It is amazing to me how condescending Microsoft writes again in another post about the future of Internet Explorer. I linked a year and a half ago to the IE8 user agent kerfuffle, when IE General Manager Dean Hachamovitch tried to explain the web to developers in the most basic terms. Well, today they've let him out again, telling web developers what's important to them.
Specifically, the post is a preview of what's to come in Internet Explorer 9, and the scorn heaped on Microsoft and IE has no end in the comments, with a lot of them asking, no: begging, Microsoft to adopt WebKit as their rendering engine instead of continuing development on IE.
I wanted to try to address the features Hachamovitch highlights as upcoming in IE9.
Firstly, on hardware acceleration:
Specifically, we demonstrated hardware-accelerated rendering of all graphics and text in web pages, something that other browsers don’t do today.
That's not entirely true, as the iPhone has hardware acceleration for CSS effects. But even amending the above to include only desktop computer browsers, the last word today is an important qualifier. The guesses I've seen for a release date of IE9 has been some time 2011, a long time away for more agile browser developer teams to get hardware acceleration in their rendering engines too. If this even is a desirable feature; all modern desktop environments are 2d accelerated, so I'd be curious to see measurements of performance gains of browser-specific acceleration.
Also, Firefox uses the Cairo vector drawing library for its graphics, which I think uses hardware acceleration when available. I'm not sure if this is enabled across all platforms yet but I think it's safe to say it's in the works if not.
The post then goes on to talk about performance. In the WebKit Sunspider javascript performance test, IE9 in its current version scores right below all other current browsers. Meaning performance is worse. That bar diagram is worth seeing for the comedy outliers alone. The future versions of other browsers (Firefox 3.6, Chrome 4 and WebKit nightly builds) all score better than IE9, Chrome and WebKit seemingly by a factor of 2. Hardly numbers to brag about for IE9, even if they've gotten within the range of reasonable performance now.
Notice also how Hachamovitch is eager to tell us that they're not just looking at benchmarks, they're looking at performance for real-world sites. So their excuse for doing terribly in benchmarks is to refer to other benchmarks that can't be tested in an objective way. Sore loser talk.
Which brings me to the Acid3 tests. In case you're unaware, Acid3 is a browser test designed to check how well a browser conforms to a select set of web standards. In the two browsers I use regularily on my work computer, Firefox 3.5 and Opera 10 (notice they're both officially released versions, not betas), I see 93/100 and 100/100 scores respectively. Both these browsers have a little work to do to make the animation totally smooth - one of the test requirements - but the scores speak volumes alone of a development team that is concerned with following web standards.
So where is IE9 in all of this? Dean trots out a screenshot where they've reached 32/100. But instead of commenting on why, he goes on to highlight other tests that they've done better in, and again points out the flaws they see in Acid3 and how the standards they test aren't really standards at all. Sore loser talk again, if you ask me.
Bringing out smoother font renderings is just sad. Anyone who has had their work reviewed by clients still on IE6 will recognise the sinking feeling of seeing their many hours hard at work pushing pixels to get a site perfect go to waste because of that browser's horrible font rendering. IE7 and IE8 is not much better, but this is more of an OS issue so I won't use it against Internet Explorer. They're working on improving font rendering, which is good, but it feels so incredibly late in the game to improve their font rendering to acceptable, never mind pushing the limits a little.
The comments on the post speak for themselves. There's the odd positive comment, and a couple I mistook as positive before I realised their snarky sarcasm. Mostly though, it's just a lot of web developers very upset at having to support yet another perceived train wreck of a Microsoft browser, yet another thing to worry about. A site I worked on recently currently has 4 separate style sheets, one for the web and one for each IE version in use: 6, 7 and 8. This is a little excessive and could probably be optimised, but someone along the way saw enough difference between even the three Microsoft browsers to create separate style sheets instead of cluttering up the main style sheets and that is a powerful vote against IE on its own.
I'd like to leave you with a quote from one of the positive comments on the post. After a lot of rants against IE in general and the contents of the post, you have to make extra sure Microsoft doesn't get you wrong:
IE Team: Great job guys. Really. No sarcasm.
Sad state of affairs.
