Make IE5, 6, and 7 standards compliant!?

Sounds like I'm halucinating, but read on

[swa66]

I'm stunned at something I found over at google's code collection. It's a bit of javascript that seems to be designed to make our life a lot easier by making IE 5 and 6 either act like IE7, or even much better: by making IE5, 6 and 7 act like a standards compliant browser.

I hope the URL is OK enough to be allowed (otherwise you'll have to search for it):

[code.google.com...]

Take a look at the examples.

I'm really starting to like the idea of using 3 lines and get rid of many of the headaches IE is causing me every time I start with editing a CSS file.

Moreover if Microsoft ever does make a IE8 that actually will pass the acid tests, then I'm all ready for it.

While I do worry about not using JS for those with JS turned off, I'm pretty confident a vast majority of my IE users have it turned on (most would not know how to turn it off anyway).

The MIT license seems more than OK.

I understand it's not fully CSS, but it's seems like something that could solve a lot of the questions in this group.

Perhaps I need to use it and get back with my feet on the ground, but the examples convinced me it's something I'll try real soon and I didn't want not to share this find.

Has anybody been using it in production?
Results?
Why not?
...

Dreaming of a life without IE "features" and flat out bugs ...

[suwanneeredhead]

Interesting... looks like they have a discussion forum ...

i'm going to read up on that... that's where most of my problems are right now...

[edited by: SuzyUK at 10:33 am (utc) on Feb. 3, 2008]

[SuzyUK]

This is an update to Dean Edwards original IE7.js [dean.edwards.name]. Originally developed when there was to be NO IE7 it was fairly stable. The project has now moved and is in the process of being updated. The IE7.js part used to make IE5 and 6 emulate compliant browsers, it now only "upgrades" IE5 and 6 to act like the real IE7 and there is now an IE8.js in development, I presume this is intended to fill the gap between what was added to MSIE7 and what still hasn't.

It seems that this betas are indeed that, the group discussion shows very recent bugs/patches (Feb 2nd) - just be aware that it's not quite the plug and play it might seem, yet.. though he's welcoming testers

..fun to use for non critical enhancements, but I'm not sure I would depend on it for crucial application layout choices, he's a clever chappie that Dean Edwards :)

[edited by: SuzyUK at 11:00 am (utc) on Feb. 3, 2008]

[Xapti]

I've heard of IE7.js by Dean Edwards, like SuzyUK said, but what's the difference between this and the old one if it doesn't make IE7 (the browser) support more than it currently supports? Perhaps some minor updates on IE6 or IE5 fixes may exist or something?

Like since you (suzyUK) mentioned IE8.js, why isn't that solely being worked on, instead of this updated IE7.js? (partly why I ask, is because I don't know what the latest IE7.js does)

Anyways, the main thing I'm bugged at, is that IE7(brwsr) still doesn't support :before and :after pseudo-elements. Not only that, but the Dean Edwards IE7.js doesn't render it the same as in firefox (or possibly other browsers, haven't checked). For instance, I think the font color wasn't kept when I tried using a character as a bullet (that's a feature that should be incorporated into CSS's list style type), and other more minor problems existed too, which are harder to explain.

[edited by: Xapti at 4:43 am (utc) on Feb. 4, 2008]

[swa66]

It seems they are working on the IE8 script more than the IE7 one, but I guess the IE focussed community will be happy to get IE7's half-way solution.
From my limited tests, it's unfortunately far for ready for production quality.

Still fixing a bunch of errors that should have taken the folks in Redmond a few minutes of their time seems a worthy goal if you ask me: The only way out of the current mess we're in is to get IE (6, 7 and 8 when/if it comes around) to support standards compliant stuff only.

As it's clear that MSFT isn't going to bother ever to fix IE 6 and 7 and that they also seem incapable of convincing their customers to upgrade away from IE6 (at least not among my visitors). That only leaves hope for those seem unwilling users to upgrade to IE8 of better. IE8 might be a solution -provided they do the right thing-, but it'll take decades to get it in the hadns of most of their customers.

So I think this project -or something like this- has a much bigger chance of medium term solution than anything else.

As it is, I'm *really* fed up with things like the peekaboo bug, doubling margins, extra whitespace, and all the rest of the junk.
Offer me a foolproof .js to add in a conditional comment and make it go away: I'll sign, even if I have to pay for a license to use it (but it must take away the current mess).

[lavazza]

Thanks for the heads-up Xapti

I notice that the '1 line, no-whitespace' version of the file weighs in at 31kB

I'm wondering... what are the bandwidth overheads with [conditionally] linking to such a file?

Will (for e.g.) an IE6 browser cache the file?

If so, will that cache persist (per session) and be 'available' (locally) for multiple domains?

If not, it seems (to me) that the cost seems to outweigh the benefit... or is it just that I have no idea what I'm talking about?

[swa66]

I'd honestly not care too much about the overhead. IE is slow in rendering anyway. And using the script is a way to cast your vote to try to stop MSFT from continuing to say "but people have fixes that deviate from the standard, so we can't go to the standard anymore". Also if we all make IE a bit slower than other browsers they start to have reasons to go for standards only.

Sure, it'll take a few seconds on a dial-up line, but if the rest of the site is lean and mean that should outweigh against a single image just as well.

[lavazza]

I live in a part of the whirled where NO-ONE has real broadband, almost half of all connections are via dial-up, and IE is still - wrist-slashing-sadly - the dominant player

So, for me, coding relative slowness into a site ain't exactly a high priority

using the script is a way to cast your vote

How will votes be counted?

they start to have reasons to go for standards only

They start to have reasons?

[swa66]

using the script is a way to cast your vote

How will votes be counted?

As far as I know and from reaction of MSFT: they monitor what we write, every time we use e.g. a hack they see it as a reason to support that hack in their next browser.

they start to have reasons to go for standards only

They start to have reasons?

They are thinking out loud to not make IE8 standards compliant on it's own, they want a non-standard tag in the html to make it act standards compliant.
But one of the keywords in my statement was "Standards only": Not support anything beyond the standard, a goal none of the browsers have today. Yet it is critical to get rid of: "it works in browser X but not in Y"
Moreover most if not all of the failed attempts at adding things outside of the standards (think VML, activeX, ...) are nothing but sources of security problems as the only ones interested in putting their time in something that at the very best will only work in IE seems to be the bad guys. (After all, they need but one hole in many systems to be successful in what they do)