Microsoft released a preview of Internet Explorer 9 the other day. The world sniggered.
My favourite comment of all of the ones on the IEBlog was by someone known only as Justin": "Your browsers are bad and you should feel bad." Justin, if you send me an email I'll send you a cookie for that quote.
Don't get me wrong; I think it's great that Microsoft's finally staring to accept that their IE8 browser was obsolete before it even shipped. Their own graph shows that IE7 is around three times slower than IE8, which is in turn around five times slower than the next-slowest browser in circulation (Firefox), which is in turn at least twice as slow as Chrome. In other words, Microsoft's own graph shows an absolutely staggering performance difference between IE7 and the rest of the civilised world, and IE8 sitting all by itself in the middle. It's worth noting that while IE6 is also still a supported browser, Microsoft appears to have simply been too embarrassed to even include it.
Microsoft claims that IE9 today is about as fast as the current version of Firefox. Oh, but wait - IE9 hasn't shipped yet, and won't for ages, whereas Firefox is out in the market, systematically eating away at IE's anti-trust-derived market share along with its younger, slimmer cousins like Opera, Chrome and Safari.
Microsoft's next claim to fame is a massive improvement on their Acid3 test score, from 32/100 to... umm... 32/100. Hang on - what? You mean you've got an entirely new browser under development, guys, but you haven't fixed the core of what was broken with the previous one?
For the record, IE's performance is awful. We as professional developers accept that and dislike it, but it's only a reason to dislike Internet Explorer. The reason that anyone who's ever had to code for it hates the bloody thing with a passion is simple: it doesn't support nearly enough of the web standards that every other browser in common use does. Oh, and it leaks memory like a bloody sieve, too.
Oh, but wonderful - I can now have 96-point Gabriola text render really quickly and without jaggies. I can honestly say that of all my complaints about having to write code for Internet Explorer, that's been number one on my all-time wish list. Wow. I can die happy, now.
I'm not suggesting that Microsoft should stop development on IE. Far from it. Competition is good, right? If everyone were to adopt WebKit then we'd only be in a whole new world of hurt five years from now. What the entire development community would absolutely love to see, however, is this:
- Fix the update mechanism. Has anyone noticed that we're up to Chrome 4.xx now? No? Why not? Because the thing updates itself, that's why. Make IE do the same, and ditch support for all legacy versions. If large corporate clients want to keep using IE6 for their intranet applications, fine. Let them. But if you continue to inflict IE6 and its misbegotten children on the world because of a few large corporate clients, the rest of the world is going to despise you for it. Rightly so.
- Just make it work. We honestly don't care how fast it runs as long as it's within cooee of the others. We do care about having to write custom code to deal with the broken rendering engine, memory leaks and other nasties.
- Stop bragging about until you've done it.
To be civil (which is less than you're getting from most of the web community), but blunt nonetheless: put up or shut up.