Never doubt that a small group of thoughtful, committed people can change the browser.

July 29, 2004 4:57 PM

Internet Explorer 7. Dean Edwards does what a team of developers with billions behind them apparently can't -- update IE to work with modern standards. Almost, anyway... as he says, it's still in alpha, and has its quirks, but check out the Pure CSS Menus demo, for example.
posted by weston (19 comments total)

Is the "Pure CSS Menus demo" not supposed to work in IE? Because it does... Am I missing something?
posted by techgnollogic at 5:36 PM on July 29, 2004

I guess I always took it as a given that the "team of developers with millions behind them" were breaking standards on purpose.
posted by RavinDave at 6:39 PM on July 29, 2004

It would have helped if the site provided a "before" example, because that's the "after". IE7 is essentially a mega-hack embedded in the page. Currently versions of IE are transparently fixed. You have to see the examples without the IE7 stylesheet to see what IE5/6 normally does to it.
posted by nakedcodemonkey at 6:54 PM on July 29, 2004

techgnollogic, it seems to work okay in IE for me as well, but it isn't as nice as it is in firefox, which I use as my primary browser. The difference seems to be that in IE, there is a tiny gap between the pop-down box and the "css/edge" and if I move my mouse even a bit too slowly, the pop-down disappears.

Mind you, the page also looks a lot nicer in Firefox, but that could be because I have taken the time to mess around with font sizes and whatnot in it whereas I only use IE when I absolutely have to (or to see whether a site works in it).
posted by synecdoche at 7:25 PM on July 29, 2004

Here's the before (Eric Meyer's original pure css menus demo).
posted by weston at 7:25 PM on July 29, 2004

Ah ha. Thanks guys.
posted by techgnollogic at 7:28 PM on July 29, 2004

I admire the noble diligence behind this, as well as the "why climb the mountain? because it's there!" that must drive it, but will this ever really go anywhere? There will never really be an IE7 until Microsoft says there is.

Does anyone know this- if you had access to the source code, is "fixing" IE all that hard, or do the guts need to be junked and rewritten from scratch?
posted by mkultra at 7:45 PM on July 29, 2004

Vigilante bug-fixing... neat. Whatever you might say about MS, they certainly did a decent job of making IE extensible if it can support something like this.

Now they just need to work on the whole security thing...
posted by reklaw at 7:52 PM on July 29, 2004

"dean - edwards 2004"...what could have been.
posted by republican at 8:52 PM on July 29, 2004

Has anyone ever investigated if Microsoft software (not just the browser - but IIS and Windows and the whole kit-and caboodle) is at least somewhat supported by the development and system administrator community precisely because it *is* so buggy? I mean in terms of job security and longevity. When you look at the amount of tech time spent fixing the stuff that should mostly work in the first place, it seems to be an impressive percentage.
It's been my distinct impression that some IT people remain very loyal to MS as a "tried and true standard" either because that is what they are familiar with and/or it keeps them busy. I'm not suggesting this is a major reason for MS's dominance, but I've seen it often enough that it seems more than purely coincidence.
posted by sixdifferentways at 10:30 PM on July 29, 2004

man, if dean had won the nomination and picked edwards as a running mate, that guy would, like, be cool if he had deanedwards.com
posted by delmoi at 11:43 PM on July 29, 2004

oh yes, sixdifferentways. the same goes for oracle and a few other vendors i could name. not that oracle is buggy, but it totally requires manual tweaking care and feeding. oracle could be as user friendly as sql server, but that would deprive many many oracle dbas of a living. (i like and prefer oracle as a developer, but honestly... )
posted by i_am_joe's_spleen at 11:52 PM on July 29, 2004

so what happens when IE 8 renders CSS on par with Firefox et al, and the million pages with IE box hacks and whatnot stop 'rendering' correctly?
posted by Satapher at 12:24 AM on July 30, 2004

IIRC, those box hacks usually take advantage of other IE bugs. If those bugs are corrected too, things should be fine.
posted by Tlogmer at 12:36 AM on July 30, 2004

Satapher, the "IE 7" hack presented here doesn't make that a danger, since a site developer has to make a conscious decision to use it, and would therefore know not to use the CSS "filters" anymore.

Still, you ask a good question, and I think that's why the best CSS filters utilize a browser's inability to comprehend certain selectors, rather than things like stylesheet parsing bugs. It's easy to imagine the parsing bugs getting fixed w/o fixing box model or something like that, but it's harder to imagine someone getting closer to selector standards w/o getting closer to the standard in other ways.
posted by weston at 12:41 AM on July 30, 2004

Yes, IE7 is just a mega-hack embedded in the page—but it's a mega-hack of convenience. Now my stylesheet can include things like adjacent sibling selectors, without me having to worry about putting in an individual hack for IE to get similar behaviour. And when the real IE7 comes out (har), I can just take out the three-line hack, and everything will work.

Okay, there may be some sunny optimism in the preceding paragraph, but this is still great. It's not some patch for IE that users won't bother to download, it's a style hack that web designers can make much use of.

Although a "before" example would've helped win over the naysayers, I agree. Still, anyone who knows what they're doing should understand that this is a great idea.
posted by DrJohnEvans at 5:39 AM on July 30, 2004

His hack to get mozilla to use IE-style dhtml behaviors is just as, if not more, impressive.
posted by mikeh at 8:30 AM on July 30, 2004

I can't see the job security angle sixdifferentways. Supporting MS crap is so tiring. I guess if you were the kind of person who previously would have been screwing caps onto toothpaste tubes you might want MS to continue to make buggy, "feature" rich software. Most IT people I know would much rather be soving new problems though.
posted by Mitheral at 9:54 AM on July 30, 2004

A better solution.
posted by Blue Stone at 2:37 AM on July 31, 2004