I have an Android phone, and its name is “moogly”.

I have acquired a Googlephone — a brand-new T-Mobile G1 to replace my eight year old and on-its-last-legs Sprint phone.

I’d say I had to get one of these because it’s got some of my software in it, but as a one-time maintainer of GIFLIB (not to mention named contributor to libpng) just about every cellphone has some of my software in it. (And every browser. And the Microsoft X-box. I am become Shiva, destroyer of worlds ubiquitous and omnipresent.) No, this one is special because it’s got Linux inside. And open source is part of the branding, though the look of polite incomprehension on the face of the perky blonde T-mobile salesbeing I dealt with suggests that there is no danger of imminent mass enlightenment from this campaign, alas.

First impressions: Pretty good, stacked up against the iPhone better than I expected. Well-designed user interface, I was flying through it within three hours of familiarization and think less technical users would too.

Hardware: The keyboard, predictably, sucks — zero-travel chiclet keys. Display very nice, crisp 320×480 with good luminance contrasts. Charges from any USB port — what a cool idea, I might never actually use a dedicated charger again. Speakers suck too but no blame attaches; in an enclosure this size, non-suck is not a physical possibility. Wider and longer than the decrepit Samsung 660 it’s replacing but thinner; it fits comfortably in a pocket. The trackball works nicely but worries me a bit because it feels fragile, a potential failure point.

Interface bugs: In the Favorites editing screen, the entry for “Name” stops accepting characters well before you get to the end of the box — screwed-up font metrics, or a silly length limit, or both. You can choose from a palette of icons for your five favorites, but contacts outside that group have to be iconfied with a picture. Why is that?

Other bugs: The OS apparently crashed and rebooted once in my first four hours of operation. Hasn’t done so since.

Omissions: Where is my fetch-ringtone-from URL function? Where is my fetch-wallpaper-from URL function? I wanted to make my default ringtone the Star Trek communicator sound; had the MP3, needed to figure out how to get it onto the phone’s ringtone list.

This turns out to be easy by a slightly different route. Plug the G1 into your Linux system; it (or rather the SD card in it) will present as a removable disk with a subdirectory. Drop your sound file in there; open your music player; click Menu and select “Set as Ringtone”. Voila! Oddly, you may have to unplug the USB cable to make the music file visible in the phone.

Setting wallpapers isn’t too tricky either. The SD will have a directory called dcim/Camera; drop the image in there, and it will become visible as a Picture under Settings->Wallpaper. My G1 now shows a nice astronomical photograph of M33. 640×480 images work nicely as all it has to do is a 2:1 scaledown in the Y direction; to save space on your SD card you may want to pre-shrink them.

Finally, a cellphone I can hack! Just superficial stuff, so far, of course, but the relative ease with which I sussed all this out within a few hours of acquiring the device is very promising. I like it. And apparently a lot of other people do, too; we had to canvas three T-Mobile stores to get one, they’ve apparently been selling like crazy

Overall: No, this isn’t quite as polished as the iPhone and lacks the cool multitouch gestures. But it’s seriously cheaper, almost as good already from a purely functional perspective, and the open-source stack will mean it gets better fast and will add value from third-party apps at a rate Apple’s walled garden won’t be able to match. Especially given that multiple cellphone vendors will be shipping Android phones; this means a broad-based, stable market sure to attract lots of developers and service providers.

Even the first-out-the-gate G1 seems perfectly designed and positioned to disrupt the iPhone’s market from below. If successors feature hardware with an even slightly slicker presentation, Apple better watch its ass.

Bonus karma points to the first commenter to correctly explain why, when I figure out how to telnet/ssh into the underlying Linux, the only possible thing for me to change the hostname to will be “moogly”.

UPDATE: and here’s an astronomical wallpaper I made which happens to fit the default placement of the clock and icons perfectly; NGC7331 gets the unoccupied center of the display.

64 thoughts on “I have an Android phone, and its name is “moogly”.

  1. Great Googly Moogly, Eric, I would have thought that was obvious!

  2. Cathy gave the game away…but, FWIW, the phrase brings to mind the Temptations Ball of Confusion for me.

    It’ll be interesting to see how well the Android platform really does take off. This is really the first time open source has gone up against closed source in a market where other factors don’t greatly control buying decisions. (I don’t think folks’ preference, or not, for AT&T as a cellular carrier makes all that much difference in the uptake of the iPhone.)

  3. >Cathy gave the game away

    No she didn’t. I’m waiting to see who first identifies the phrase we’re referring to, and its author.

    Update. Hmmm…your usage may predate the one I had in mind.

  4. Well, I was thinking that you thought it was ugly, uggaly, moogly. I still blame Zappa, though.

    So, who are you going to vote for? Obamccain or Barr? There are only two choices, of course. Every human decision must be brought down to two choices for it to make sense. Up/down, left/right, in/out, forward/backward, positive/negative, old/new, demopublican vs libertarian.

  5. Russ wins the prize. “Great googly-moogly!” Frank Zappa, Overnight Sensation, 1974. (Full name of the album: “Apostrophe’/Overnight Sensation”)

  6. Interface bugs: In the Favorites editing screen, the entry for “Name” stops accepting characters well before you get to the end of the box — screwed-up font metrics, or a silly length limit, or both. You can choose from a palette of icons for your five favorites, but contacts outside that group have to be iconfied with a picture. Why is that?

    Other bugs: The OS apparently crashed and rebooted once in my first four hours of operation. Hasn’t done so since.

    Things like this should be showstoppers. Yes, even the text field thing. They may seem minor, but to your average cellphone customer they make the interface look amateurish and unfinished. With an attitude like this, the “aw, let them live with the bugs, they do no harm and we’ll fix them later” attitude (endemic to open source), Android poses little challenge to the iPhone.

  7. Russ wins the prize. “Great googly-moogly!” Frank Zappa, Overnight Sensation, 1974.

    See, I knew it was from “great googly-moogly” but I’m not up on my Zappa. I should have guessed though; Weird Al used the phrase in his Zappa tribute “Genius in France”.

  8. > when I figure out how to telnet/ssh into the underlying Linux,

    something tells me you’re going to be dissapointed to come to learn that Android runs a linux kernel (the part that rms would refer to as ‘linux’), but the rest of android from there on up is a virtual machine (Dalvik) over a very thin (libc, but non glibc) layer.

    Typically apps for Android are written in java, with the java bytecode files translated to ‘dex’ file (with ‘dx’).

    You *could* drop a sshd onto the phone, of course, but I don’t think you would see much once you get a shell for it to execute. (This based on a quick peer around the source tree.)

    Since its “open source”, yes, you could eventually craft enough userland to make a nice little nest for yourself, and yes, the sethostname() system call is there (and even in the libc) but… to what end?

  9. Got mine, too. What is this “Favorites editing screen” of which you speak? I can certainly see that one of the four tabs in the “Contacts” app is labeled “Favorites”, but I can’t find any obvious way to invoke an “edit” command from there.

    How did you determine that the display was 320×480? That’s good to know, since I can pre-shrink wallpaper images.

  10. Personally I’ve been a little disaapointed in how slowly apps have appeared for the GNU/Linux-based Nokia N8x0 devices. It contains a pretty complete version of Linux (people have also gotten it to run KDE, Debian, Android, etc.) OpenSSH installs easily from its package manager. (It also has apt at the xterminal.) There are still no decent PIMs for it that don’t have some kind of bug. Being a phone hopefully you’ll get much better market penetration and more developers.

  11. I am torn. On the one hand, I desperately want this phone (and being an early adopter addict is only part of it). On the other hand, my phone is an essential part of my communication with the world outside–a great percentage of my interactions through email, txt, IM, and phone never near an actual PC–and I am loathe to jump in without knowing it will work for me.

    Posts like this make it all the more tempting though… keep us up to date! :)

  12. I’d like to get my hands on an Android. Even though it’s open source, Google seems to actually design their programs for users to use, instead of just trying to please themselves.

    I’ve just finished reading The Design of Everyday Things, (required reading for designers and hackers!) and I think the problem with FOSS is the tendency for designers to design for themselves rather than for users. Open source is always touted as a boon for developers instead of users. I wish more hackers would abandon configuration fetishism and design around user’s needs. Google and Canonical seem to be taking a step in the right direction.

  13. >What is this “Favorites editing screen” of which you speak?

    Home -> myFaves -> Any of the plus icons.

    >How did you determine that the display was 320×480?

    Googling led me to some user forums where this issue was being discussed.

  14. >>How did you determine that the display was 320×480?
    >
    > Googling led me to some user forums where this issue was being discussed.

    Android only supports HVGA screens (currently).

  15. Jim T:
    Typically apps for Android are written in java, with the java bytecode files translated to ‘dex’ file (with ‘dx’).

    Well, that’ll hold up the Wesnoth port. (Actually, I looked very briefly at that several years ago for the Nokia N770, and found the memory footprint was a lot bigger than I was expecting before I punted. Never did figure out if that was for real or e.g., unnecessary duplication of images that was irrelevent on desktop platforms…)

  16. > Well, that’ll hold up the Wesnoth port.

    The world can’t wait, time for the re-write in Java.

    > [footprint]

    Just load the images (on-demand) from a remote server.

  17. Just load the images (on-demand) from a remote server.

    Lawl. Don’t have enough CPU juice or storage? Just offload it to the cloud.

  18. >> Well, that’ll hold up the Wesnoth port.

    >The world can’t wait, time for the re-write in Java.

    Java? How about Python + Pygame?

  19. > Will the phone run Python?

    Only once someone makes a python port that will run over the top of the Dalvik vm.

    I think it likely that someone (likely @ google) might have that working already. Guido works there, no?

  20. Jython won’t run because Dalvik won’t yet accept dynamically generated bytecodes; you can put busybox on the G1, and run those using the adb debug shell over USB, but I still haven’t found an obvious way to a) get root, b) re-flash the kernel.

  21. Hm. I suppose you could knock out a basic shell in java pretty quickly, but I’m not sure what you’ld see from there (aren’t android apps in some kind of sandbox?)

  22. Do they offer a refund for the phone if it’s rendered unworkable by forcibly removing critical pachages?

  23. if a phone was just a phone, it would not take pictures, it would not be a schedule, a phonebook, a handheld console, an alarm clock, a clock, a currency converter, a primitive chat device… I don’t think this is Linux’s fault.

    If someone wants to hack away at his phone, all the best to him.

  24. I didn’t buy a cellphone until 2006. When I did, I got the most basic model they had. It still has Java installed, but I never use it for much of anything besides as a phone (and perhaps an alarm clock). I much prefer to use computers as computers, and handheld consoles as handheld consoles (I <3 my DS).

    But I speak only for me, not for the gadget-hungry throngs out there (which, given the Palinization of America, increasingly amount to passive consumers of technology developed in China, India, and Europe).

  25. Anyway. To get ontopic. What the hell is the reason behind this boom in smartphones, mobile apps, and mobile web browsing? Exactly why would you want to do anything on a screen the size of a box of cigarettes? Exactly when don’t you have a laptop within 3 yards of you? Sigh. The first rule of usability is the physical constraints and comfort of the human body. Keys the size of the fingertips and NOT smaller, fonts the size you can comfortably read and NOT smaller, at least 50 chars per line so you don’t have to scroll for every tenth word etc. Why would you want to do anything on a device smaller than a laptop? (A small laptop.) Is it that hard to carry an ultra-portable around?

  26. > What the hell is the reason behind this boom in smartphones

    Answer: Moore’s law.

    Your desktop is probably “fast enough” if you’re in the 90% / first two std deviations. That market is all but saturated with the exclusion of replacement.

    But Moore’s law marches on, so getting sufficient MIPS in your hand to run a *nix variant complete with TFT, touch-screen, various radios (bluetooth, 802.11, cellular), all implemented in CMOS (rather that discrete designs or designs that require GaS or ECL) has recently become “quite possible”, and selling sand for O($100) is a great business IFF you’ve got a market for millions of the resulting “chips”.

    Thus, cellphones that run browsers over the top of a *nix kernel. The software is, to be kind, “last generation”, at best, but it enables a familiar “product”.

    As for those who insist on ‘breaking root’ on every

  27. crap, managed to ‘click’ the mousepad with the mouse pos exactly wrong…

    Anyway…

    As for those who insist on ‘breaking root’ on every bit of hardware they own, my take is that they either have nothing better to do, or don’t know how to do actual interesting work on the device. Certainly there are interesting things to do with an Android phone or an iPhone, and those who liberated the iPhone may have a point that they don’t want to have to “go through” Apple for access to the API (now resolved) or live with the restrictions of the API (not yet solved) for a device they own.

    Anyone who spends inordinate time trying to get a shell prompt on an Android phone is likely just taking a break from his 20x/day fapping habit.

  28. I guess I grok what you mean – but that’s not even new. I remember installing Linux onto a HP palmtop – don’t remember which model, it bricked long ago, I think it started with 76 – and amusing myself with running mc on it over ssh over wifi from another computer that booted Linux from a thumbdrive just for the “live the bleeding edge” kicks out of it, but when was that? I don’t quite remember but at least three years ago. Maybe four or five. (And that was exactly the time when I realized that for short-term fun it’s OK but it’s pretty useless, due to the utter inconveniece of using the tiny screen and keyboard.)

  29. ObPedantic: You guess what you grok. Grokking implies a full understanding of the subject. :)

  30. … and obviously, I had to make a misteak there. You _don’t_ guess that you grok something, is what I meant.

  31. @shenpen an iPAQ running Linux useless? Au contraire! I used mine with the Navman GPS sleeve, and some Python GPS software of my own design, and I was able to display moving maps, keep a GPS track, take waypoints, and … record georeferenced voice clips. Even the n810 can’t do that right now.

  32. > Even the n810 can’t do that right now.

    but the hardware is all there, so its just software. What you mention doesn’t sound like a big add to Maemomapper (doesn’t do the georeferences voice clips)

    Everything else you mention has been available since just after the 770 was covered in Linux Journal by Doc and I. Years ago.

  33. @JimT As long as you’re taking your meds, I’ll respond to you. Yes, the ITT has the capability. Shenpen was asking why anyone would bother. My answer is, as always, pocketability. All the keys, all the pixels, and all the mips in the world do you no good unless you have your computer with you. Who is going to bring something with thm everywhere if they have to carry it?

    Once you decide on pocketability, you have a small screen, small hard keyboard, and small barrery, yes, which constrict usability. Oh well. They’re compromises I’ll make for th computer that I have in my pocket.

  34. >And about ESR, I am surprised that someone as smart as him is a Wiccan. Usually highly smart guys like him who reject that Christian faith are atheists, as least the ones I have encountered.

    Well, in the strict philosophical sense I am an atheist – I don’t believe anything resembling the omniopotent/omniscient/omnibenevolent God of the Abrahamic religions exists. In fact I think it is necessarily true that such a god cannot exist — the traits humans ascribe to it are self-contradictory.

    Almost no Wiccans believe in such a god (there are some sort-of exceptions; the topic is complicated). When David Brooks writes, in the article Shenpen cited, “That’s bound to lead to new movements that emphasize self-transcendence but put little stock in divine law or revelation.” he could be talking about us. Well, “self-transcendence” is a misleading way to label it….but this is too large a topic for a comment. Perhaps I shall blog about it.

  35. Re: actual google phone info

    A google developer claims you *can* write native apps, though they’re concentrating on JVM stuff for simplicity for now:

    link
    Actually, nothing requires you to write code in the Dalvik VM. That’s just
    what’s published at the moment for compatability. I was at a tech talk the
    other day where it was explained that a native interface extension for
    Dalvik is coming out (so you can code against libc and libm at least. Have
    to provide other libraries yourself), and that doing fully native
    applications is possible, but not the focus right now.

    All the applications on the G1 already run as their own processes with
    permissions enforcement done by the kernel, so the security model is
    already setup to handle this.

  36. I have deleted a large number of contentious and off-topic comments from this thread. In future, I will be much stricter about this. I blog about politics and religion often enough that those of you who want to argue such things can do so in the threads hanging from those postings.

  37. >If Google’s phone locks any user into anything, then it is not open.

    (Transplanted from here.)

    I’m not aware of any actual lock-in in the G1. The entire software stack is open, and it’s not even clear that Google intends to close the root hole; I’ll be surprised if they do, actually, as I can’t think of anything in their business strategy around Android that would require it.

    In any case, the right question about the G1 is not “Is Google angelically perfect?” but “Does the market presence of the G1 tend to pull the world in an open-source direction?” I think the answer to this is very clear; Android has Sprint and the rest of the lockin-mad mobile phone vendors shit-scared and scrambling to adapt. The smart ones will join Android rather than trying to beat it.

  38. I’m not aware of any actual lock-in in the G1. The entire software stack is open, and it’s not even clear that Google intends to close the root hole; I’ll be surprised if they do, actually, as I can’t think of anything in their business strategy around Android that would require it.

    Open: have you replaced the kernel you’re running yet? If not, you might inquire as to why nobody else has, either.

    As for R30 (which fixes the root hole), it started shipping 11/9, and you too will get this ‘fix’ OTA
    http://news.cnet.com/8301-1009_3-10092407-83.html?tag=mncol;txt

    Or just google for Android and R30.

  39. > Sprint and the rest of the closed-source mentality carriers

    Yes, they’ll adapt to Android, if only to compete with the iPhone (and other Android phones). Note explicitly that counter to your argument, (in the other thread) they didn’t do this in order to avail themselves of the open source development model. In fact, the ASL v2 allows them to take every advantage while contributing nothing back.

    ‘lockin-mad’: Your T-Mobile G1 is still sim-locked. Can you install an AT&T (say, pre-paid) SIM and get service?
    Not unless you’ve waited 90 days from purchase (provided you’re a customer “in good standing”), and even then, you’ll still be under contract with t-mobile. The other option is to have purchased the phone for $399, sans any contract, and even then, you’re likely to have to wait 90 days to get it unlocked.

    And note that Android licensing (via the APL v2) explicitly allows the carriers to modify the phones without giving the source code to the changes.

  40. > it’s not even clear that Google intends to close the root hole

    They’ve already fixed this particular one, and rightly so, because it’s a horrible security issue. But you seem correct in asserting that they have no vested interest in preventing users from getting root on their own phone.

  41. >If not, you might inquire as to why nobody else has, either.

    Er…possibly because we don’t need to yet?

    >Note explicitly that counter to your argument, (in the other thread) they didn’t do this in order to avail themselves of the open source development model.

    T-Mobile did. They have a justified expectation that the openness of the stack will grow their business, which is exactly the argument for open-source hardware drivers I’ve been making for ten years. As for the others – their fear comes from their justified suspicion that T-mobile is right. It’s open-source winnage all the way.

    And if other phone vendors are going open out of fear rather than altruism or respect, I’ll take that. I’ve often said that an open-source evangelist’s worst enemy is his own idealism and his best allies are fear, greed, and vanity. What’s happening now as the iPhone pushes other vendors into the arms of Android illustrates this maxim perfectly. Their reasons don’t matter a hill of beans compared to their behavior.

    At the end of the battle, all cellphones will be as open as the G1 , forced there by competitive pressure. They want to co-opt us, but we will have co-opted them.

    (Eric now feels like kicking up his heels and cackling “Bwahahaha! My sinister master plan for world domination, it is working!”)

  42. T-Mobile did. They have a justified expectation that the openness of the stack will grow their business, which is exactly the argument for open-source hardware drivers I’ve been making for ten years.

    Citation, please. (For T-Mobile’s position, not yours.)

    (Eric now feels like kicking up his heels and cackling “Bwahahaha! My sinister master plan for world domination, it is working!”)

    I suggest being careful while you’re ankles are in the air.

    What you’ve got with Android is an unlimited ability to write programs for the device (as long as you like writing in Java, at least for now), some decent hardware, and a behemoth corporation (Google) backing the thing. Honestly, its not that different from the situation with the iPhone, except Google/T-Mobile haven’t constrained access to the hardware in the same way that Apple has with the iPhone SDK, and its easier to distribute an application for Android than it is for iPhone. Apple has certainly shown its ass in these regards.

    However, for reasons unfathomable, you seem convinced that the phone in your hand
    is truly open (in the same way that say, the Open Moko / Open Runner are.)

    It isn’t.

    The most interesting development for Android (from my POV, anyway) would be a release of Android for the iPhone. Who wants to bet that there aren’t a small group of Google engineers beavering away their 20% against this goal? The mere existence of such a port would be good for a drop of 10% in AAPL’s share price, if not more.

    The far more likely development for Android is its use in other consumer electronics, and even small ‘webtops’ or ‘netbooks’. This will cause a *major* ‘tip’ toward “open source” in these devices, though on Google’s terms.

    As an example, how long will it be before someone makes the Android equivalent of an iPod Touch sans the bullshit iTunes / ITMS lock-in? How long before the MythTV people release a remote viewing front-end for Android? And then how long until *the* MythTV front-end *is* Android?

    I’d have thought by now you would have scratched the obvious itch and sent up a flaming flare in the direction of Jonathan Schwartz on how the GPL licensing of Java was a Class-A mistake, because Google (via Dalvik) engineered around the issues with JME and then released that under APL v2, *specifically* citing the issues with the GPL in the process.

    Finally, Google isn’t that interested in saving the cell phone world via Android. Its more the case that Android helps them fight for Net Neutrality. What carrier is going to suggest that Google pay them for access to their customers when a very large percentage of the handsets in-use by said carrier’s customers run software from Google?

    There are other benefits, of course, but in the end, Android is far more strategic for Google than you have allowed.

  43. >However, for reasons unfathomable, you seem convinced that the phone in your hand
    is truly open (in the same way that say, the Open Moko / Open Runner are.)

    Nobody has explained to me how it fails to be. Yeah, sim-locking, blah blah blah. If I wanted my phone unlocked now, rather than in 75 days, I’m sure I could find someone who knows how. I really don’t give a crap. The T-mobile contract is even less relevant to the reasons I care about the openness of the stack.

    >As an example, how long will it be before someone makes the Android equivalent of an iPod Touch sans the bullshit iTunes / ITMS lock-in? How long before the MythTV people release a remote viewing front-end for Android? And then how long until *the* MythTV front-end *is* Android?

    I was going to blog about these possibilities, actually. Though I think you’re not giving Google credit for a long-term vision that’s even more expansive.

    >I’d have thought by now you would have scratched the obvious itch and sent up a flaming flare in the direction of Jonathan Schwartz on how the GPL licensing of Java was a Class-A mistake, because Google (via Dalvik) engineered around the issues with JME and then released that under APL v2, *specifically* citing the issues with the GPL in the process.

    Thought about it. Don’t need to; it’s 2008, not 1998, and I’m certain Schwartz’s own people are telling him that, if he hasn’t figured out himself. The only purpose sticking my oar in would serve is to raise my visibility, and I have no present need to do that.

    >There are other benefits, of course, but in the end, Android is far more strategic for Google than you have allowed.

    What the heck makes you think I’m discounting Google’s strategic interests? I know what they are; they’re flippin’ obvious. I’ll blog about this.

  44. Though I think you’re not giving Google credit for a long-term vision that’s even more expansive.

    Eric,

    You delete my words, you say I’m public enemy #1 on your blog, Russ Nelson pokes every chance he gets, and you expect me to write down everything ?

    Feh.

    The end game? The end game for Android is the replacement of the “linux desktop”. Oh no, not for you, or me. But Aunt Tillie… she would get everything she needs (and a lot more) via an Android Desktop distro, and whatever software she wants via the Android Marketplace.

    Think of it as the death/resurrection of “embedded linux”, too.

    Or think of it this way. There will shortly only be a handful of languages that matter:

    C: required deep down, but few apps will be written in C
    Java: unless python comes to dalvik
    JavaScript: will soon become *the* scripting language of choice.

    Lisp/Python: back-ends (on servers), though python needs help to deal with the onslaught of multi-core (*)

    TCL/Perl/PHP: dead meat
    C++: the choice of the last generation
    C#: crib death
    Objective-C: will only matter if Android fails

    (*) btw, that whole “64-bit will change the world, tipping the balance to linux” thing didn’t pan out, did it? :-)

  45. >(*) btw, that whole “64-bit will change the world, tipping the balance to linux” thing didn’t pan out, did it?

    Did, and didn’t. We got the transition timing right, pretty much nailed it to within 90 days. What we didn’t foresee is that the market momentum we thought would result in large deployments of Linux on conventional PCs would shift to netbooks. Interestingly, however, the 30-40% Linux now has in netbook market share is pretty much identical to our best we’re-trying-not-to-be-fanboy-optimist guesses about what we could achieve in PCs by 4Q 2008 – and it is enough to tip the power balance in the PC market from Microsoft to the Taiwanese. So I’m actually not feeling like we were seriously wrong; the Linux win looks like it’s going to have timing and dynamics pretty similar to what we projected.

    By coincidence, I was actually going to blog about netbooks and the state of Linux on PCs next; I’ll lay out the power-shift argument in detail there.

  46. Be careful, Asus has said that its standard netbook is going to be the 10″ screen with XP loaded. Everything smaller will be history.

    I’m not sure the 30-40% figure will hold for a “standard linux desktop”. What it did do is starve MSFT on two fronts:
    1) they get paid a (*whole*) lot less for XP on these machines,
    2) Vista isn’t happening elsewhere (though they’re aiming to re-try with ‘Windows 7″.

    So, they’re making less money, but they also have a long way to go until they’re broke.

    Meanwhile their only ‘partner’ in the cellphone space (rim) is getting beat up. The crackberry is a has-been, or will be as soon as the Enterprise and SMB spaces give up on Exchange and either a) host with Google Apps or b) something better comes along.

    64-bit seems to have had nothing to do with netbooks, and as for ‘right timing, wrong subject’: Correlation is not causation.

  47. >Asus has said that its standard netbook is going to be the 10″ screen with XP loaded. Everything smaller will be history.

    Three weeks ago I would have agreed with you — I’m the guy who still has a huge clunky VDT because I can’t get 2048×1536 in a flatscreen. The G1 has changed my mind; it suggests that the utility peak for browsers may be much further in the direction of small and light at the expense of display size than I had previously assumed. If that’s so, the vendor (and Microsoft) push to upsize netbooks will not in fact kill the smaller models.

    Six months from now, it would be really interesting to plot netbook sales volume versus display size. If we can get the information.

  48. Interestingly, however, the 30-40% Linux now has in netbook market share is pretty much identical to our best we’re-trying-not-to-be-fanboy-optimist guesses about what we could achieve in PCs by 4Q 2008 – and it is enough to tip the power balance in the PC market from Microsoft to the Taiwanese.

    It was initially 100%.

    Consumer demand pushed the netbook manufacturers back into Microsoft’s waiting arms.

    Why?

    Because Windows is easier. And, for what most users want to do, simply better.

  49. <blockquote.
    Three weeks ago I would have agreed with you — I’m the guy who still has a huge clunky VDT because I can’t get 2048×1536 in a flatscreen.

    Perhaps not, but you can get 2560 x 1600, which is, you know, *better*, and better for you, and these days less than $1k. Excuse denied.

    utility peak for browsers may be much further in the direction of small and light at the expense of display size

    Watch Javascript as it gets faster than python and adds support for things like Google’s “Gears” start to become dominant and we get “browser” “windows” that are really a complete, stand-alone application.

    Jeff Reed> Because Windows is easier. And, for what most users want to do, simply better.

    More likely, it is more familiar, and runs the software they already have. (e.g a 10th generation copy of the MSFT Office ’97 Install CD(s), with the “license key” on a sticky note inside the jewel case.)

    There are also some “worse is better” (aka “good enough”) arguments to be made.

  50. > I’m not aware of any actual lock-in in the G1. The entire software stack is open, and it’s not even clear that Google intends to close the root hole; I’ll be surprised if they do, actually, as I can’t think of anything in their business strategy around Android that would require it.

    I would have thought you’d realize, Eric: Google’s strategy isn’t in play there: *TMo’s* is. You (should) know how large call carriers act. The dev program isn’t subsidized by them, locked to them, nor (and here’s the big one) do they have to provide the same level of technical support to customers on it, so it’s a different story.

  51. And, incidentally: don’t you hate it when someone says “Great Googly Moogly!”, when, deep in your heart, you know it’s really only a pretty good googly moogly?

  52. So I know this thread is months old, but…

    With the recent release of the Motorola Droid on Verizon Wireless network, maybe it’s time for an update?

    It seems that another of Eric’s predictions is at least partially true: Android has Sprint and the rest of the lockin-mad mobile phone vendors shit-scared and scrambling to adapt. Though I wouldn’t describe Verizon’s behavior as “scared and scrambling.”

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong> <pre lang="" line="" escaped="" highlight="">