This morning I stumbled over a comment from last September that I somehow missed replying to at the time. I suspect it’s something more than one of my readers has wondered about, so here goes…
Edward Cree wrote:
If I’m really smart enough to impress esr, I feel like I ought to be doing more with myself than toy projects, games, and an obscure driver. It’s not that I’m failing to change the world, it’s that I’m not even trying. (Not for want of causes, either; there are plenty of things I’d change about the world if I could, and I suspect esr would approve of most of them.)
Obviously without Eric’s extroversion I won’t be as influential as him, but… dangit, Eric, what’s your trick? You make having a disproportionate effect on the course of history look easy! Why can I never find anything important to hack on?
There are several reasons people get stuck this way. I’ve experienced some of them myself. I’ve seen others.
If this sounds like you, dear reader, the first question to ask yourself is whether you are so attached to having a lot of potential that you fear failing in actuality. I don’t know Edward’s age, but I’ve seen this pattern in a lot of bright young people; it manifests as a lot of project starts that are potentially brilliant but a failure to follow through to the point where you ship something that has to meet a reality test. Or in an opposite way: as self-constraining to toy projects where the risk of failure is low.
So my first piece of advice is this: if you want to have “a disproportionate effect on the course of history”, the first thing you need to do is give yourself permission to fail – as long as you learn something from every failure, and are ready to keep scaling up your bets after success.
The second thing you need to do is finish something and ship it. No, more than that. You need to make finishing and shipping things a habit, something you do routinely. There are things that can be made to look easy only by cultivating a lot of self-discipline and persistence. This is one of them.
(The good news is that once you get your self-discipline to the required level it won’t feel like you have to flog yourself any more. It’ll just be habit. It’ll be you.)
Another thing you need to do is actually pay attention to what’s going on around you, at every scale. 99% of the time, you find important things to hack on by noticing possibilities other people have missed. The hard part here is seeing past the blinding assumptions you don’t know you have, and the hard part of that is being conscious of your assumptions.
Here’s my favorite example of this from my own life. After I described the many-eyeballs-make-bugs-shallow effect, I worried for years at the problem of why nobody in the hacker culture had noticed it sooner. After all, I was describing what was already a decades-old folk practice in a culture not undersupplied with bright people – why didn’t I or anybody else clue in faster?
I remember vividly the moment I got it. I was pulling on my pants in a hotel in Trondheim, Norway, idly chewing over this question yet again. It was because we all thought we knew why we were simultaneously innovating and achieving low error rates – we had an unexamined, unconscious explanation that suited us and we never looked past it.
That assumption was this: hackers write better software because we are geniuses, or at least an exceptionally gifted and dedicated elite among programmers. Our culture successfully recruits and selects for this.
The insidious thing about this explanation is that it’s not actually false. We really are an exceptionally gifted elite. But as long as you don’t know that you’re carrying this assumption, or know it and fail to look past it because it makes you feel so good, it will be nearly impossible to notice that something else is going on – that the gearing of our social machine matters a lot, and is an evolved instrument to maximize those gifts.
There’s an old saw that it’s not the things you don’t know that hurt you, it’s the things you think you know that ain’t so. I’m amplifying that: it’s the things you don’t know you think that hurt you the most.
It’s not enough to be rigorous about questioning your assumptions once you’ve identified them. The subtler work is noticing you have them. So when you’re looking for something important to hack on, the question to learn to ask is: what important problems are everybody, including you, seeing right past? Pre-categorizing and dismissing?
There’s a kind of relaxed openness to what is, a seeing past preconceptions, that is essential to creativity. We all half-know this; it’s why hackers resonate so strongly with Zen humor. It’s in that state that you will notice the problems that are really worth your effort. Learn to go there.
As for making it look easy…it’s only easy in the same way that mastery always looks a skill easier than it is. When someone like John Petrucci or Andy Timmons plays a guitar lick with what looks like simple, effortless grace, you’re not seeing the years of practice and effort they put into getting to where that fluency and efficiency is natural to them.
Similarly, when you see me doing things with historical-scale consequences and making it look easy, you’re not seeing the years of practice and effort I put in on the component skills (chopping wood, drawing water). Learning to write well. Learning to speak well. Getting enough grasp on what makes people tick that you know how to lead them. Learning enough about your culture that you can be a prophet, speak its deepest yearnings and its highest aspirations to it, bringing to consciousness what was unconscious before. These are learnable skills – almost certainly anyone reading this is bright enough to acquire them – but they’re not easy at all.
Want to change the world? It’s doable. It’s not magic. Be aware. Be courageous. And will it – want it enough that you accept your failures, learn from them, and never stop pushing.
Considering that I have to get back to doing this (since I’ve kind of slacked off on this sort of thing for about at least the last Four or so years…sigh…) I heartily agree with what Eric’s suggesting here.
The problem with changing the world is that it is it’s diaper is what needs changing, and it’s a stinky, dirty, job. The hand that rocks the cradle rules the world.
In the general economy there are trillions in debt. That will collapse.
In cyberspace, there is a similar ponzi in Technical Debt. Unpatched things, inefficiencies (that make our phones and tablets run 10x slower, and the greens don’t care that it means 10x more energy – but we’ll have wind and solar soon).
But technical debt doesn’t result in real debt, at least not reliably. I could easily code things in the IoT space that wouldn’t’ have holes or have fewer holes. I was offered a job at a supplier but I was too expensive, but someone managed to hack an entertainment system to do some really bad things like shut off the engine over the cell network. They obviously saved meatspace money. (self driving cars?).
The problem is you can’t change the world, at least not human beings that are marginally greedy and slothful. Look at what happened to NTP. Only when things are collapsing do people pay attention, and hope the barbarians won’t be able to cross the gate before it is reinforced.
(My current situation is specifically wonderful because a company had years of technical debt, and finally had a melt-down, and I spent a year just going through the code fixing simple, often silly things).
One of the seminal works I read AS A TEENAGER was “The Mythical Man Month”. The world hasn’t changed although that has been tribal knowledge for DECADES. But is ignored. As is the further learning on the topic like Demarco and LIster’s Peopleware. THREE DECADES and we still have an idiocy, even in the high-tech area. This is some kind of Women’s day, and the Tech giants, or at least most of them, are more worried about “diversity”.
Change the world? Isn’t Climate Change doing that? Faster please!
“You’re not Don Knuth.
It’s okay if you ship something with bugs, as long as it works pretty well.
Fix it in a later release.
Ship it.”
Another place of refuge is a big team where you’re a cog no more responsible for the outcome than a spare wheel in a parked Jeep.
(@esr: did you _not_ mark an excerpt point on purpose?)
For years now I’ve struggled with much the same thing, though my problems stem mostly from having interests which are profoundly diffuse and fractured. But last year I committed to a major, long-term endeavor and successfully completed it, writing a lot about my process along the way.
I called it “The STEMpunk Project; the name is a portmanteau of ‘STEM’ and ‘steampunk’, and the project was aimed at spending a year teaching myself as much computing, electronics, mechanics, and artificial intelligence as humanly possible. As of now I’m actively writing a book about the experience, but in the meantime I thought I’d share some of the most germane essays I’ve written to complement ESR’s advice.
“So my first piece of advice is this: if you want to have “a disproportionate effect on the course of history”, the first thing you need to do is give yourself permission to fail – as long as you learn something from every failure, and are ready to keep scaling up your bets after success.”
This is an enormous hurdle which must be overcome if one is to tread the path of greatness. An important first step is knowing what to do you when you feel inadequate. The central message of the linked post is that failures are almost always better understood in the broader context of your efforts. Anything you’re good at is something you started off doing poorly, and some of history’s greatest titans were beset constantly by self doubt.
Moreover, it’s vital to remain in what is popularly called a ‘growth’ or ‘incremental’ mindset. Your failures form a ladder, with each failure forming a rung getting you ever closer to the goal hanging above you.
When you’re undertaking something which is completely new to you it can be difficult to know how best to go about learning from your mistakes. I’ve written a bit about performing a failure autopsy, and I think the advice there has the potential to help people leverage failure in a productive way. I haven’t gotten around to writing up the real version yet, though Eric has seen a rough sketch of it and believes it to be compelling.
I have also benefitted from using ‘mantra stacks’ as a means of sharpening focus, curtailing frustration, and sustaining motivation. And it can’t hurt to spend time studying the group of people, like Sebastian Marshall, Scott Young, and Cal Newport who I’ve dubbed The Ultrapraxists. They’re geniuses at manipulating workflow and tackling enormous projects. Finally, don’t be concerned if you spend a lot of time in a frustrating semignostic state, in which you spend hours grappling with a difficult problem with nothing to show for it but a vague sense that something has been accomplished.
I’m making rapid progress on the STEMpunk Project book, but in the meantime I’d welcome anyone who has questions to check out rulers to the sky, my personal blog, or reach out to me personally.
Hi, querent here. I should note that in the original comment I may have been subconsciously saying things I knew Eric would want to riff on, and this led me to perhaps emphasise the wrong things. I am able to ship finished code (sometimes, though perhaps not quite “as a habit”); I’m even sometimes able to see past assumptions and come up with the kind of solution that afterwards makes me wonder why no-one saw it before. Indeed, it’s more the elbow grease than the flashes of insight that I lack. (My motto: “The impossible we do immediately. The difficult takes a little longer.”) As for reality tests, I’m one of the maintainers of the Linux driver for the network card that runs the world’s stock exchanges (and yes, that responsibility does sometimes keep me awake at night).
I really hope the above doesn’t sound like boasting, because that’s not what I’m aiming for. Instead, I’m trying to clarify that the Eric-skills I’m trying to understand are those that aren’t about engineering, the ones that get hurriedly mentioned in the penultimate paragraph. Actually, even that’s not quite true, because I understand enough about comparative advantage to know that it’s inefficient for me (somewhat more aspie than the hacker average) to spend my time “grasp[ing] how people tick”. What I’m really looking for, as encapsulated in my closing question “Why can I never find anything important to hack on?”, is some way of changing the world by writing software, and it wasn’t until I actually thought about it just now that I realised that isn’t what Eric did. Not to devalue the code he has written, but it’s essentially orthogonal to his anthropology and culture-hacking; sure, he had to have written something in order to have standing, but it didn’t really matter what.
And sure, he may have also changed the world by shaking up DNS, but from the way he told the story it sounds like he was just in the right place at the right time; although it also sounds like he manages to make a habit of being in the right place at the right time, and if that’s a learnable skill then teach me, oh guru.
>although it also sounds like he manages to make a habit of being in the right place at the right time, and if that’s a learnable skill then teach me, oh guru.
Teach you? Gladly, if and when I ever get a clue about how I do that.
I agree that this sort of thing is a frequent enough occurrence in my backstory to justify a suspicion that I’m doing something productive of right-place-right-time moments. Fscked if I know what, though.
Here are a couple of possible theories:
1. People have a lot more potential RPaRT moments than they realize, but usually don’t notice them. On this theory my ability is simply an above-average propensity to notice.
2. It is possible to increase your incidence of RPaRT moments by having personal traits that both draw you to those nexus points and improve your ability to Be That Guy when you’re at one. Obvious candidates include curiosity, an interest in being at the cutting edge of change, and a broad knowledge base.
3. Malcom Gladwell’s age-cohort theory of what happens around technology transition points may be relevant.
On reflection, I’ve decided that Edward reached two conclusions that aren’t quite right.
>What I’m really looking for […] is some way of changing the world by writing software, and it wasn’t until I actually thought about it just now that I realised that isn’t what Eric did. Not to devalue the code he has written, but it’s essentially orthogonal to his anthropology and culture-hacking; sure, he had to have written something in order to have standing, but it didn’t really matter what.
But I have changed the world purely by writing software, at least once. (I think there are other cases, too, but they’re less clear-cut.) Porting GIFLIB to Unix, and improving it in the process, was the enabling technology for the graphical web browser. If you want to attribute that to Marc Andreesen for being smart enough to pick up my code and use it in Mosaic, I’m OK with that. One of us changed the world purely by writing software, or both of us did. It is possible.
As for “having standing” and my code being orthogonal to the anthropology and culture-hacking, I think this is also subtly wrong. It did matter what I wrote, because one can imagine my having written code that would not have the longevity and pervasiveness to give me standing. I don’t think it’s irrelevant that any hacker curious about me is going to rapidly find out about things like GIFLIB and Emacs VC mode and nethack and GPSD. Not all of my projects have outlasted the specific circumstances they were written for, but several have had what the publishing biz calls “long legs”.
I don’t think you can plan for that, really. But you can put your effort in directions where it’s more likely to have long legs. I think my longstanding interest in infrastructure and service code is therefore relevant. What I’m mainly working on now, NTPsec, seems likely to have very long legs. So is reposurgeon, though it’s only six years old now.
So I’m going to say that if your objectives include either (a) changing the world through software, or (b) developing the cred to project your influence the way I have, then try to choose objectives that aren’t tied to this month’s or this year’s fad. A good rule of thumb, maybe, is that if you can’t realistically imagine people wanting to use something in ten years, put your effort elsewhere.
Humans tend to focus on our strengths and not our weaknesses, for it is easy to do. The tougher thing is to drill on our weaknesses. It takes discipline and effort. ESR seems to have a habit of addressing his weaknesses and adding skills in those areas. Example: I remember he mentioned he has a congenital muscle weakness (CP??) but has worked to master the martial arts. Many choose pour more and more of their energy and focus into their areas of strength and accomplishment. I believe this actually tends to narrow our ability to influence and lead. Remember what defines a leader….FOLLOWERS. If people will not follow you, you are not a leader. If you are a talented developer, lets say in the 95th percentile compared to your peers and work extraordinarily hard and take it to the 97th percentile your influence is just not going to appreciably change. At some point you are excellent enough. Further effort in the subject is fine, but if your goal is to move the needle, it is effort best spent elsewhere.
The ability to write, the ability to speak, the willingness to make everyone your teacher and the breadth of your understanding about the world in my opinion are some of the other talents that can help increase ones scope of influence. Read, listen, be a lifelong learner. Seek to understand ideas, philosophies and other perspectives as a true student. Do that for 20 or 30 years.
Finally, insight often comes from other fields. If you have never been exposed to those ideas, you will never bring that flash of brilliance to the table.
Heinlein said it best:
“A human being should be able to change a diaper, plan an invasion, butcher a hog, conn a ship, design a building, write a sonnet, balance accounts, build a wall, set a bone, comfort the dying, take orders, give orders, cooperate, act alone, solve equations, analyze a new problem, pitch manure, program a computer, cook a tasty meal, fight efficiently, die gallantly. Specialization is for insects.”
Charisma is a dump stat for a lot of geeks. It doesn’t have to be.
You put a geek in a sales position, and tell them that their income depends on Making The Quota, and you have a miserable geek, who develops a life-long aversion to that skill set.
You put a geek in a sales position, and tell them “Your job is to listen to the customer, identify the problem, and help them solve the problem…” and you teach a geek the two skills needed to be an effective salesman:
Confidence. Talking to people is a skill and a learnable one.
Communicating your problem-solving capabilities and how knowing you helps benefit those around you. That eventually leads into the following:
You’ll have a broad enough base in solving interesting problems, problems that are important to other people, that eventually you’ll start spotting the problems that nobody else is seeing…and you’ll have the street cred to build teams to solve those problems.
You want to change the world?
Learn sales and marketing and sell your ability to solve problems. Eventually, you’ll be marketing your vision for choosing what problems need solving.
Most of the instances of RPatRT are “having a rep for solving this type of problem” and “being around when the discussion of a solution set is engendered.”
Here’s an amplification of something Eric stated:
Learn from mistakes, not just mistakes you made, but mistakes others make. You don’t have the time to make all the mistakes yourself. Likewise, never externalize your failures, and do not assign your failures to meddling from some outgroup.
Sometimes, things fail for factors that seem beyond your control. I say ‘seemingly’ because you’re usually letting your assumptions about whose back needs scratching and how hard blind you. Or, you’re applying sales when you should’ve been applying marketing, and the transition from the tactical (sales) to the strategic (marketing) is a fuzzy one that’s easy to miss.
Particularly if you’re one of those people who claims disdain for politics because they take away from doing ‘actual work’. As your ability to define what problems other people will work on increases, more and more, your job isn’t the ‘actual work’ – it’s getting people to do it.
>Charisma is a dump stat for a lot of geeks. It doesn’t have to be.
I’m going to amplify this from a different angle. Charisma is a learnable set of behaviors, as I described in Charisma: a how-to.
You don’t have to be particularly smart to learn them – a lot of charismatics are, in fact, quite stupid. You just have to be paying attention.
Pay attention.
> Here are a couple of possible theories:
Additionally I’ll toss out some not very well thought out ideas:
4. Voracious, insatiable, nonstop divergent thinking and imagination without fear of entertaining what in hindsight were obviously stupid ideas, coupled with the ability to distill quickly to the generative essence and recognize which ones are important to pursue.
5. Being dissatisfied with the status quo and accepted norms. Questioning everything. Rebellious. Yet also valuing humans and being social, as both a motivation and a source and interaction nexus for spawning and prioritizing ideas. Have any introverts positively changed the world? @Ken Burnside’s comment may apply here.
6. Not having tunnel vision and bogging yourself down in work that doesn’t foster those activities and attributes. Carpe diem. Time is wasting…
7. Not being too unbalanced on the side of sensing and logic, i.e. S on the Myers-Briggs personality type. Having a strong visceral intuition, yet well integrated with the prefrontal cortex. Interest in creative arts and athletics might be indicative.
8. An insatiable desire to understand how things work, not just machines but everything including human psychology.
9. Passionate about your vocations.
10. High energy level. I think Eric has mentioned before the importance of strong immune systems and other physiological attributes. Diet may be very important also. Vegans starve their body of nutrition. Refined sugar is destructive, and HFCS even more so. Eric blogged about that not too long ago.
>The insidious thing about this explanation is that it’s not actually false
Ah. I need to quote you here “all interesting human behavior is overdetermined”.
That means we can cherry pick the explanations we like, and we won’t even be entirely wrong because in overdetermined causation the explanations we picked CAN cause the effect, the problem is is that in practice other patterns contribute too.
I think this can explain the kind of “religious” debates that never get solved. Have overdetermined causation, have two factors that could each cause the effect on their own. One fits into your worldview, another in mine and we are both emotionally attached to it: the likely result is that we are going to argue absolutely forever about it because neither of us is actually wrong.
>That means we can cherry pick the explanations we like, and we won’t even be entirely wrong […] we are going to argue absolutely forever about it because neither of us is actually wrong.
Yes, I think this happens a lot.
@tz
I don’t see the old Man-Month problems. I see new ones, but my perspective is producing technology for internal use, not for sale, keep that in mind (i.e. not actually software firms). Back then it was understood that if you need something, some needs to actually write a program for it, which needs to be specified, developed, tested etc. But today we have large systems in place and the laypeople and managers don’t actually know their capabilities. They have no idea if a change they ask is a small configuration change or developing an entirely new feature from scratch.
And I guess we still did not get better at being extroverted and explaining things :)
As a result today it is not “we need X, please develop software for it, how many months?” it is more like just assuming our systems can do X because my buddy’s cousing worked in a place where they said their systems could do X and sure ours can too so they just casually say things like “hey, please set it up so that our salespeople cannot sell under a minimum price, in the campaign they start next week” (because our Chinese supplier apparently has a system like this) and then I am in panic thinking “oh no there is no such feature maybe I could code it up but not for next week and how exactly, what should I block, shipping or invoicing?” etc. etc. so I cannot give a yes or no answer, no, there is no such feature, yet it is not impossible to make one, and if one would be content with crappy untested undocument code maybe next week is doable at that point nontechnical manager loses patience “so can it be done or not?” So no one is asking for an estimate to do development, rather they no longer think development is necessary, we have magic systems where everything is a matter of a bit of configuration because everything that could be developed actually is.
This is not simply stupidity nor ignorance. Gigantic amounts of software features were developed since than. They are really not supposed to know the current feature set the current systems of the enterprise have. Moreover, a lot of things can be tweaked with a little scripting, which is a gray zone between “yes, we can do this, just a bit of configuration” and between “no we need a new feature developed for this”. I lean for the later, as even little scripting needs proper testing and documentation, but if you keep pulling those kinds of brakes while people insist and insist it is all to easy to give in and say “fuck I am an introvert, it is easier to code that crap up tonight than argue here with all these normies”.
I say it is a very different problem from what we had back then.
Back then there was a clean line between programming and everything else: if you need to recompile, it means it is programming. Today in the scripting age the difference between adding a new feature in Rails or turning on an existing one in a config file is small, so one can easily say fuck proper process and testing and all, and just do it outright if the normies insist so hard on this.
We don’t have this line of resistance anymore. Everything is just an easy script, everything can be tweaked therefore must be tweaked all the goddamn time. Not this discipline that this is V3.31 Build 123 and this has so and so features and others not.
@esr:
With a lot of hackers/hackeroids/geeks, I think it’s not so much that we don’t realize that CHA is a learnable set of behaviors, its that, however useful they might be for the advancement of this or that cause we would like to see advanced, there seems to be memetic contagion there that we don’t really understand, so we’re afraid to touch that skillset. We’ve seen too many Obama’s and Trump’s and bad managers and transparently manipulative advertisements not to wonder if CHA is a tainted stat. Now, we’ve all seen Reagans and ESRs and have all kinds of winsome friends and relatives that seem to have high CHA without being tainted, but we sort of wonder if we’ve just had the wool pulled over our eyes like the Trumpeters and Obamites. This probably has to do with the amount of sociopathic charismaticism that’s out there: we tend to see CHA as an indicator for sociopathy. I think it also has something to do with wondering whether we ourselves are sociopaths or close to it: I have the distinct impression that most supposed “anti-intellectual” purges are actually cold wars between factions of intellectuals that turn hot: whichever faction gets an army on its side exterminates the other.
> They have no idea if a change they ask is a small configuration change or developing an entirely new feature from scratch.
I’ve been the non-programmer asking for a change with no idea of how difficult it will be to implement, more times than I can count. On the other hand, at least I’m aware of my ignorance. On the gripping hand, there have been one or two occasions where the request had been to implement a CRITICAL! NEED TO HAVE!! FEATURE!!! and if the programmers need to flush the entire code base and rewrite the whole thing in INTERCAL to implement it, then by the gods, the right thing to do is to flush the entire code base and rewrite the whole thing in INTERCAL.
When I think about things I’ve managed to get done, I think the following strategy seems to have worked:
1. Be the person you want to be. Presumably, you have traits you want to dial up to 10. Do so. For as many dials as possible.
2. Watch for when other people start to ascribe a reputation to you. It will be for some of those traits.
3. Ease off the other dials and lean on the dials you’re noticed for, dialing them to 11.
Examples: I’m good at programming, games, mental math, piano, lashing, handwriting, graphite art, proofreading, sprinting, volleyball, picking out fine detail, mental proofs, headline writing, test taking, English vocabulary, about four Trivial Pursuit categories, plus maybe a few dozen other things. I recall hitting all those through college. I got *noticed* for a subset – programming, games, proofreading, and fine detail – so I’ve let the rest go, mostly.
The result is that I get known as the programmer, the gamer, the proofreader, and the guy you want to beat on something to find what will break. (I’d like more of a specialist’s rep, but I ain’t done yet.) When people see you as something, microproblems in those areas often solve themselves, because you fix them while barely trying, *and* you don’t have to prove it. (Some microproblems will even evaporate – you did nothing, everyone assumed you fixed it, and they turned out to be unimportant.)
I have a theory that most projects of significance have at least one person with a rep for taking care of large parts of it, but mostly just rolled through and focused almost wholly on only the Important Stuff.
…and again, they got away with that because of that rep.
I’ve read remarks here and there that regard GIFLIB as the thalidomide to Eric’s Food and Drug Administration. They think it’s a tangible but relatively small thing whose coattails Eric’s been riding ever since. I think they’re wrong.* I see GIFLIB as one of many feet in the door. I think it contributed to Eric getting a reputation, upon which he capitalized, whether consciously or not, to get more done, because people trusted that he could solve software problems. If Eric hadn’t made GIFLIB what it was, shrug. He likely woulda made something else, like a more secure Telnet or a more efficient logic gate or a better eggbeater or whatever, and would have been advertising that, and parleying that into open source cutlery or something and we’d all be commenting on Aproned and Dangerous.
*Mostly wrong. I agree that GIFLIB wasn’t *that* big a deal. I believe it didn’t make or break Netscape. If Eric hadn’t written (version 2 of) it, someone else likely would have. Might not have been as good, but none of this matters IMO as much as what Eric did with the fact that he built what he had built.
>They think [GIFLIB is] a tangible but relatively small thing whose coattails Eric’s been riding ever since
Yeah, but you know what’s funny about that? The relative timing. It didn’t become generally known that GIFLIB was what Andreesen had picked up until I was already famous for a bunch of other stuff. I myself didn’t have that fact until rather late in the game.
So these people, whoever they are, have latched on to a counterfactual in which I hung my cred on something that, in fact, nobody (including me) actually knew when I was becoming famous. Which is pretty amusing if you think about it.
It is indeed amusing.
To be fair, I doubt I could turn up those remarks again, meaning they’re held by vanishingly few people. And the people who do, might not be referring to the Andressen connection at all – they might only be referring to GIFLIB by itself (which gets around the counterfactual thing, almost – it doesn’t explain why it was good enough code that he used it in the first place). My memory might even be playing tricks on me, and conflating those remarks with mutterings about some relatively minor Unix contributions you might have made.
To all of this, I’m inclined to declare, “whatever – open source software would have been a good idea even if my dog had barfed it up this morning”. Good ideas aren’t about the person who had them – although you might want to look at that person for more.
>Good ideas aren’t about the person who had them
That is true. The flip side is, as I have often said, you haven’t won your battle until your crazy radical ideas are such conventional wisdom that they no longer have your name attached to them.
The final price of that kind of victory is that people will eventually forget what you actually did. I accepted that cost in 1998, with my eyes open, just as my peak fame was beginning.
>I accepted that cost in 1998, with my eyes open, just as my peak fame was beginning.
Even before, at the very beginning. At the Linux conference in Bavaria ’97 and the O’Reilly Perl gathering later that year, the first two times I gave the CATB talk. The stunned silence from the audience as they took it all in and knew their own power for the first time and realized everything they believed about themselves and their craft had changed forever. The wild cheering that followed…
…and as the cheering mounted, some part of me knowing that this must necessarily pass – that for me to win (and I was already pretty sure I could) it would be necessary for me to make my my centrality in that whole phase change at least half and perhaps entirely forgotten. I think I even said out loud in Bavaria (there’s a recording somewhere) something very much like “The day will come when all this seems obvious to everybody, and you will forget that it was me who said it, and on that day I will have won.”
It is a very strange feeling to be worshiped in the moment as a culture hero and at the same time know that if you allow that to continue for too long you will have sabotaged your purpose – that you can’t allow it to be about you, not for any longer than it takes you to make tactical use of the fame and get the mission done. Not quite like the slave at a Roman triumph whispering in a victorious general’s ear “Remember you are mortal”, but not entirely unlike that either.
UPDATE: Er, two of the first three times. I gave a very early version of the talk at a Philadelphia LUG meeting in, I think, late ’96. The ideas and the rhetoric weren’t fully developed at that time. I don’t remember anyone cheering.
> The final price of that kind of victory is that people will eventually forget what you actually did.
But for Satoshi, not soon enough.
[Y]ou haven’t won your battle until your crazy radical ideas are such conventional wisdom that they no longer have your name attached to them.
A close second-place consolation prize might be if others refer to your idea by name, but as if it weren’t a name.
An acquaintance of mine was once surprised to find that the “pareto” in “pareto optimal” was actually some guy (presumably, he thought it was some sort of Latin term).
Then there was the article a few years back, mentioning how David A. Huffman found out about some journal referring to “huffman coding”. No capitalization. (An example.) IIRC, he was rather amused.
> realized everything they believed about themselves and their craft had changed forever
Years downstream from the domino effect of that epiphany, I am hoping governance and cultural evolutionary strategies are the next to be disrupted by Linus’s law and the only know positive scaling law (decentralized paradigm) of groupwise coordination.
@Jon Brase, “think big or go home“.
Regarding right-place-right-time:
I think that part of it is having good intellectual taste, both in ideas and in thinkers; and part is being sufficiently extroverted and proactive to connect with the people doing the really interesting work.
Two common mistakes are working on what is hot right now, or working with the people who happen to be around you or that you already know. Instead, you want to work with the people who are working on what is going to be hot a few years from now.
To summarize: The right place is in the company of smart people working on cool ideas. The right time is a few years before those ideas really take off. You get there by learning to recognize smart people and cool ideas, and having the gumption to go work with/on them.
You can be charismatic without all the baggage inherint with niceties. You can simply be competent, and that works with anyone you’d prefer to do work with. You can even be a serious prick, as long as you’re not personal about the whole thing.
I often have to tell clients that what they want is feasible, but completely inappropriate to their business model, and there’s no way to do so without just being a prick about it.
Or there is, and I don’t know it. My findings are that if you’re competent, being an asshole means you avoid doing work for idiots, which is worth the lost income.
You don’t have to change your persona when working with clients (or potential clients), is my point.
>My findings are that if you’re competent, being an asshole means you avoid doing work for idiots, which is worth the lost income.
This works only if you not only have competence but can signal competence. All too many people (especially introverts) who have the former have trouble with the latter.
Regarding the OP, questioning your assumptions is certainly important, but like other commenters, I’d note that you must question everything, or at least whatever you can. Now, that doesn’t mean being a contrarian for contrarian’s sake, but stepping through the claimed facts and logic of whatever model you’re thinking about adopting very carefully. Often, it is a model that worked in prior times but doesn’t work well in the changed realities of today, say if you’re thinking about becoming a travel agent. ;)
Creativity is often just such clear-eyed thinking through of the details, where you bother to put all the pieces together, whereas others just assemble a mishmash of conventional wisdom in their head, without ever bothering to think about how all the pieces fit together, forget about whether each piece even makes sense anymore.
Regarding some stuff esr said,
>I think I even said out loud in Bavaria (there’s a recording somewhere) something very much like “The day will come when all this seems obvious to everybody, and you will forget that it was me who said it, and on that day I will have won.”
I wonder what you think is so obvious today, that all source will be open? Because I’d say we’re far away from that goal, with most source being closed even today. I ask because I wonder what you think is the endgame, that someday all source will be open or just some majority?
>It is a very strange feeling to be worshiped in the moment as a culture hero and at the same time know that if you allow that to continue for too long you will have sabotaged your purpose – that you can’t allow it to be about you, not for any longer than it takes you to make tactical use of the fame and get the mission done. Not quite like the slave at a Roman triumph whispering in a victorious general’s ear “Remember you are mortal”, but not entirely unlike that either.
I think you’ve done a good job of getting out of the way, whereas I wonder if for Stallman, much of the goal isn’t to be idolized by a cult of true believers. Of course, he could just be principled in believing all source must be free, but given the irrational and semi-theological nature of his claims, that would be tough to maintain.
>I wonder what you think is so obvious today, that all source will be open?
No, and I didn’t argue for that claim then or later. There are specific though unusual niches where closed source makes sense. I wrote about this in my original papers.
What seems obvious to most people today that didn’t then is that the lower error rates observed in open-source software are driven by the many-eyeballs effect. Even people who (correctly) observe that it’s not a panacea because sometimes you plain don’t get the many eyeballs where you need them generally concede that it works when you do.
In 1997, with the long shadow of the “The Mythical Man-Month” over us, that was a shockingly radical idea.
esr on 2017-03-09 at 07:53:55 said:
> > is a dump stat for a lot of geeks. It doesn’t have to be.
> I’m going to amplify this from a different angle. Charisma is a learnable set of behaviors,
> as I described in Charisma: a how-to.
In an interesting bit of synchronicity Seth Godin says:
> People who are open, empathetic, optimistic, flexible, generous, warm, connected,
> creative and interesting seem to have a much easier time. They’re more able to
> accomplish their goals, influence others and most of all, hang out with the people they’d
> like to be with.
> The best part is that this is a skill, something we can work on if we care enough.
http://sethgodin.typepad.com/seths_blog/2017/03/emotionally-attractive.html
>The best part is that this is a skill, something we can work on if we care enough.
Hmph. That post is a useless tease; Godin stops there without saying anything about how. I’d never do that to my readers.
> It’s not enough to be rigorous about questioning your assumptions once you’ve identified them. The subtler work is noticing you have them.
I have printed out and hanging side-by-side on my cubicle wall at the office two signs: “Numquam ponenda pluralitas sine necessitate ~William of Ockham” and “ASS|U|ME”. The latter is there to remind me to be aware of my assumptions, and to challenge them as soon as I become so aware.
As an added bonus, I find myself challenging others’ assumptions, which does wonders for my popularity, let me tell you.
(I doubt any readers of this blog would question why Ockham’s Razor is there.)
>(I doubt any readers of this blog would question why Ockham’s Razor is there.)
The owner of the premises certainly doesn’t.
esr:
That’s repetitive, don’t you don’t you think? :-P
You certainly do, but a bit o’ proofreading would do ya good. Just sayin’. :-)
Hey, you gotta give him some credit for making the statue of the guy who’s thinking. No, wait – that’s Rodin. Never mind, then.
@ESR:
Hmph. That post is a useless tease; Godin stops there without saying anything about how. I’d never do that to my readers.
Most of the stuff he posts to his blog is motivational in nature (he posts pretty much every day) not “howto”. This is the internet, if you can’t use a search engine then you either need to work on that, or you’re pretty much not going anywhere anyway,
>Most of the stuff he posts to his blog is motivational in nature
Yeah? What the fsck is “motivational” about being told you will do better if you are more “emotionally attractive” if you don’t have a clue how to get there and the person making the claim just leaves it hanging in the air without actionable advice? That doesn’t seem “motivational” to me, it seems clueless or cruel.
I’d never do that to my readers, and I have zero respect for anyone who does it to his.
> Yeah? What the fsck is “motivational” about being told you will do better if you are more “emotionally attractive” if you don’t have a clue how to get there and the person making the claim just leaves it hanging in the air without actionable advice?
My theory is that the target audience knows, or is assumed to know, how to be emotionally attractive but tends to forget the magnitude of the importance of being emotionally attractive. To hackers it sounds like “Be more emotionally attractive. It’s a Simple Matter of Programming”. Then again, you’re more extraverted than the typical hacker or geek, so you’re probably more competent to know if my theory holds water than I am, so your “what the fsck” probably indicates that it doesn’t (on the third hand, I’m not sure I’m really so much introverted as socially underconfident).
On a side note, would “chkdsk” be a euphemism for “fsck”, or even stronger language? What about “chkduk”?
“Windows found problems with the filesystem. Quack!”
>your “what the fsck” probably indicates that it doesn’t
There should probably be a word for advice or instructions phrased so that the likelihood a person will process it correctly is inversely proportional to their need of it. “Unhelpful” is not strong enough.
esr:
Paul said that “[g]ood ideas aren’t about the person who had them” and you agreed, eventually stating that “you can’t allow it to be about you”. You were discussing the open-source cause; do you feel the same way about your other causes? For example, would it be okay if people started using your pro-gun arguments without giving you proper credit?
>For example, would it be okay if people started using your pro-gun arguments without giving you proper credit?
Absolutely.
>you agreed, eventually stating that “you can’t allow it to be about you”
I’ve been thinking about this, and I now consider that it depends on what scale of change you’re interested in.
Let’s say. for example, that you’re an architect with a radical new idea about how to structure-frame buildings. There are at least two strategies you might choose: one is to couple the idea closely to your personal brand, the other is to try to turn it into part of the toolkit that every other architect uses.
Which you choose depends on your objectives. If what you want is to be famous and do a lot of lucrative commissions, then trying to keep the new idea tied to your personal brand makes sense. There’s nothing illegitimate or wrong about that.
On the other hand, if your overall goal is to minimize building collapses and you think your method accomplishes that, then you don’t need the ego conflict other architects will have with using an idea tied to you to be getting in the way of that.
> There should probably be a word for advice or instructions phrased so that the likelihood a person will process it correctly is inversely proportional to their need of it. “Unhelpful” is not strong enough.
dissonantics
esr:
Thanks. I asked because I had imagined the (unlikely) scenario of my discussing the issue of guns with someone in meatspace and thought that it might be persuasive to quote your dictum, “…when it’s a crime to own guns, only criminals own them…” (from the Libertarianism FAQ); but I wouldn’t want to lead anyone to know you and eventually visit A&D, because I’ve repeatedly embarrassed myself here. :$
How do you feel about, say, the inclusion of Sid Meier’s name in the titles of his games? (Shenpen once made comment that seems relevant. “Raymond’s Fetchmail”… heh, heh.)
>How do you feel about, say, the inclusion of Sid Meier’s name in the titles of his games?
I don’t have any problem with it. I’m a fan of his games, so the name is useful information.
“…when it’s a crime to own guns, only criminals own them…”
That exact wording might be Eric’s, but it’s known better by far as “when guns are outlawed, only outlaws will have guns”, which is as much in the public domain as “the cat is on the mat”.
As for linking sites: I think the Libertarianism FAQ is far from the best source for gun control discussion (it’s only got the one paragraph, after all). Ethics from the Barrel of a Gun is better, and has insights you’re unlikely to find elsewhere, but if you want something more complete, you’re probably going to need a combination of GunCite and the bolus of things available from NRA.
Which is kind of a shame, and when you think about it, it’s another example of how name associations can short-sheet you. NRA-ILA is the 800-pound gorilla of lobbying groups, and if that puts you off, and you’re new to the gun debate, you might be tempted to assume there’s something sketchy about the gun rights position.
I’d always been of the impression that those games were namedSid Meier’s Civilization in large part to avoid naming conflicts with other games, such as Avalon Hill’s Advanced Civilization. (It’s certainly partly a branding thing as well; cf. Sid Meier’s Alpha Centauri.)
> I’d always been of the impression that those games were named Sid Meier’s Civilization in large part to avoid naming conflicts with other games
Of course, we do this in hackerdom too: “GOSMACS”, “K&R C”, “CLC-INTERCAL”… authors’ names are a handy disambiguator, all the more so for multiple implementations of a common standard. Heck, “Linux”.
Then again, there are cases of perhaps more ego-driven namings, such as “reiserfs”. But they’re rare and often divisive.
OTGH, I wonder what bucket “awk” belongs in?
@Edward Cree –
> OTGH, I wonder what bucket “awk” belongs in?
Well, as you are probably aware, the program’s name is an acronym of the authors’ names – Alfred Aho, Peter Weinberger, and Brian Kernighan. “Ego-driven”? Don’t know, but would bet against.
Personally, I always thought it was an abbreviation for awkward, on account of the syntax. YMMV….
Paul Brinkley:
Thanks, I didn’t know that. But that one seems problematic regarding its translation to Spanish, so I prefer Eric’s version.
The Wikipedia article about Meier tells a different story (bear in mind that Sid Meier’s Pirates! predates Civ), but it was certainly wise to signal that it was a different game.
esr:
I loved Civilization II. It was my “gateway drug” into the strategy genre, at the age of nine. I discovered it thanks to a friend of my father’s. Amazingly, that person has several things in common with your wife: both are female lawyers who wear glasses, like cats, and used to play Civ2. (I must have put it together some time after this came out.) Small world indeed.
It’s surprising because I came to know those two similar ladies in very different contexts and nearly two decades apart. I guess Heinlein was right: “To be ‘matter of fact’ about the world is to blunder into fantasy – and dull fantasy at that, as the real world is strange and wonderful.” :-)
>I loved Civilization II.
So did I. It’s one of only two proprietary computer games I ever cared about enough to own a copy of. (The other one is Spaceward Ho!, the only Android app I’ve ever paid money for.)
On a different note from my previous comments on this post, changing the world seems like hard and tiring work, and I don’t really want to do it, but there’s change that I think needs to happen that nobody else seems to be pushing, and I need advice on how to actually do it. A summary of my thoughts on the matter follows:
I subscribe to what you have called the “libertarian critique” of IP law, which you do not subscribe to but have summarized as the belief “that so-called ‘intellectual property’ rights are illegitimate because they grant state-enforced monopolies that would neither arise nor be defensible as natural property rights in a free market.”
Furthermore, I hold that IP law is actively corrosive to rights in physical property, especially in a computer-heavy society like our own (where DRM is a tempting means for IP holders to enforce their supposed rights), and that our society is at risk of collapse / descent into dystopia on a one-to-two decade timescale as long as using IP law to prop up sale-by-the-copy remains the primary means of monetizing information production.
The preferable way to deal with this would be to abolish IP law, but this would likely require more time than we have and a degree of vigrorous activism that would be infeasible, maybe even societally dangerous, in the current American politcal climate.
So I think the best way to shift the balance of open vs. proprietary information is to help the information industries find ways so monetize the information they produce aside from sale-by-the-copy. The most direct way to monetize information production without sale-by-the-copy is by rattling a tin cup. This, as I see it, is the primary importance of your Patreon drive: the particular software you are working on isn’t so important as the concept of funding open source development by micropatronage. In a comment on “In which, alas, I must rattle a tin cup”, you said:
Surely you would find it more distasteful to take a salary from $PROPRIETARY_VENDOR to develop a proprietary NTPsec equivalent, or to work for the US government and have your salary for developing NTPsec come from tax money? Take heart! You’re doing what Microsoft, Oracle, Apple, Hollywood, Houghton Mifflin, et al. will eventually have to be doing if we’re to avoid dystopia, and you’re among the pioneers. That’s not cause for humiliation.
So the question, then, is how to get the information industries to try adopting micropatronage? My idea is to approach vendors with legacy hardware and software of historical/retrocomputing interest, and to propose that they crowdfund open-source releases of their legacy products. Free access to source code and hardware designs for such systems (or even the ability to directly reverse engineer without clean-room techniques in cases where there is no surviving source) would be of great help to efforts to emulate those systems, so by doing this, companies could gain good PR and make money from a product that was no longer making them money, without the risk of crowdfunding entirely new code. Low-risk releases of legacy code would help get them used to the idea of crowdfunding software, and, if their crowdfunding campaigns went well, they could hopefully be convinced to crowdfund the release of more and more recent products, and eventually to experiment with releasing a new product funded entirely by crowdfunding. Once this had been done in the computing industry, hopefully it could be extended to other information industries like Hollywood and the publishing industry as well.
The most obvious candidates to approach first would be the vendors for the early PC platform: Microsoft, IBM, associated component vendors, etc. A large portion of current geekdom grew up on this platform and has a fair amount of nostalgia for the games of their childhood, so public response if one of these companies were to announce a crowdfunding campaign as described above would likely be fairly enthusiastic. On the other hand, the PC platform is fairly well studied and the quality of projects like DOSBox and FreeDOS is already fairly good, so maybe a different micro platform would be better to target.
Given that you have prior experience advocating open source to corporations, I’d like to ask you for advice on how to implement this. That said, given said experience, your better people skills, and the fact that you are somewhat better known than I am, you’d probably do a better job than me if you were to take it on. However, my impression is that your evaluation of the threat IP rights pose to actual property rights is less dire than mine, and you certainly have quite a bit on your plate already. You’re enough better suited to the task that I have to be bold enough to ask if you’d take it on, but I’m certain enough that your answer would be “no” that I have to be ready to take it on myself.
So, the advice I’d like:
If your evaluation of the threat of IP were the same as mine, is this how you’d tackle the problem? Is my strategy, as outlined above, likely to be effective? What company or set of companies would you approach first? How would you go about getting their attention? Do you think the FOSS community would respond positively if, for example, Microsoft were to offer to release the source code to DOS 6 or Win 3.1 to the public under something like the BSD license in exchange for $DOLLAR_AMOUNT in donations, or would groundwork need to be done to warm the community up to the idea? What do you see as the primary obstacles to something like this? Has anything like this ever been attempted before? If so, presumably the results weren’t what was hoped for, or the project was insufficiently ambitious in scope, or it already would have changed the world. So, if this has been done before, what failed?
And with that, this comment has been over a week and three rewrites in the making, so I’ve got to force myself to finish and ship.
>Is my strategy, as outlined above, likely to be effective?
Well…I don’t think it’s doomed to fail, but it will take a lot of work to implement effectively. Plan on giving it at least five years of your life; that’s what I had to do.
>What company or set of companies would you approach first?
Novell. I’d try to get them to release the heritage Unix sources.
>Do you think the FOSS community would respond positively if, for example, Microsoft were to offer to release the source code to DOS 6 or Win 3.1 to the public under something like the BSD license in exchange for $DOLLAR_AMOUNT in donations, or would groundwork need to be done to warm the community up to the idea?
I don’t think I know enough about the potential level of interest in DOS or 3.1 to say.
>What do you see as the primary obstacles to something like this?
Territorial inertia on the part of the rights-holders. In situations like this they have a tendency to way overestimate the present value of the asset.
>Has anything like this ever been attempted before?
Not that I know of.
@Jon Brase
> the belief “that so-called ‘intellectual property’ rights are illegitimate because they grant state-enforced monopolies that would neither arise nor be defensible as natural property rights in a free market.”
Afaik, many or most of the indie producers understand it is counter productive to enforce IP law on your supporters (since the entire point is build popularity and not destroy your PR). The threat of IP law comes from rent seeking middle men who abuse both the producers and the consumers. Direct purchases from indie artists are ostensibly in actuality a gift culture with the benefits to the giver of receiving in return a token of pride in the reputation, such as a customized Tshirt with the indie’s imagery. I noted Eric is featuring the names of his top donors in the credits section of his works.
> that our society is at risk of collapse / descent into dystopia on a one-to-two decade timescale as long as using IP law to prop up sale-by-the-copy remains the primary means of monetizing information production
IMO, you’re erecting a strawman by fighting dinosaurs who are on their deathbed, along with downstream dependents of the rent seekers in (afaik particularly entrenched in American) society.
Apparently it is the capitalistic funding model which is the difference between commercialization of an economic revolution and the failure of for example France to win the industrial revolution. I’m looking into changing the world by decentralizing funding models with a blockchain. They key insight of my Rise of Knowledge, Demise of Finance is that non-fungible production in the Internet era isn’t amenable to aggregated finance (e.g. loans), Theory of the Firm substitutability, nor rent seeking distribution funnels. All centralized databases much die. Thus, Github must die. StackExchange must die. Facebook must die. Google Playstore must die. Etc..