The Anti-Competitive Google
Google is betting that the Web is going to continue be a very strong platform. They are betting that web apps will provide a large chunk of our productivity and pleasure, where traditional desktop programs used to.1 With their strong hand in HTML5, their open-source Gears browser extension, and the drafting of the Wave protocol, Google is loudly proclaiming the Web to be an open environment that no one entity should control. Their actions, however, belie this vision in favor of anti-competitive corporate strategies.
Google’s public vision for an open web
Google has made the following statements to lend credence to that position:
- We are working on (and an employee of ours is controlling) the HTML5 specification, a collaboration between major browser vendors, reducing or eliminating the need for awkward and closed plugins like Flash or Silverlight; this makes the Web an open playing field, and any device and any operating system can now have the full web experience without Adobe or Microsoft’s consent.2
- We are writing a protocol called Wave to power the next generation of Web communication and synchronization, and we won’t control it; it will be email or HTTP, a specification anyone can implement without our consent or help.
- We wrote Gears and made it open-source, so all browsers (not just ours! cough, cough, Microsoft) may run web apps with features that once only native apps could have, and to speed up certain kinds of sites.
- In the meantime, we released Gears as a plug-in for popular browsers such as Safari and Firefox.
- We built Gears into Chrome, and hope other browsers include it by default in the future.
- We believe the Web is going to be so powerful that we announced plans for an OS based on Chrome (WebKit) and Linux (Just like Palm’s Web OS.)
Inconsistencies and other anti-competitive red flags
Let’s examine a couple ways Google has betrayed their publicly stated vision with contrary actions:
Google has a policy of not supporting or testing their web apps in Opera. Opera is a browser that has driven much web innovation (inventing not only tabs but Chrome’s new start page, to name a few) and always supported open Web standards (such as the effort to write HTML5, and in fact, their CTO invented CSS). Opera has very few inconsistencies with other websites.
The problem is so bad that Opera users can’t add Gmail events to Google calendars, and the emails that Opera users send in Gmail suffer collapsed line breaks so that the paragraphs run together. It’s been like this for years.
While I have called on Opera to step it up and fix these issues themselves to enhance their own market share (low market share is a vicious cycle for browsers and other platforms, because low market share means fewer developers support it, which inclines users to avoid it), I would argue that for Google to create and ignore these problems is anti-competitive. (I have discussed this before.)
How is this anti-competitive? It means that a company who makes the world’s best webmail says, “We choose to force 100% of the people who like our webmail to use one of the browsers on our list, and not Opera. In faaact, you will get the best experience in our own browser, Chrome.” What got Microsoft into trouble was throwing its dominance in one software category (OSs) around to dominate other markets (browsers, office suites, media players). To cite market share as a reason to avoid supporting Opera is a cop-out because of the vicious cycle I mentioned above: When you’re a big dog and a trendsetter with lots of cash, you better do the right thing and play fair.
Why is not supporting Opera inconsistent with an open web strategy? Because claiming to support an open-standards-based Web that works on all browsers, including Opera, with whom Google is working on HTML5, but not actually supporting that very open-standards-compliant browser just doesn’t make sense.
Seemingly dropping support for the Gears Firefox plug-in
As I write this, it has been a full 2 weeks since Firefox 3.5 officially launched (on June 30, 2009). There had been completely public betas and release candidates (the last of which, per Mozilla policy, became the stable 3.5 release, bit-for-bit) available for some time. And yet, to date (July 14, 2009), Google has not updated their Gears plugin for Firefox 3.5 (which, among other things, enables offline Gmail and speeds up WordPress publishers’ experiences).
You might be thinking, dear reader, that it could take significant time to update the plugin for Firefox 3.5. You would be mistaken, though quite understandably: Users have taken the open-source code, tweaked it, and compiled it to work with Firefox 3.5 on Windows, Linux, and Mac OS X. (This didn’t solve the problem, because Firefox does not know about these unofficial, untrusted fixes and can not/should not make them available via its add-on update routine.)
So it’s not that the upgrade takes too long. Google knew ahead of time Firefox 3.5 was coming and its users showed that the fix is fairly trivial. It’s just that Google is dragging their feet. Because they aren’t replying to users, no one knows why. (Does Google hate their customers? Or are they just arrogant?)
But users have their suspicions. One wonders, and I join him, if this isn’t a deliberate (and shady) attempt to get Firefox users to switch to Chrome. Is it? What other reason could be at play? The only alternative I can think of is just developer laziness and an incompetent manager or two. Then again, many people at Google use Firefox, and there is no way this issue could have gone unnoticed.
How is this anti-competitive? Why is it internally inconsistent? Google says it wants an open, interoperable, Web with a diversity of browsers all supporting the same open standards and ideally with enhancements like Gears (which is, remember, open source) that make for a better experience. But by denying Gears support for Firefox, they are pressuring Gmail users to switch to Chrome; using one product to artificially support another is anti-competitive and any attempt to homogenize browsers toward one (Chrome & the WebKit engine) betrays the vision of the open Web.
True feelings about the open web
I wish I could summarize Google’s true position, but the contradictions here are too great for an outsider like me to resolve. The least we can conclude is their actions are considerably less benevolent than Google would have us believe.
Let me know if you have anything to add.
Update (February 14, 2010): Google Buzz, which I believe will fail, so aggressively competitive that Dare Obasanjo notes they are using privacy as a pawn in a competitive chess game, even if it jeopardizes their very safety.
Hence the Chrome OS initiative. My reaction is between the unthinkingly obsessive Arrington-style “OMG this will be big” and the piffling “won’t be anything” response from others: I think Chrome will be just what it is planned to be, a competent entry for netbooks. That is, no, it won’t ever gain much if any traction on traditional personal computers; but if done well, it could provide a smooth (if limited) experience on small, cheap devices used mostly for email and casual surfing. ↩︎
Sounds a lot like what Apple’s gung-ho true-web-technology push, as they embrace HTML5 wholeheartedly in Safari, extend CSS in innovative and backwards-compatible ways (transitions, gradients, transformations), make WebKit (the core of Safari) available for anyone to use for pretty much anything (from Chrome to Palm WebOS to NetNewsWire), and reject plugins (as they should be able to do!) on the iPhone. The inconvenience of missing Flash, in my view, is not a poor reflection on Apple but on everyone who has used Flash as if it were a Web technology. It is not. It does not integrate well with the rest of the webpage and needs drawn in a separate pass from the rest of the rendering engine, unlike pure web technologies like HTML text, images, SVG, and even