When Groundhog Day ends

Saturday, 17 May 2008

Joel Spolsky:

Almost every web site I visited with IE8 is broken in some way.

Isn’t it funny how that works? Because almost no website I visit with Firefox 3 is broken if it worked in Firefox 2. Ditto between recent versions of Opera, and very close to ditto between recent versions of Safari.

Maybe we’re not looking at quite as universal a principle as Joel would have you believe?

These are not web pages with errors. They are usually websites which were carefully constructed to conform to web standards. But IE 6 and IE 7 didn’t really conform to the specs, so these sites have little hacks in them [to fix the resulting problems].

That’s really the crux of the matter, isn’t it? Internet Explorer was stuck at version 6 for over six years. Six years!

Here’s a little secret about the web that you may not have noticed before: most of my years-old bookmarks are now 404. The web has a comparatively short memory. (Future generations of historians will hate us.) How many pages that are still being visited today have been entirely unchanged for even 6 years? I know that the majority of pages I visit are only months old at most, and even at those sites where the content is older than 6 years, the “chrome” in which the content is presented has generally been touched much more recently. The sort of ultra-legacy content that is the crux of Joel’s entire argument hardly exists on the web.

What about those sites that do break? The most prominent example given by Joel is Google Maps. Is that legacy content? Has it been untouched for years? Then what are the chances that it will be updated to work in the most recent version?

The reality is that almost all currently available content on the web is targetted at recent browser versions. Browser vendors are not really forced to support ultra-legacy content – except, of course, when the most recent version of a vendor’s browser has an engine so old that content created now is targetted at browser technology from half a decade ago.

For much of the time in which Internet Explorer was stuck at version 6, it had more than 90% market share (though its hold started declining quite palpably toward the end of that era). Even now, no web developer can afford to ignore it. Piling hack upon hack is the only realistic option. The bind that the Internet Explorer team finds itself in, therefore, is purely of Microsoft’s own making. They willfully neglected the browser to the point that they will not be able to do right by either the idealists or the pragmatists. Congratulations.

You made your own bed. Now you will lie in it.

(My sympathies to the people in the trenches at Microsoft who, as individuals, are trying to take the forward-facing view, and have to deal with the fall-out of the past corporate course.)