Google Chrome Frame: Why they did it and why it probably won't work

On September 23, Google released Chrome Frame, an add-on for Internet Explorer (IE) 6-8. Chrome Frame allows websites to request that IE visitors use the rendering engine behind Google’s speedy Chrome web browser instead of IE’s native engine. A TechCrunch synopsis and the Chrome Frame page provide further explanation. This article offers strategic insight into…



On September 23, Google released Chrome Frame, an add-on for Internet Explorer (IE) 6-8. Chrome Frame allows websites to request that IE visitors use the rendering engine behind Google’s speedy Chrome web browser instead of IE’s native engine. A TechCrunch synopsis and the Chrome Frame page provide further explanation. This article offers strategic insight into why Google is aggressively pushing their own browser technology, whether Chrome Frame will succeed, and how Chrome Frame should be seen by web development clients.

Chrome Frame

Ask any web developer what they think of Internet Explorer 6 and you’ll hear an earful. The 8 year old web browser still commands nearly 20% of the browser market and is woefully inadequate at supporting modern standards, incurring millions of dollars for legacy support every year. IE 7 and 8 were big improvements, but as we’ve opined on before, even IE8 fails to support forward looking techniques supported by the competition.

In the 6 month since IE8′s release, competitors Firefox, Chrome, Safari, and even Opera, have all seen major updates. All of them introduced performance upgrades, in particular to their JavaScript engines. JavaScript is increasingly the engine for dynamic content on websites, from animations to on the fly content loading without page reloads (via AJAX). Google’s browser, Chrome, positioned itself from day one as focused on performance, JavaScript performance in particular. At least in theoretical tests, it more than delivers on its promise.

Microsoft argues – not without merit – that the real world performance difference on today’s web is negligible or even in IE8′s favor. But Google isn’t thinking about “today’s” web: they’re focused on building the next generation of web based products. And they’ve made it clear that web browser technology – on which they depend – isn’t moving where it needs to at a fast enough pace.

Google claims that’s the reason Chrome exists: to help push other browsers forward. Chrome’s rendering engine is built on the modern, speedy, and open source WebKit that also powers Apple’s Safari and is increasingly dominating the mobile space.  But Google built their own JavaScript engine, coined V8. The fact that they’ve made V8 source code openly available suggests that their rationale for Chrome’s existence – and the newly announced Chrome Frame – is believable.

Don’t believe it? Check out this Nintendo emulator, built entirely on JavaScript. In the about section, the author points out that its pretty playable on Chrome, and minimally playable on Firefox and Safari. On my 2 year old Dell laptop, Super Mario Brothers runs at about 60 frames per second in Chrome 3, and about 8 fps in Firefox 3.5. It even runs at about 2.5 fps on my iPhone 3GS! Internet Explorer 8? Doesn’t even run. Is it unreasonable to argue that we should be able to run web applications at least as powerful as a 24 year old gaming console?

Google’s rationale for Chrome (improving web browsers) may seem selfless at first blush, but don’t be fooled. Google knows that their continued growth depends on more eyeballs, more often so they can deliver more ads and – let’s face it – collect more information about our habits. The path to more eyeballs more often is paved with killer web applications, and being able to traverse the path of killer applications depends on the web browser.

I doubt Google wants to be in the business of building web browsers any more than Honda wants to manufacture tires. They’re just dependent, and unsatisfied, most of all, with Microsoft’s dominant web browser – the most “behind” of the batch. Enter Chrome Frame: a plug-in for Internet Explorer 6-8 that, combined with a meta tag in the HTML’s header, will let Internet Explorer use Chrome instead of Microsoft’s Trident engine to render web pages. To be clear, Google is making no bones about their beef with IE6-8. On the Chrome Frame download page, they’re upfront:

Start using open web technologies – like the HTML5 canvas tag – right away, even technologies that aren’t yet supported in Internet Explorer 6, 7, or 8.

Take advantage of JavaScript performance improvements to make your apps faster and more responsive.

So we’re clear on – and quite sympathetic to – the rationale behind Chrome and Chrome Frame. A couple of our our more technically sophisticated clients reached out to us about Chrome Frame today, wondering if they should implement the meta tag that tells Internet Explorer to use Chrome Frame. Admittedly, without having time to experiment with all 3 compatible browsers, we have to admit we’re hesitant to endorse this, and skeptical about its success.

It seems to me, there are a few reasons people are using Internet Explorer.

Especially with Microsoft aggressively pushing and incentivizing upgrades (for which they deserve credit), the only likely reason that someone is using an out of date version of Internet Explorer (particularly 6) is either because they just don’t care (even going so far as to ignore Windows Update requests), or because the browser is imposed on them by IT departments for legacy or business reasons. If the user doesn’t care, it seems fairly unlikely that he or she would ever care to install Chrome Frame (let alone even hear about it).

If the browser is imposed by IT on a user who is interested in installing an alternative but locked out, it seems doubtful the user would be allowed to install an Internet Explorer add-on. Note: I don’t have a locked down environment to test this in, so if anyone can confirm whether the Chrome Frame add-on can be installed even on a machine locked down from installing applications, let me know. If so, and I’m wrong about this assumption, there may well be a significant niche to consider.

With Internet Explorer 8, some may genuinely prefer the browser. It may be comfort with Microsoft products, a preference for the user interface and features, or just a choice to use the browser that shipped with the operating system. There may be a small audience that prefers the look and feel, but would be happy to use a different rendering engine. But it seems to me that most users who choose to use IE8 because they believe it’s a better product wouldn’t want to install the engine for a product that they actively chose not to use (Chrome).

In short, if you want to use the Chrome engine, and you’re allowed to install add-ons and applications, why not use, well, Chrome “proper”?

The best answer is that the user is simply unaware of the option. Let’s face it: the fact that I’ve seen a basic question (answer?) on Jeopardy about Chrome in a higher dollar category reveals what is probably obvious: outside of the moderately technically savvy – those who make a conscious choice about their browser – most people probably don’t have a clue what Google Chrome is.

Google tries to solve this by allowing your website to offer Google Chrome to the visitor if the site is using the special meta tag. But do I feel comfortable telling clients that they should force 65% of their audience to be greeted by a “would you like to install Google Chrome” message? Can’t say I do. Seems like for most it would be an annoyance at a best, and at worst, scare off visitors who don’t know what “Chrome” is and wonder if some ad is being aggressively pushed on them.

Chrome is a great browser – my personal primary choice for casual surfing (Firefox is still the best for development) – and I appreciate the idea behind Chrome Frame. Will the idea work? For now, color me skeptical.

Categories: /

9 Responses

  1. klamer says:

    As to be expected, on our secured and heavily firewalled network where I am not allowed to install Chrome I also am not able to download the Chrome Frame add-on… Too bad, this means there is no way I can here try the Javascript SNES emulator ;-)

  2. Daniel says:

    Very well written. I’m also having a hard time seeing “the frame” as a success story. However, it’s likely that Microsoft will be really annoyed with this, stepping up the battle by improving Internet Explorer in a faster pace, because clearly, things have been moving way to slow. So yay for browser wars!

  3. John P says:

    I don;’t think it’s just that people are unaware of Chrome. Another reason to offer chrome as an IE plug-in is to sell services into corporate/government environments. While you won’t be allowed to install a plugin, IT services can role it out if they want.

    For a corporate, it makes much more sense to install a plugin to access enhanced Google Apps and keep everything else as is, than to change browser wholesale — no risk to your intranet services tested only on IE6; no problem with that tool you use based on an ActiveX component; no retraining to a new interface.

    Corporates are already used to installing and managing plugins. If they’ll put up with Oracle’s JInitiator then Chrome Frame shouldn’t be problem. If Google salespeople came through the door saying their new service barely worked on IE and they must switch browser entirely it would be a different matter,

  4. Oomph says:

    John – you raise a very good point. There may be a government or corporate niche comfortable installing an IE plug-in so that they can adopt Google Apps without breaking legacy apps. Google may be trying to address exactly that market, especially in light of recent news concerning government adoption of their services.

    At the same time, that use case worries me a bit. Systems like Google Apps have the potential to make the cost/benefit case to drive upgrades from browsers like IE6. This could provide an excuse to continue to hold back on making IE6 a bad memory.

    I suppose you could make the case for adding that proprietary meta tag to as many sites as possible, to provide that niche the best experience. But I need to make sure I can turn off the “please install this plug-in” message that – as noted in the article – seems like bad form for a marketing website.

  5. felipe says:

    Just adding the tag will not show the message, actually, it will do nothing to any user, unless he has IE with GCF installed, which then it would render on WebKit.

    So, if your application is Chrome compatible, you should have no fear to add this tag, as it will only do good.

    To show the message, you have to add some JavaScript, so it’s up to you.

  6. michael says:

    Until http://www.nsslabs.com give different results about the poor performance of Chrome, I’ll be sticking with IE8.

  7. Shayd says:

    @Daniel
    RE:
    “it’s likely that Microsoft will be really annoyed with this, stepping up the battle by improving Internet Explorer in a faster pace”

    Historically, actually improving their product to make it more stable, standards compliant and or less resource intensive is an absolute LAST resort for MicroSoft. It’s far more likely that they’ll respond with scare tactics, defamatory PR and official recommendations NOT to install Chrome Frame for “security reasons” in direct mailing to IT managers at big enterprise shops to try and shut it down before end users can do a side by side comparision.

    Microsoft didn’t get where it is today by having the best products. Their strength has always been in the basic 3Ls or marketing: Lies, Lawyers and Leverage.

  8. Harry says:

    @Michael
    http://www.nsslabs.com shows that Chrome is not as good at Phishing prevention as IE8. But Chrome Frame sits INSIDE IE8 – so with ChromeFrame you get BOTH IE8 Phishing protection AND Chrome performance!!!
    Are you gonna install it now?

  9. Lin says:

    This allows them to “control” the HTML5 experience. For example, they can now use canvas tag in their app. If the user’s browser doesn’t support it (IE doesn’t support canvas, all other major ones do), the website will prompt the user to install Chrome frame. It’s still way more likely for a user to install an IE add-on than switch the browser completely.