Daniel Eran Dilger
Random header image... Refresh for more!

Ars’ Jon Stokes hails Chrome OS as the second coming of the PC

Daniel Eran Dilger

In his latest effort to cram most of a thesaurus into an Ars Technica article, Jon Stokes has written a glowing tribute to the web search emperor’s new clothes: Chrome OS, and outlines why Google will save the PC from itself. He’s wrong, here’s why.
I can’t help myself

I’m not excited to be writing another article critical of something from Google, lest I further embolden my circle of critics who love to write me off as a fanboy of Apple, without regard to the fact that everything I write pretty much turns out to be spot on and deadly accurate.

Fake Steve Jobs (aka Daniel Lyons) recently castigated me for knocking Android (well, ostensibly, but really for my having assailed his career arc as a Linux basher and complicit Microsoft stooge) in a parody that was actually pretty funny if you haven’t seen it yet.

There are others who are less comical in their efforts to discredit me, like the retired Android hobbyist who wrote me scathing hate mail about how insulting it was to see me say out loud that the Android Market is pretty much just a hobbyist shareware rummage sale. He threatened to have ZDNet look into my background to answer the burning question “does Phil Schiller sign your checks?”

I’m also wary of being silenced by Google (both in search and in ads, because Google monopolizes both on the web) for voicing any alternative viewpoints on its various strategies. But I have to write this, and can’t stop myself, so here it is.

The Secret Diary of Steve Jobs : Guest blogger Daniel Eran Dilger on why Android will fail
Daniel Lyons: Fake Steve Jobs and the SCO Shill Who Hated Linux
Forbes’ Fake Steve Jobs Is Also Fake On Apple

Core competencies

Chrome OS is an interesting attempt to pull off something that has failed in the past: delivering a web appliance as an alternative to the low-end PC. Perhaps with Google’s ad muscle, the company can make the new venture work. I’d like to see Chrome OS succeed, because it would be a strong validation of HTML5 as an open platform for web developers.

But given that Chrome OS is still unfinished and unproven and will be for another year, it’s a bit much to be touting it as having vanquished the status quo and solved all the problems of PCs, as Stokes recently did in his article on Ars, particularly when the new web OS really does nothing to solve any problems and instead only raises several new and significant limitations.

There’s lots to like about Google in general. It’s very successful at web search and monetizing web content (like mine) and apps (like the excellent Maps and so on) and building core web technologies (note that I’m a big fan of HTML5, and Google can be considered the main proponent of that). But Google is no Apple. It has no hardware expertise, no consumer savvy, no real capacity to create intuitive user interfaces. Like Microsoft, Google has been quite successful at being itself, but really bad at pretending to be Apple.

This being the case, I’ve been quick to criticize pundits who arrive at the conclusion that any company that decides to compete against Apple on its home turf will be wildly successful, despite lacking any experience, customer loyalty, and so on. Like the Microsoft Zune or…

Why Apple is betting on HTML 5: a web history

Take Android

For example, when it was first rumored that Google had bought Android and was developing a new open source strategy related to mobiles, I wrote that those who thought this meant that Google would be developing an iPhone-killer were flat out wrong. I explained that Google has no interest in killing the iPhone, and instead that it wanted to kill Windows Mobile and another platforms that might block Google’s access to mobile web dollars. This appeared prescient in 2007 when I wrote it, but it’s pretty obvious now.

The Great Google gPhone Myth

I have repeated this over and over as it slowly became more obvious to pundits who make their living stirring up false or sensationalized information to generate ad clicks. There is nothing wrong with Android as Google intended it be, it’s just that Google didn’t create Android to be “an iPhone” in a world-changing sense of new technology designed to shake up the industry.

Android was designed to replace WiMo and in its place install Google’s adware/spyware platform on the same limited hardware phones that Microsoft designed. It’s that simple. The Android platform is going to continue to be a Windows-style mess, but it won’t benefit from the monopoly protection that Microsoft enjoyed because there’s too much competition within Android (because Google isn’t policing things like Microsoft did in PCs) and outside of the platform (iPhone, RIM, Symbian, and Bada all have bigger names in smartphone hardware behind them than Android does, and even Palm is in the running with its own WebOS).

Inside Google’s Android and Apple’s iPhone OS as core platforms
Inside Google’s Android and Apple’s iPhone OS as advancing technology
Inside Google’s Android and Apple’s iPhone OS as business models
Inside Google’s Android and Apple’s iPhone OS as software markets

Chrome OS

Chrome OS is a very different ball of wax compared to Android, even if both share a lot some of the same wax. Many of the problems I see in Android do not necessarily apply to Chrome OS, because while Android is trying to beat Windows Mobile at its own game, Chrome OS is trying to take on Windows XP as being the way people launch their web browser on netbooks, a completely different kind of effort.

Stokes wrote an impassioned piece about Chrome OS for Ars, and describes it (I’m paraphrasing here) as Google ‘starting with the Internet and making an OS, rather than starting with an OS and trying to figure out how to make the Internet’s series of tubes fit together, which is currently an experience like trying to eat flan while skydiving.’ He writes as if Apple and Microsoft have been unsuccessfully falling all over each other for the past 15 years in trying to figure out how to allow PC users to browse the web, let alone The Cloud.

This is such complete nonsense that its simply embarrassing to read. First off, neither Mac OS X nor even Windows 7 have hit any wall of diminishing returns that requires a new approach to PC operating systems answered by HTTP access alone. But secondly, Google hasn’t invented something brilliantly new here. Chrome OS is a distribution of Linux that only runs one app: a web browser.

It’s that simple.

Google is adding value by stripping away conventional services that a PC normally runs (and which aren’t necessary on a netbook, the idea goes), and building support around HTML5 to enable web apps to do things that a conventional browser hasn’t been able to do before. That includes offline browsing, local thin client style storage on solid state disk, and a stronger security partition around individual web pages, so one tab’s browser exploit can’t affect things happening on another browser tab.

Chrome OS: Internet failing at PC > PC failing at Internet
Why Apple is betting on HTML 5: a web history

Rather than being anything like Android, Chrome OS is another take on the Palm Pre’s WebOS platform, but for netbooks rather than iPhone killers. Unlike Palm’s new OS, Chrome OS is backed by Google’s vast fortunes and can be implemented by a variety of hardware makers. Or from the opposite perspective, it’s backed by a company with no hardware or consumer platform expertise and set to be deployed across a mess of fractionalized, squabbling competitors two years after the Palm WebOS was first announced.

Features are flaws written upside down

Chrome OS is a laudable, interesting project, but Stokes gets so excited about things that he begins to gloss over reality and spew completely asinine stuff. Among the new systems “features” are a series of flaws written upside down, such as:

“The OS will support only a limited number of Google-blessed devices and peripherals, which is Google’s way of ensuring reliability and security.” Yeah this was always a popular feature on other platforms that would only work with a few vendor-approved devices. It was such a drag for users that Windows could work with pretty much anything.

The OS will also be awesome because Google will ostensibly “optimize it for Internet HTTP traffic: very high numbers of simultaneous connections and high latencies. Thus Google can make tradeoffs down in the networking stack that give a better user experience on ChromeOS, without worrying about how the OS will perform on a LAN with different file protocols and such.”

In other words, Chrome OS won’t be able to connect to network file shares. Or torrents. Yeah that’s awesome. And it’s cool Google started with the Linux kernel, which offers so much more that needs to be rewritten and optimized compared to the superior networking stack of BSD. Clearly, Google wants to deliver the most highly optimized network technologies possible for its cheapskate users browsing the web on low cost mobile netbooks, where the network stack is so much more important than the scant amount of RAM and the available Atom/ARM mobile processor resources.

And what about the core weakness of the web as a platform, that it is abstracted so far from away from the underlying hardware? Well, there are emerging standards to bridge this gulf using open APIs for accessing things like the GPU. But rather than backing the WebGL 3D API standard pushed by Khronos (OpenGL), Mozilla and others, Google went out and bought a company, in the style of Microsoft, and plans to use that company’s competing O3D API instead.

This isn’t necessarily bad, and Google’s 3D API efforts are not proprietary like Microsoft’s, but Stokes writes it all up like it’s a feature that Chrome OS devices won’t have any real graphics power, and that it’s a good thing that Google isn’t embracing community standards. This reeks of unhinged fanboyism rather than rational discussion of a technology and its suitability and prospects.

Next up: “while ChromeOS has a filesystem of some sort, you’ll never see it. I, for one, couldn’t be more thrilled.” Yes, this is very innovative. Just like the iPhone. And when that came out, everyone in punditland blew a gasket about Apple ‘not being open and exercising too much control’ by taking away users’ view of their files. But when Google comes out and announces plans to do the same thing four years later (and a year from now), it’s the bee’s knees.

Ugly is a virtue if you use a thesaurus creatively

Stokes gets so carried away trying to express his deep passion for a vaporware concept still a year away that he resorts to blowing out what is perhaps the most opaque paragraph ever written on a tech website. The setup: a backhanded acknowledgment that web apps are generally all really ugly and unusable compared to the native apps (Mail and iTunes and so on) that users are accustomed to using.

“The unsightly way that Chrome OS implements Web-based knock-offs of regular desktop apps will strike many users as crude,” Stokes writes, before generating a paragraph with the clarity and brilliance of a lump of coal dipped in tar:

“But the awkward visual effect of Chrome OS is unlovely in the way that VisiCalc was unlovely, and not in the way that, say, many things about Apple’s iTunes or Windows Vista are unlovely. The former’s Spartan-looking imitations of traditional desktop apps are awkward because they represent an embryonic stage of something new (ontogeny recapitulates phylogeny), while the latter’s frenzy of buttons and dialogs and sync options are less awkward than they are exhausted.”

Sometimes using a big word is the only way to get across the idea you have in mind, but that paragraph is so overwrought that it just insults the reader. And the message he conveys is not even significant. He could also have said, “Chrome OS currently looks like awkward crap, but there are things we don’t like about iTunes and Windows, too, and it looks like Google will somehow solve all its user interface issues from scratch faster than Apple and Microsoft can improve theirs, just because we have blind faith that Google can do no wrong, even in areas where it has no particular expertise nor proven track record.”

A slow and painful system of tubes

Microsoft and Apple are, Stokes writes, “currently involved in a slow and painful process of trying to stretch and push ‘the PC’ out towards the Internet and towards a more useful and integrated relationship with the cloud as a new type of server.”

Actually, no. Microsoft’s painful stretch occurred in 1996, and Apple’s in 2002. Microsoft is now sitting on the web like a satiated buffet client in Vegas. Apple is delivering innovative Internet and web services that millions of people actually pay for, unlike Google’s adware services that most people will only use for free. It’s not Apple or Microsoft that is struggling to put together a cohesive push strategy for mobile and desktop users. It’s Google stretching to deliver this with an adware business model. That’s not even really novel, and not without competition.

“So while I could write a book about all the ways in which Apple and Microsoft have failed at the Internet over the past decade,” Stokes concludes, “as of today I can only write a few lines about the ways in which ChromeOS may or may not fail at the PC.”

Google has never failed in war with Eurasia

Well, that’s because Google has yet to fail at anything. Except for failing to compete against YouTube with its own Google Video service, and subsequently having to buy out its much smaller rival. And its various products that didn’t work out, like Google Catalogs, Google Notebooks, Google Mashup Editor, Dodgeball (an SMS location reporting app), and a Twitter competitor called Jaiku. And Knol, that thing that was supposed to replace Wikipedia.

And never mind that Android is just Windows Mobile without the Windows part. No doubt web users’ brand familiarity with Google is going to cause them to buy Chrome OS devices from netbook makers that they wouldn’t otherwise buy were they only running X11/Linux or Windows, and everyone will overlook the fact that netbooks running nothing but a web browser have significant disadvantages never before experienced by users. Some of those disadvantages are the reason Linux hasn’t worked out, even running on netbooks where there isn’t nearly as much in the way of expectations as conventional PCs.

I personally hope Chrome OS takes off and offers a suitable competitor on the low end to make cheap, useful appliance netbooks. It will be interesting to see how these stack up against devices like the iPod touch, which do the web but also do games and apps and offer iPod features. Maybe it will force Apple to offer Bluetooth keyboard support for the iPhone OS, to enable a mini-laptop sort of feature that I’ve requested before. I’m all up for competition.

However, while gushing specious accolades for vaporware still a year away is somewhat irresponsible, and stating things like “And in 2009, it’s a much bigger deal for a PC company to fail at the Internet than it is for an Internet company to fail at the PC” are downright ridiculous. It’s 2009 and Chrome OS is nowhere to be seen. And PCs are not “failing at the Internet” in any way, shape or form. Windows PCs and Mac are the way 97.8% of the world accesses the Internet!

Time is on Google’s side?

Not content to let that be the most absurd line in the article however, Stokes finished with, “Right now, at least, Google has time on its side, and the company can afford to release cloud clients (Android and ChromeOS) that start out awkward and immature, and let them develop as the user-facing part of a still-evolving cloud.”

Right, Google has lots of time to wait around before Microsoft recovers from its Vista disaster (remember when the same window of opportunity was presented as dire and fleeting for Apple two years ago?) and before Apple takes out an even more solid claim of the market both in PCs with the Mac and mobiles with the iPhone and iPod touch. It’s a waiting game! Perhaps Google would do even better if it holds out to 2012.

I appreciate Ars writing original articles rather than just closely paraphrasing mine and reprinting them without attribution (I wish they’d stop doing that entirely), but this kind of boldly-stated puffery should not have made it past the editor.