evidently historical evidence is not required but merely the assumption that humans would behave in a logical fashion.”

    “Rational” is a better concept in this context. Homeopathy is logical, but based on irrational assumptions.

  313. > And you claim that after the retraction of people will suddenly get informed and shun dangerous and ineffective drugs? I simply do not believe you.

    I do not doubt that there will be people who make bad decisions, some of which will cause them harm, up to and including losing their lives.

    Their lives. Their choices. Not mine, nor yours.

    > What would happen would be that people would be drowned in conflicting information that would make rational decisions all but impossible.

    And somehow these idiots who believe in homeopathy, and are incapable of making rational decisions, are expected to elect the people to make such decisions on their behalf. The Fundamental Contradiction can’t just be hand-waved away.

    >Evidently historical evidence is not required but merely the assumption that humans would behave in a logical fashion

    Legislators, administrators, regulators, police, wardens, prison guards, and other humans comprise government. Why should we assume that they will be any more logical than the voters who elect them?

    >To disprove my statement logically you would have to show that having MWB&Gs is NOT a requirement for western civilization through logical proof. Good luck with that. It would be easier to do proof by counterexample.

    MWBAGs are essential to “government”. When one advocates for “government” to do ${THING}, one is in fact proposing that MWBAGs compel those who do not wish to comply voluntarily with their part in the Great ${THING} Project. I do not argue against the existence of MWBAGs themselves, for unlike ESR and a few others around here, I am not an anarchist. The argument is over what MWBAGs should be empowered to do.

    Governments exist to protect against predation and punish predators. I am just fine with having MWBAGs use their guns against murderers, rapists, and others who (attempt to) transgress against the person or property of others by force or fraud, and against those who act in a reckless manner likely to cause such harm inadvertently. That all flows logically from the right to self-defense. But we do not grant them this power arbitrarily. When they exceed their authority, we punish them just as we do any other aggressor.

    I can’t find the moral justification to extend that authority to the crusade against self-destructive behavior. Vices are not crimes. Irrational, self-destructive behavior is self-punishing. The sole function of MWBAGs against such behavior should be to limit collateral damage. Thus, I can’t support laws requiring motorcyclists to wear helmets to protect their noggins, but “You must wear either a helmet with a face shield OR goggles to keep crap out of your eyes” is legit because it protects others on the road. (And if they drive on their own property, I can’t even go that far. No one else is forced to join them.)

    In the rare cases where someone really is just plain incompetent (such as minors or severely mentally-handicapped) then the burden of proof is on those who wish to have them judged thus. I will err on the side of liberty if the evidence isn’t beyond a reasonable doubt, and accept that some people are going to qualify for Darwin Awards as part of the price we pay for freedom. (If we’re talking about not vaccinating children, or treating them with homeopathic “medicine”, the standard’s a bit lower due to the fact the child by definition isn’t a consenting adult. So a helmet law for minors is justifiable.)

    And if they really aren’t competent to make decisions for themselves, they damned sure aren’t competent to vote to make them for us all. Laws that forbid the general public the freedom to do things some of us think are bad for them (while simultaneously insisting they have a “right to vote” those laws upon all) fail due to the Fundamental Contradiction.

  314. Copyright itself is also such an “anti-concept”, grouping ownership of character and setting ideas with ownership of text.

    Not in the United States, where (in the law, at least) specific expressions of an idea can be protected while the underlying idea itself cannot.

  315. > Homeopathy is logical, but based on irrational assumptions.

    Diluting something to the point that there is less than one molecule of it in an average dose (and therefore many “doses” literally have none of the allegedly-active ingredient), and expecting that to have any beneficial effect, is neither logical nor rational.

    But I can’t justify shooting someone who does it, so I can’t justify paying a guy to wear a badge and shoot them on my behalf for doing it. And that means I can’t justify making a law that says he can’t do it, because that’s what a law does.

  316. The Monster:

    I don’t think [pyridostigmine bromide] could be developed today, however.

    So we do need the State to correct market failures in healthcare? I ask honestly; while I have libertarian leanings, I also have doubts.

    The ideal solution, I think, would be a private effort to develop unprofitable drugs with the aid of crowdfunding. But until that materializes, it may be legitimate to turn to the State for a temporary solution (and hope the solution isn’t rendered ineffective by the nature of bureaucracy). After all, the State won’t disappear anytime soon; we might as well make the best of it.

    My knowledge of these issues is negligible, so I’m more-than-open to criticism (I’d say “counterarguments”, but what I wrote isn’t an argument – it’s more like brainstorming).

    “IP” is an anti-concept. It subsumes patents, trademarks, copyrights, and trade secrets, each of which is a different thing.

    I see you agree with RMS, at least on that issue. I suppose you’re both right, but – as usual – I don’t know enough to be sure.

  317. Well I guess since people still use homeopathic treatments despite evidence that they don’t work, the obvious solution is to regulate them off the market, right? I mean, since it works so well in other cases, like heroin… uhh, or, umm, cocaine…

  318. @Paul Brinkley
    “Well I guess since people still use homeopathic treatments despite evidence that they don’t work, the obvious solution is to regulate them off the market, right? ”

    No. But it is prohibited to sell homeopathic products with the claim that they are medically effectivd and cure disease.

    What is your problem with that?

  319. @The Monster
    ” accept that some people are going to qualify for Darwin Awards as part of the price we pay for freedom. ”

    Implicitely, you are subjecting the gullible and ignorant to murderous fraud. There are limits to what can be done to protect the stupid. But it is not nothing.

    Note that evolution works by killing of a large fraction of the population. Not a society I would like to live in.

  320. Well I guess since people still use homeopathic treatments despite evidence that they don’t work, the obvious solution is to regulate them off the market, right? I mean, since it works so well in other cases, like heroin… uhh, or, umm, cocaine…

    Or just have laws that say you can’t sell them as medicine but as it’s own (very odd) drug category? Nah.

    All the caterwauling about how the FDA restricts everything so you can’t do anything out of the mainstream ignores that a) you can sell your fake cure as homeopathic “medicine” b) you can sell your fake cure as a “dietary supplement” or c) as an individual you can bring it into the country for personal use while the FDA tells the MWB&Gs to ignore that as long as it’s not some other restricted substance and you aren’t trying to sell it to anyone else.

    Meanwhile, it’s a whole lot safer here than in China or Pakistan where fake drugs are prevalent. And if they can fake drug packaging, they can fake a UL sticker or any other kind of private accreditation that isn’t backed with MWB&Gs.

    Who, by the way, will arrest you and (probably) not shoot you for this kind of crime.

  321. @nht
    Because everything can be expected to be cheaper, certification will be cheaper, even at whatever the “gold standard” of certification is. And of course there will be certifications that are less than gold, as well as non-certified treatments, also, due to the previously mentioned wealth it will be easier in general to fund research. The easier it is, the lower the minimum price of doing that research, and that is what determines if you need to pyramid build in order to fund it. All other solutions are handwaving over the problem of insufficient wealth.

    The idea of a private moon shot was unthinkable 50 years ago. Now any billionaire who’s fancy is tickled by the idea can try it. When you understand why that is the case you will understand medical economics.

    @winter
    Waaaay upthread you commented on 3rd world countries having bad medical infrastructure as evidence that The Market Can’t Do Meds Bro. You are confusing matters: it takes wealth to grow wealth. And if you have very little wealth then no amount of pretending otherwise, or dumping it into massive Public Works projects is going to change that. You just have to grow it the hard way. Now what system will grow that wealth, that is a legitimate question.

    I would think that this error couldn’t stump a 2yr old child, and yet I see it on a regular basis from putatively intelligent adults. Presumably it is caused by the same mechanism that results in 90% of Tech Support stories.

  322. @Foo
    “You are confusing matters: it takes wealth to grow wealth.”

    Still, some well organized states are able to do much better in public health and medical care than equally poor countries that are less well organized with a “free” unregulated market.

    There is not much wealth needed to do basic health care if you organize it well, e.g., 1950s NHS in the UK. But a lot of money is needed to replicate a US style system with a free market in medical care. It is even questionable whether there is enough wealth in the US to do it.

  323. All the caterwauling about how the FDA restricts everything so you can’t do anything out of the mainstream ignores that a) you can sell your fake cure as homeopathic “medicine” b) you can sell your fake cure as a “dietary supplement”

    Oh, look, you’re lying again. Short rebuttal: kratom.

  324. > So we do need the State to correct market failures in healthcare? I ask honestly; while I have libertarian leanings, I also have doubts.

    This presumes someone who is smart enough to recognize that the market has failed, who has superior knowledge to the billions of people in the market. Time and time again when people insist “market failure” has occurred, it’s actually due to government trying to apply that allegedly-superior knowledge and screwing things up.

    > The ideal solution, I think, would be a private effort to develop unprofitable drugs with the aid of crowdfunding

    I agree. Now ask yourself why it isn’t already happening. One obvious answer is that capital markets are heavily regulated to protect people against their own stupidity, making raising capital in conventional ways prohibitively expensive. Crowdfunding itself is only allowed to happen now because those regulators haven’t figured out how to cripple it yet. But given time, agencies like SEC will demand (and get) regulatory authority over crowdfunding, at which point they’ll burden it with so much compliance costs that it’ll largely go away.

    – – –
    > Implicitely, you are subjecting the gullible and ignorant to murderous fraud. There are limits to what can be done to protect the stupid. But it is not nothing.

    Fallacy of the Excluded Middle. 15 yards and loss of down.

    I’m not “subjecting” anyone to anything of the sort. I’ve already said that actual fraud is something government morally can act against. So if it’s fraud, prosecute it as such. But that can’t be a blank check to silence people who question the “scientific consensus”, or we get Galileo being forced to recant heliocentrism.

    I am all in favor of trying to educate people, but I draw the line at denying them the freedom to do “stupid” things just because someone thinks they’re stupid. Lots of people enjoy things I think are stupid, and I’m sure I enjoy things plenty of others think are stupid.

  325. But it is prohibited to sell homeopathic products with the claim that they are medically effectivd and cure disease.

    What is your problem with that?

    My problem with that is relatively mild, but exists to the extent that it requires providing evidence which you had *just* claimed does no good.

    Why are you advocating busywork?

  326. @The Monster
    “This presumes someone who is smart enough to recognize that the market has failed, who has superior knowledge to the billions of people in the market.”

    People die from fake medicine in considerable numbers. Sounds not like you need a genius for that.

    @The Monster
    ” I’m not “subjecting” anyone to anything of the sort. ”

    You tell us we should remove all government regulations to be replaced by free market solutions.

    @The Monster
    ” So if it’s fraud, ”

    If a person really believes in faith healing, exorcism, or radium water, it is not fraud if he sells it. It is still dangerous and will not work. Fraud is only of very limited value. Moreover, without registration and transparency, it is impossible to decide whether a drug is a fraud.

  327. The Monster:

    This presumes someone who is smart enough to recognize that the market has failed, who has superior knowledge to the billions of people in the market.

    Normally, I would accept this Hayekian reasoning. But you’d said:

    There isn’t much happening with new drugs to treat [Myasthenia Gravis] because it only affects 50~200 out of every million people.

    Is that a matter of knowledge? If I understand correctly, you’re saying that the research on new drugs is unprofitable not because of a lack of knowledge, but because there are relatively few MG patients.

    If no market actors are going to invest in the innovation that would help your wife and others like her, then we need a community effort to make it happen, either via the State’s apparatus or – preferably – the voluntary charity of crowdfunding. (Alas, the former appears to be unreliable, and you’ve made a compelling case that the latter’s going to die out.)

    In any case, I’m sorry that your wife suffers from a chronic ailment. If, in my ignorance, I’ve misrepresented the situation in any way, I apologize.

  328. If I understand correctly, you’re saying that the research on new drugs is unprofitable not because of a lack of knowledge, but because there are relatively few MG patients.

    You omitted the crucial fact that the regulatory regime imposes crushing non-recurring capital costs on drug development.

  329. @Christopher Smith
    ” You omitted the crucial fact that the regulatory regime imposes crushing non-recurring capital costs on drug development.”

    Just to confirm I understand you. You are deploring the fact that companies must prove that their drugs are safe and effective? You prefer a world where it is not necessary to prove safety and effectiveness of medication?

  330. > You tell us we should remove all government regulations to be replaced by free market solutions.

    No, I do not. Stop saying I do. Government’s role should be confined to dealing with force and fraud. That is not the same thing as “remove all government regulations”.

    >Moreover, without registration and transparency, it is impossible to decide whether a drug is a fraud.
    If that were all the FDA/DEA claimed to do, I wouldn’t have much problem with them. But it’s not, and you know it’s not. They claim the authority to forbid people to consume certain substances to protect them against their own ignorance.

    >Is that a matter of knowledge? If I understand correctly, you’re saying that the research on new drugs is unprofitable not because of a lack of knowledge, but because there are relatively few MG patients.

    No, it’s the combination of the research and the high costs of regulatory compliance. It’s easier to bear those costs for a drug that treats a common illness. Every additional regulation that is promulgated adds to the costs, and at some point pushes a disease into the “orphan” category.

  331. @The Monster:

    >I do not doubt that there will be people who make bad decisions,
    > some of which will cause them harm,
    > up to and including losing their lives.
    >Their lives. Their choices. Not mine, nor yours.

    And I agree with you entirely when it comes to things like homeopathic drugs. But when it comes to vaccinations for ANYTHING but the flu, or for anyone but doctors, then it is no longer entirely their lives. Those who refuse to take MMR or DTaP are jeopardizing herd immunity and harming those who can’t take it or who are not immune even though they took it. Those who refuse to take flu shots (if they aren’t doctors) are harming only themselves, and as such should probably only be punished with marginally higher premiums. (Yes, flu shots should cost negative amounts from the insurance company. That’s just good business. It’s cheaper to pay for a flu shot than to pay for flu care.)

  332. You are deploring the fact that companies must prove that their drugs are safe and effective?

    To what standard?

  333. @Chriztopher
    “To what standard?”

    Better than what was asked of Thalidomide and DES I presune?

  334. Christopher Smith said:

    You omitted the crucial fact that the regulatory regime imposes crushing non-recurring capital costs on drug development.

    The Monster said:

    No, it’s the combination of the research and the high costs of regulatory compliance. It’s easier to bear those costs for a drug that treats a common illness. Every additional regulation that is promulgated adds to the costs, and at some point pushes a disease into the “orphan” category.

    In that case, I recant. Thank you both for showing me my error. There’s no justification for governmental intervention in this area after all (or any other area), for it’s hindering progress in life-saving technologies. To make matters worse, Obamacare hinders access to the life-saving technologies that already exist, by making healthcare outrageously expensive and destroying the very jobs people need to pay for it (and for food and many other things).

    And I insist that even if some honest people has benefitted from Obamacare, that doesn’t atone for the harm it’s doing to other honest people.

    I regret having thought that there might be a legitimate role for the State, and hereby ratify my anarchism.

  335. @The Monster
    ” No, I do not. Stop saying I do. Government’s role should be confined to dealing with force and fraud. That is not the same thing as “remove all government regulations”.”

    As we were discussing the FDA, I was refering to regulations on medications. And you just confirmed that you indeed want them removed completely.

  336. > As we were discussing the FDA, I was refering to regulations on medications. And you just confirmed that you indeed want them removed completely.

    No, what I want removed completely is the FDA’s power to ban medications. I don’t have a problem with the FDA requiring unapproved medications to be labeled as such, which is a form of regulation. Give people information, but leave them free to choose what to do with that information.

  337. In that case, I recant.

    I’m curious why? They made two unsupported assertions that IF true may lead you to believe that an unregulated market MIGHT be better but only IF they could also show the harm asserted from the cost of the clinical trials outweighs the good. They would also have to show that privatizing drug trials is effective and actually cheaper.

    Otherwise you recanted based on no evidence or new information.

    My issue with the discussions is that much of the argument is based on easily disproved “facts” (no coral snake venom, DDT banned for killing mosquitos, deaths from deregulated markets is only in the double digit order of magnitude, etc) that there isn’t an honest assessment of whether one should “recant” and whether “recant” isn’t actually the proper word since these beliefs appear to be based on faith in a system that has never been shown to work at scale with actual humans and on dubious “facts” that “everyone knows”.

    It’s really a lot like arguing with extremely religious about God or the literalness of the Bible. Note that I am a Christian just as I am a Conservative nominally for smaller government, less regulations, etc. I’m fully aware that my biases and beliefs are not all backed by data and discriminate between what is fact and what is faith.

    Most importantly when given data that doesn’t correspond to my “facts” I examine them and I never have to “recant” or claim there is “no justification” for something I believed but just change my opinion based on new data. But it has to be real data. Not just assertions, handwaving or theory.

  338. nht,

    “When the facts change, I change my mind. What do you do, sir?” –John Maynard Keynes

    God, no wonder he gets market fundies’ panties in such a twist.

    1. >http://quoteinvestigator.com/2011/07/22/keynes-change

      That line is much older than Keynes. I’m certain I’ve seen it attributed to a prominent Victorian, possibly Disraeli or Gladstone. Or could be Charles Babbage.

  339. It saves lives.

    My point in that very post is that it doesn’t, by your own argument.

  340. I quit, I give up. Nothing’s good enough for anybody else, it seems. And being alone is the best way to be.

  341. @Jorge Dujan

    I regret having thought that there might be a legitimate role for the State, and hereby ratify my anarchism.

    If you never come across data that challenges your beliefs you aren’t pushing hard enough.

    I quit, I give up. Nothing’s good enough for anybody else, it seems. And being alone is the best way to be.

    The simplest solution is to create a basement universe for yourself, then pull the wormhole access in after you. I’d lend you my equipment, but once the process is complete I will be the only entity that exists in the universe, Solipsism FTW!

  342. I’m interested in why? They made two unsupported affirmations that IF genuine may persuade that an unregulated market MIGHT be better yet just IF they could likewise show the mischief stated from the expense of the clinical preliminaries exceeds the great. They would likewise need to show that privatizing drug preliminaries is powerful and really less expensive. and check my website dailyimagefunda.

  343. What a line:

    “Trust simple, crude metrics over complex ones because the simple ones are less brittle. KLOC is best, though a poor best.”

Leave a Reply to trying2b-amused Cancel reply

Your email address will not be published. Required fields are marked *