Friday, March 14, 2014

Re: Chromebook belongs to computing's past, not its future

About the author

Jeff Nelson invented Google's ChromeOS operating system while working at Google in 2006.  Nelson is the author of two books and many magazine articles.  He has over 20 years experience building world class teams in the Internet industry.

Responding to Joe Wilcox's recent blog which often focuses on Chromebook, Joe Wilcox has written a book, 'Chromebook Matters'[] and several blog posts about Chromebook.

 "Twenty Fourteen isn’t Year of the Chromebook"[]

 "Why Chromebook is good for students", []

 "Chromebook belongs to computing's past, not its future"[]

He raises a number of interesting points that I'd like to respond to.  However, rather than responding to many points in a water-ed down, rambling blog, I'm just going to respond to perhaps the most important point.

Wilcox writes, "Chromebook is an aberration, only made possible by Google’s sync smarts and supporting services platform."

I think you are focusing too much on the specific platform and not enough on the big picture.  The industry has been trending more heavily to a fully connected world and client-server architectures in which data lives on the server.  Further, regardless of whether you are using a smartphone, PC, Mac, or ChromeOS, a large portion of your data is actually residing off your device - on a remote server.  ChromeOS may be the most pure form of that client-server architecture, but I hope you will agree that the trend away from stand-alone architectures toward Internet architectures is only accelerating.

Wilcox goes on, "As voice and touch replace keyboards, devices like smartphones and phablets make more sense."  I don't like to draw hard lines based on hardware configurations.  The physical keyboard does not define what's a Chromebook and what's not a Chromebook, or whether or not the platform will be successful in the market in the long run.  Quite the contrary, you can add a physical keyboard to just about any device.  Visa versa, you can run ChromeOS without a keyboard.  In my view, the OS should be viewed as agnostic to the hardware configuration, particularly when such minor issues as whether or not a keyboard is included in the packaged product.

Arguably, the industry has already seen the market diverge into two or more sub-species of hardware configuration, based primarily on context of how they are being used.  Smartphones are for around town, because they are easily carried.  Tablets are for home-use, because they aren't easily carried.  Laptops and PCs are for 24/7 professionals who need the speed and ergonomics of a full-sized keyboard and monitor.

As for voice, I am entirely convinced, as someone who has tried to rely heavily on voice and has done some work in the voice recognition industry, voice is not suitable for replacing keyboards on a full-time basis.  If you think it might be, I'd encourage you to try it for a week, yourself.  I found after only a few hours, my throat was already noticeably irritated.  After about a week, my voice was starting to give out.  Speaking continuously for hours and days just places too much strain on vocal cords. Further, it's not necessarily faster than a skilled typist at data entry.  Is there a place for voice? Yes. Is it going to replace keyboards for professionals? Not a chance.

I can't say if the specific Chromebook product will be a marketing success in the long run. That is, in fact, entirely out of my hands since I no longer work for Google, but I am confident Chromebook's Internet reliant, client-server architecture is here to stay.


  1. Thanks for the email and link to the post.

    Please take a look at the majority of my writings about Chromebook, which are more in line with what you write in your post and also with the tone of my book, which I presume you haven’t read based on your blog post.

    My writing style is to look at topics from different perspectives, and I take the more negative vantage when my colleagues in the news media start sounding an Echo Chamber. My recent Chromebook posts are meant to counterbalance sudden media fan frenzy — bloggers and journalists relating the same points of view because they think it’s vogue.

    I have a reputation for being anti-Apple, which I am not, for similar reason. When Apple was the tech press’ darling, I wrote harder stories.

    That said, Chrome OS is an aberration. No other new PC operating system succeeded after Windows 3.1’s release ( Mac OS came earlier). Even Linux is largely confined to servers. Microsoft’s monopoly is insurmountable for many reasons, with the applications barrier to entry being right near the top. People don’t buy a new OS without apps, and developers don’t create them without adoption. It’s a chicken-and-egg scenario Chrome OS beats because Google brings so many apps and services and the browser is an established app platform, while being a familiar motif for user interface. No wonder there is amazement among tech bloggers and journalists.

    As for mice and keyboards, they are familiar but unnatural constructions. The most valuable user interface is you. Your fingers, voice, and eyes.

    I know lots of small business people who never use computers, but depend on smartphones every day. In emerging markets, Gartner, IDC, and other analyst firms say that the first Internet-capable device purchased is typically a phone. In 2013, the second switched to tablet from PC. What sometimes is referred to as "technology skip" is common phenomenon, where users in a new market jump over one product category for another. That’s the pattern with PCs.

    I really appreciate your thoughtful response and do agree with many of your points.

    1. I appreciate your feedback as well.

      Technology skip in the emerging markets was much more about economics, than it is about technology trends. Technology professionals working on software, IT, or data entry based in places like China and India still own or have a computer provided for them; they don't do their work on a phablet.

      In fact, I would say one of the problems with the tablet/phablet/smartphone platform right now is that so few authoring tools are available directly on the platform. That is to say, if you want to develop software for iPhone, you MUST purchase a Mac. There's little or no content authoring, aside from the most primitive documents or video upload (as you mentioned in one of your columns), on the devices themselves.

      That's a problem and probably also a market opportunity for a software organization that is able to create a simple enough set of tools to create complex applications directly on devices.

    2. True enough. But can you develop for Chromebook on a Chromebook? Is the situation similar to tablets? I'm not a developer and so wouldn't first-hand know.

    3. On whether people can develop software with just a Chromebook, just using a Chromebook I made this software which runs on a Chromebook and supports writing more advanced software:

      So yes, you can do it. That uses IndexedDB to store data locally on the Chromebook, which you can the export as JSON text to store in files on the Chromebook (I like the Chrome extension "Caret" editor) or cut-and-paste to GitHub.

      On getting credit, I've been tangentially involved with some big ideas from WordNet which ultimately made Google billions (George Miller was my undergrad adviser at Princeton and I was talking about semantic networks of triples and graduated just as he started it at the end of his career) through Watson/Jeopardy (I developed a speech system at IBM Research around 1999 where people could direct queries to a 3X3 display wall of recycled ThinkPad laptops, even made a presentation on it to IBM's legal department) -- and probably a bunch more. What I've seen is that it is pretty hard to get credit or be remembered years later for being part of the spark or muse for anything someone else works hard later to perfect or expand on. And that often is not out of any failing on the part of the later people in terms of morality or scholarship other than it often being hard to recall or even recognize where the sparks for ideas come from -- and many ideas come from multiple place, like George interacted with Allen Newell and Herbert Simon at CMU. Although, to be fair, I've also realized that I'm often forty years behind the thought leaders in sci-fi (like Theodore Sturgeon with "The Skills of Xanadu" which amazed me about shared knowledge, or Isaac Asimov with his wall-based display of data in the Foundation series which amazed me as a concept) -- while I am then often maybe ten to twenty years ahead of the mainstream by liking those same ideas. And like for almost anyone, there are no doubt ideas I see once somewhere and which then go into my subconscious forgetting the original (like maybe "Data and Reality" by William Kent relating to triples which I might have seen at an IBM library around 1980, and inspired my Pointrel triplestore work in the 1980s that predated both RDF and WordNet). Certainly writings by Victor Serebriakoff in "Brain" and Gregory Bateson's "Steps to an Ecology of Mind" also informed my own thinking on semantic networks, just like "Silent Running" inspired my work in robotics and interest in speech recognition (along with other sources from "Lost In Space" and the Robot to "Wonderama" which once showed a device that made pictures of words from the audio signal). Our thoughts are shaped by so many influences recognized and unrecognized...

      (to be continued for size limitations...)

    4. (continuing the previous comment)

      Getting credit for anything is such a slippery topic given how often we get inspirations or implementations from other places (like Puppy Linux for "Guppy" Google OS, as mentioned?), and given how often people then build further on what we say or do in a "stigmergic" way. Minecraft is another example, given so much of it is taken from "Infiniminer" by Zachary Barth, and yet not everything about Minecraft comes from there. Big funding rarely goes to the muse, it seems. :-) Otherwise I'd have some claim to Google's billions that have come from WordNet-powered Adsense, which I would have poured into implementing a social semantic desktop. :-) It can be so time-consuming and emotionally and financially draining to be ahead of the mainstream times -- if still beyond the true visionaries, including for Chromebook of Alan Kay and the Dynabook or Theodore Sturgeon and The Skills of Xanadu short-story (well worth reading as a 1950s vision of mobile computing and social networking and manufacturing on demand, and which inspired Ted Nelson and that Xanadu which pioneered Hypertext), and of course the oft-forgotten Doug Engelbart and "The Mother of All Demos" or Vannevar Bush's earlier "Memex" (although roots go back to "The Walking People" and their "Standing Bear" cave paintings mentionedin their oral history from 1000s of years ago for instructing the young on dealing with bears and such, or probably for other older peoples if the idea of abstract representations was several times re-invented). The people who come next often don't realize what it took to prepare the soil so their own "thought seeds" could sprout and grow. In today's society, it's also a fairly different mindset (often requiring broad imagination) to push a new idea for the current culture through some working demos versus to then implement that same specific idea in a big way for a big audience as a big enterprise once it catches on (often requiring narrow focus and unhappy tradeoffs related to economics and central control). And since "the rich get richer", those who put their time into innovation for the public good (an exhausting process of banging your head repeatedly against the wall of mainstream short-term economics which often socializes costs while it privatizes gains), those innovators are most likely statistically to see fragments of such ideas developed in a big way by others with deep pockets (financially, socially, politically) who take few risks but just wait to see which way the wind is blowing and then try to own all the results as much as they can. Microsoft perfected that approach it seems in the 1990s...

      That's all one reason I like C. H. Douglas' suggestion of "Social Credit" or later a "Basic Income" as a way to fund invention and innovation by ensuring everyone gets some of the fruits of our society's productivity, given so much of today's productivity relies on our common knowledgebase which was developed in all sorts of tangled ways over thousands of years. But there I go again, :-) being ten to twenty years ahead of the mainstream in the USA (although the idea is getting more popular in Germany). As it is now, the patent lottery for funding innovation has all sorts of problems, not the least of which is the fundamental immorality of creating "artificial scarcity". And I say that despite my name being on a software patent from IBM Research related to embedded speech recognition for command-and-control. That was from being part of the IBM "Personal Speech Assistant" project pushed by IBM Master Inventor Liam Comerford which predated Siri and Google Now etc. by about a decade. And frankly, the "Dynabook" by Alan Kay as prior art is way before so much of what people patent these days about distributed personal computing and such...

      (to be continued...)

    5. (Continuing previous comment as 3 of 3)

      Anyway, I'm writing this on a weekend sitting in an easy chair beside our cat using a lightweight $250 Samsung Chromebook, despite having a much more powerful MacPro in my home office with a couple big screens as part of a treadmill workstation, and also a couple heavier harder-to-use-and-maintain laptops with shorter battery life lying around. Now if only it was easier to run Minecraft on a Chromebook, like your original Google OS vision... :-) At least there is voxel.js though which runs on the Chromebook, and which I could perhaps hope someday to improve via the Chromebook to be better than Minecraft... Thanks for pushing the Dynabook idea further along via your own take on it at Google, Jeff! We'll hopefully get there fully someday...

    6. Postscript to the above three items:

      As above, I do feel I can lay some claim to being a spark indirectly behind WordNet and so even more indirectly to aspects of Adsense. I tell more of that here: "On college and space habitats"!msg/openvirgle/PdK35mSNoSU/3zLpZuljHiMJ

      *BUT*, what I should mention to be fair is that, beyond other ideas from George A. Miller that may have made it through my stick skull at the time, his generous example of setting WordNet *free* is no doubt part of the spark for all the free software I (and my wife) later developed and gave away (including our Garden Simulator). That also includes that version of Pointrel software linked to which is under a FOSS License. At the time he decide to make WordNet free (supported with funding by the government), so many other academics were closing up their government-funded work to commercialize it under the Bayh-Dole act. And I had had similar commercial hopes myself for the Pointrel system, based on commercialization examples I had been shown by others earlier. It took a long time and further examples to move beyond that, but George's example was an early part of that. So, sparks and examples can go both ways.

      And searching via Google (however Google funds itself via Adsense or other ways) has made possible so many essays of mine (including aspects of this reply). That is a big "win" for the Pointrel system, even if in general it has never gone much past a sort of "doodle" I've been redrawing for about thirty years. So, again, benefits can go both ways, even if not directly acknowledged. As is said on a favorite notecard I've run out of from the long-defunct New Alchemy Institute, "We all live in one world, and each act we do affects the whole."

    7. Typo: That should be "thick" skull, not "stick" skull.

      That was back when I though essentially that learning anything might "contaminate" my creativity by biasing my perceptions (not totally without some grain of truth, see Jeff Schmidt's book "Disciplined Minds") -- without realizing how much I had learned from others already, and how much of creativity is putting together building blocks you got from others in new ways appropriate to current needs...

  2. The ChromeOS is an interesting competitor of Windows and OSx, it's RAM based framework allows for the cheapest of computers to become what you need them to be. Though all technical innovations aside (ie. touch, voice), software innovations also need to be focused on. People have a distinct recognition of Windows for it's simple executable installation, and not even that, a multitude of proprietary library's allow for applications such as games to take off marvelously.

    The ChromeOS lacks these libraries and thus lacks support of the gaming community which I feel is the largest investor in the personal computer industry. Easy and Interesting implementation of virtual machines would allow users who require software to use it when needed. Though in the sense of non gaming, the esthetic that OSx can offer are in par with what the post request in the sense of home entertainment.

    The ChromeOS however could be targeted at more specialized fields, such as cryptography and software development. Selling USB sticks with a ChromeOS image and enough free space for industrial programs would allow microcomputers to be easily handled on the field, and cryptographers easy access to evidence.

    So depending on the market google is looking to approach they can really make a killing or become the killed. Linux distro that is.

  3. It amuses me that people criticise a product because it isn't exactly the same as the one it challenges, forgetting that if it was, it would be pointless.

    Maybe Mr Wilcox isn't 'anti-Apple' or anti-Chromebook'; maybe it is just that he writes 'hard stories' while other go for hype ... But if that were so, why isn't he just a little harder on Microsoft?

    Most of his criticism appears on forums all over the tech world, from the mouths of Microsoft fanbois, 99% of whom have never used a Chromebook.

    To argue "Microsoft’s monopoly is insurmountable for many reasons" is really simply saying 'no-one has done it therefore it can't be done'. And the market is proving him wrong. And as Chromebook develops, he'll be even wronger.

    Tech cemetries are full of technologies which couldn't be beat ... and they are over there digging a hole for Windows as we speak.

    "I see Windows everywhere, and it doesn't know it's dead"