Internet interprets IE as damage and routes around it

9/23/09

Google just released a way for users of Internet Explorer to switch browsers, without switching browsers.

Nerdgasm ensues; I can’t find anyone who doesn’t think this is totally awesome.

(Actually I did find one lone contrarian view, but it’s from someone who appears to have gotten confused by the fact that “Google Chrome Frame” has the word “Frame” in it:

Framing Web pages that belong to someone else remains a practice that publishers hate and many see as copyright infringement.

…which has nothing to do with anything, but that’s the sort of crack reporting you get from InformationWeek. Emphasis on crack.)

(Edited to add: another naysayer points out that GCF — as well as Chrome itself — doesn’t use the MSAA Accessibility API. But we’re talking niche within niche within niche at this point.)

So what GCF is is a browser plugin, which users of IE6, IE7, or IE8 can download and install. Website developers can then tell the browser to render pages using the (standards-compliant and fast) WebKit engine instead of IE’s (shall we politely say “quirky?”) native rendering engine.

It’s a brilliant idea, and a pretty slick piece of technology. (The fact that it runs using an API that Microsoft originally developed years ago as an embrace-and-extend attack against Netscape is just gravy. Sweet, ironic gravy.)

But I’m not sure it really changes anything.

Yes, a lot of people (40% of them) still use IE for one reason or another. And yes, web development would be a lot easier if we didn’t have to spend the extra time accommodating IE’s rendering problems and box model errors. But I don’t see how this new plugin really makes much of a dent in that number.

People still use Internet Explorer because:

  • They don’t care, or don’t know better. The GCF plugin is simply not going to reach these people — if they don’t know the difference between a browser and a search engine, you’re going to have a tough time explaining to them even what the heck this plugin does, let alone why they would want it.
  • They can’t upgrade because they’re still running an ancient version of Windows. (Thanks to Jason Kershell for pointing this one out.) The plugin is no help here.
  • They can’t upgrade because their work depends on a legacy web program that only runs in IE (or, worse, that only runs in an old version of IE.) In theory, these folks could continue to run IE6 for their legacy web app, and install this plugin to use for the rest of their daily websurfing. But in theory they could already be doing this by just running Chrome or Firefox side-by-side with IE, and generally the reason they aren’t doing so is because:
  • They work somewhere where the IT department has standardized on IE, either because of the legacy app problem or because that’s just the way they’ve always done it, and the end users lack the privileges to install new software. Offering people a plugin they can’t install isn’t any more helpful than showing them a browser they can’t install. I guess I can imagine a scenario where some IT department needs to keep IE around for legacy apps, but sees the need to deploy a more up-to-date browser as well, and doesn’t want to confuse their users by giving them multiple browsers, so deploys the GCF plugin instead. But that doesn’t sound like the kind of scenario that affects mass numbers of people, at least not unless some killer app surfaces that can’t run in IE. (Google Wave is not that killer app, sorry.)

So — slick and brilliant bit of technology, but I don’t expect it to actually lead to anything. The only thing that will really solve the IE problem is time.

The real problem is IE6: web development for IE7 requires some extra attention but not nearly the same number of hacks and workarounds as IE6. And I find that, most of the time, a standards-based layout will work in IE8 with only a handful of tweaks.

And IE6, though milking an impressively drawn-out, operatic death scene, is dying. It’s down to 15% of users now; a year ago it was 25%. If usage drops off at the same rate that IE5 did, it’ll be completely gone by, um, the year 2013.

Hm. Maybe I should rethink this Google Chrome Frame thing.