Adding CSS3 Support To IE 6, 7 and 8 With CSS3 Pie

rsk writes "Internet Explorer 6, 7 and to some extent 8 have been the bane of every CSS-loving web developer for years. With the spreading adoption of CSS3's fancier rendering effects, like rounded edges, drop shadows and linear gradients, the frustration of needing to deal with IE compatibility is growing. 327 Creative's Jason Johnston has created the CSS3 Pie library to address this. CSS3 Pie adds support for CSS3's most popular rendering techniques to Internet Explorer 6, 7 and 8 by way of the IE-only CSS property 'behavior.' CSS3 Pie is open sourced under the Apache 2 license and can be accessed from its github repository."

Comments

IE? Seriously?

[jon42689]

Honestly, while I realize that there are some people out there using IE, I almost never make it a priority in development. "Oh, it doesn't work? What browser are you using? Internet explorer? Oh, that's the issue then." Why are we trying to fix something that is broken by design and is about as closed as a nun's c**t?

Re:IE? Seriously?

[SashaMan]

The fact that this was modded insightful is even more proof of why most tech people should never run a business. "Oh, the majority of users on the internet can't use your site? Tell 'em it's their problem."

Almost never make it a priority in development

[Anonymous Coward]

Me either. But on those rare occasions when I'm not developing for unicorns, I have to consider the real world.

Re:Almost never make it a priority in development

[man_of_mr_e]

Web developers have this really weird reality perception filter. It's almost like 1984 "doublethink". We have always been at war with Flash, CSS 3 is our friend.

In Web developerland, whatever the current standard is, has always been the standard, and thus anything that doesn't conform to it is "broken". This ignores the fact that other standards existed before the current standard, and that the meaning of the standards have even changed (CSS2.1 for instance, redefines a great deal of CSS2).

Granted, IE6 is broken, but not in the way most developers seem to think, or want to claim. It had bugs, and when it was designed, the W3C had not clarified how the box model was supposed to work, and IE6's assumptions were were wrong.

However, IE6's major failing is simply that it did not evolve. People like to claim IE6 today was intentionally designed to violate standards that didn't even exist when IE6 was created (or were at best ill defined). Mozilla was likewise broken in many such ways, but they evolved and fixed their problems over time. It's like calling a car that requires leaded gasoline "broken" because all you can find is unleaded gas today. It's not broken, it's just out of date.

Yes, it's frustrating that there is this huge legacy burden on web developers, but please people, stop rewriting history. Stop forming the perception filter that turns you into conspiracy theory spouting retards with no concept of how the web actually was created. (appologies to any real conspiracy theory spouting handi-capable people reading this message, i'm an insensitive clod).

Re:Almost never make it a priority in development

[Anonymous Coward]

Sorry, that isn't the reality - its revisionist and misleading. At launch, IE6 did not support the standards in place then. Not close to completely, not correctly (even in places where no interpretation was needed), and where they could interpret - they did so poorly (as you acknowledge).

Since then, plenty of standards have finalized that were in draft, in wide use, and easily adopted by dozens (yes, dozens) of other browsers. IE did not.

I was there before IE4. Back when Microsoft was actually trying to keep up. They took a decade off, and IE6 was near the beginning of it.

Re:Almost never make it a priority in development

[99BottlesOfBeerInMyF]

People like to claim IE6 today was intentionally designed to violate standards that didn't even exist when IE6 was created (or were at best ill defined).

Seeing as Microsoft went to court and lost over intentionally plotting to break compatibility with both published standards and other browsers... I don't think it's really too big of a stretch to believe. It's the "extend, extinguish" part in it's original sense from the e-mail evidence.

Re:Almost never make it a priority in development

[99BottlesOfBeerInMyF]

You can read the internal memos they discovered during the trial yourself, where the phrase "embrace, extend, extinguish" originates. Claiming they didn't intentionally break compliance, when they say they are planning to do so in their own e-mails requires absurd bias on your part.

Re:Almost never make it a priority in development

[man_of_mr_e]

Technically, they were convicted, it was just overturned later. Of course the latter part of the case was a sham.

You're the one with "creative interpretations". One cannot be "convicted" in a civil trial. The mere terminology you use is intended to deceptively imply criminal actions. Why not try using, you know, stuff that's actually true, and you know, stuff that's actually reality rather than what you want it to be.

They were found by judge jackson to engage in browser tying (among other things, none of which had anything to do with your claim) They were *NOT* found to be deliberately breaking browser standards).

The term "embrace and extend" goes back further, the DOJ simply added the "extinguish" part, or rather one of the witnesses did, and when cross-examined that witness acknowledged that the phrase wasn't in his notes. In fact, there is no evidence the "extinguish" phrase actually existed, other than his testimony.

But, in any event...

That's a very, umm, creative interpretation. I'll let the memo stand on it's own. Anyone that wants to read it and believes your interpretation, well, they must be equally... creative. You're a sad sack.

Sad sack? Who's the one that's making things up to validate his point? That would be you.

It should be quite easy for you to quote the section of the memo that talks about violating internet standards if it truly says what you claim it does. How about a few sentences? No? Didn't think so.

Re:some people?

[bunratty]

It depends on the site. Some sites have 95% of users using IE. Others have 30% of users using IE.

Re:Pointless.

[yoyhed]

This isn't pointless - do you understand what this is? This isn't for people that use IE, it's for people that develop websites for IE. This is a Godsend for me.

Developing nice-looking websites in Chrome/Firefox/Safari/Opera is easy as PIE (pun intended), but when you want that same site to look good in IE it's a fucking nightmare. This provides some easy ways of making a site look nice in all the major browsers without a huge coding headache.

Re:Pointless.

[A Friendly Troll]

If you were intelligent enough to know how to install CSS3 Pie*, you wouldn't even use IE.
*not exactly "install" per se seeing as it's a library, not an application but you know what I mean. Hopefully.

You seem to have misunderstood what this is about... The end-user doesn't install anything. The .htc resides on the server and it's the developer who includes the library and makes it work.

To be fair, however, CSS3 Pie isn't something that you should actually use, considering that it slows down the browser massively and it just adds the ability to display useless visual cruft.

This library, on the other hand, is several orders of magnitude more useful (and I'm dead serious about it): http://code.google.com/p/ie7-js/

LET IT DIE

[v1]

the last thing we need is more people coming up with hacks to give PHBs another excuse not to leave the dark ages.

If anything, we need more of the web dev tools to make pages that are outright guaranteed not to work with IE6-7.

Re:i dont want my site to work with ie

[zim2411]

tehn the morons wont show up as much as they do

Oh the irony.

Forget CSS3, work on CSS2

[Zarel]

A lot of CSS2 features don't even work correctly in IE6 and IE7: http://en.wikipedia.org/wiki/Comparison_of_layout_engines_(Cascading_Style_Sheets)

A lot of the really useful selectors, for instance, aren't available in IE6. Not to mention min-width/max-width, and white-space:pre. And using left and right in the same rule makes IE6/IE7 ignore right. In IE6/IE7, there's plenty that goes unimplemented, like :active and :before and outline and display:table; and border-style:dotted; and vertical-align:middle; and background-position:fixed;.

These aren't obscure features no one uses, these are all features I've wanted to use while designing my webpages that are supported by every other browser that IE6 and IE7 don't support.

We should really be looking to fix those, first.

This is the best way to get people off of IE6!

[iamsure]

Designers win, because they use one design (compliant, too!), for all browsers. Users win, because everyone sees the same design/version/look.

Best yet, when you pile on a library that fixes CSS3 (this article), and one to fix the CSS box model ( http://webdesign.about.com/od/css/a/aaboxmodelhack.htm ), and then another to fix the png transparency issue ( http://code.google.com/p/ie7-js/ ), and another to add canvas support ( http://excanvas.sourceforge.net/ ), and another ...

Then you explain that everyone will see the same design (yay!), and people using older browsers will experience a VERY SLOW page load. That is why they should upgrade to a more up to date browser.