Microsoft.com Makes IE8 Incompatibility List 358
nickull writes "Microsoft is tracking incompatible Web sites for its upcoming Internet Explorer 8 browser and has posted a list that now contains about 2,400 names — including Microsoft.com. Apparently, even though Microsoft's IE8 team is doing the 'right' thing by finally making IE more standards-compliant, they are risking 'breaking the Web' because the vast majority of Web sites are still written to work correctly with previous, non-standards-compliant versions of IE."
Where's the story? (Score:5, Informative)
I don't get it. Why is everyone so surprised by this? Microsoft has been the biggest consumer of their own non-standard web technologies in both an effort to tie services to Windows and to convince other web developers to use their 'neato' technologies.
Has no one ever noticed that Microsoft.com had various effects, direct system access, and other features not found anywhere else on the web? Or that Windows Update only worked through Internet Explorer? Microsoft WANTS to be as non-standard as possible. And if you don't believe me, check out this wonderful document [annevankesteren.nl] penned by none other than Bill Gates himself:
MS made their own internet (standards) (Score:5, Informative)
About ten years ago, as Web-1.0 was beginning, I decided to learn to write HTML for a personal website. At that time, MS released a beta program (I forget its name) to automate HTML authoring and I signed up, downloaded and installed it. Then I found its output while great for IE, did not render pages well in Netscape or even Opera. So I uninstalled it and wrote with WordPerfect-7, correcting the code by hand.
Some weeks later, MS emailed me (the beta program, of course, required registration with an email address) with a special offer: a free year-long subscription to an upcoming MS magazine if I would document my use of a feature on my home web page that worked under IE but not under Netscape -- that is, I would get a worthless pile of MS propaganda every month if I would break web standards to the benefit of IE.
It was always MS' plan to dominate ("embrace and extend" was what is was called then) the internet.
I believe if there was one event that caused them to change their minds and become web-standard compliant it was their losing fight with the EU monopoly courts and their punishment: to become standards-compliant with respect to APIs, networking and, apparently, at least in MS' mind, the internet as well.
Perhpas MS could take a feature from the Opera browser -- user agent spoofing, and let IE-8 users impersonate another brand so they can view standards-compliant sites as the designer intended them to be seen.
Re:Google.com?! (Score:5, Informative)
http://validator.w3.org/check?uri=http://www.google.com/&charset=(detect+automatically)&doctype=Inline&group=0 [w3.org]
Re:Where's the story? (Score:4, Informative)
http://www.iol.ie/~locka/mozilla/plugin.htm [www.iol.ie]
Lord only knows why that even exists...
Re:Breaking IE-specific sites is a GOOD thing (Score:1, Informative)
Easy fix (Score:5, Informative)
The problem is that many sites will check if the browser is IE, and then do various workarounds. So Microsoft is stuck: they can fix the browser, but then the sites have to be modified to say (if browser is IE, but version 7 then do the hack)
I think the only good workaround would be for Microsoft to change their user/agent string so it reports itself as Firefox :)
The list seems accurate (Score:3, Informative)
I've checked the main page at a few of them including:
tom.com
qq.com
mozilla.com
google.com
wikipedia.org
They seem to either:
1) Fail w3c [x]html standards
2) Fail w3c css standards
Google's rarely been standards compliant, failing to publish doctypes. Even if they did, many of their pages are built with javascript which do not create w3c-valid documents either. (But that goes for most javascript toolkits.)
Mozilla uses several "-moz" prefixed CSS attributes that are not w3c either. Even Wikipedia has a minor CSS error.
Comparing websites to a standard depends on the standard. Microsoft doesn't have to write or test IE8 to the W3C's standards, but it would be great if they did. How many of the mainstream browsers even pass the ACID tests (v2 & 3)?
I think that microsoft.com being on the list shows a changing side to Microsoft. They may never be the friend of free and open source software, but everyone would appreciate Microsoft adhering to an open and popular standard. Of course they will always have their own quirks and extras beyond any standard, but raw web development could become pleasant again.
Re:Firefox and Opera work 99.9% 0f the time (Score:2, Informative)
MSIE need to change it's UserAgent ID and CC-rules. Name themselves something new and be done with that. Simple as that.
Re:Where's the story? (Score:4, Informative)
Look, you can talk about bad security all you want, but the only difference between ActiveX and an xpcom plugin in firefox is that ActiveX would auto-install. Other than the fact that IE allows/allowed for auto installs, the two technologies are practically identical.
The problem is not ActiveX, its that IE would automatically install them. Then they made it prompt by default (it was always an option) before installing, but most users blindly click whatever they think will get them the free prize. Then they started with the unsigned warnings, but nothing was signed initially, so that was useless for a while, which again trained users to ignore it. Of course the fact that signed doesn't mean it wasn't signed by a bad guy, and since no own really does anything to the bad guys, they just make sure they are signed and go on.
I could list probably 20 things that could be changed that would have made ActiveX components not a threat, and none of those changes would actually involve changing an ActiveX component or the API in any way.
If you prevent IE from installing ActiveX components on its own you are functionally equivalent to Firefox. That doesn't mean that you can't be exploited via a bug in the browser which allows for an unauthorized install, nor does it protect you from installed components that have exploits which have not yet be found. Those problems effect Firefox as well.
Make a way for Firefox to have a page automatically install an extension and you've got the exact same problem.
Note: I pick on Firefox here because I've developed plugins for Firefox and IE. I do not use Opera, nor do I have any experience with Chrome plugins so I really can't comment as to how they may handle things differently.
Also I'm not saying you should use IE or that ActiveX is great. IE and ActiveX are crap for to many reasons to list here, but many of those same problems apply to the Firefox/XULRunner/XPCOM world as well, fortunately they just tend to be something you can fix in OSS software which takes away a lot of the validity of developers bitching about bad code, they should just fix it themselves :)
Re:The list seems accurate (Score:1, Informative)
Re:Google.com?! (Score:4, Informative)
Re:Google.com?! (Score:2, Informative)
For starters, no !DOCTYPE.
It does have a doctype: {!doctype html}
It may look unfamiliar to you because that's the HTML 5 doctype.
Re:Options (Score:3, Informative)
Re:Where's the story? (Score:3, Informative)
Look, you can talk about bad security all you want, but the only difference between ActiveX and an xpcom plugin in firefox is that ActiveX would auto-install. Other than the fact that IE allows/allowed for auto installs, the two technologies are practically identical.
The problem is not ActiveX, its that IE would automatically install them.
No, the problem is that Microsoft promoted ActiveX as a way for web developers to add extra functionality to their sites. Their goals were to compete with Java and lock users and web developers into IE and Windows. The auto-install bit is only a side effect of that, so that the user experience would be as seamless as possible, security be damned.
So now there are still lots of sites (especially on large intranets) that require ActiveX for some business-essential functionality. I guess Microsoft succeeded to some degree in that these companies can never move from IE.
By contrast, Mozilla never promoted XPCom extensions as something that should be added to a website. In fact by default you can no longer install an extension from anywhere besides the official Mozilla Add-Ons -- Mozilla intentionally makes it more difficult to do this.
So yes, from a technical perspective they both allow native code to be executed on the system but the point is that Microsoft wanted everyone to use ActiveX as widely as possible instead of reserving it only as a means of adding functionality to the browser itself.
Re:Options (Score:2, Informative)
So no one setup the GPOs to have Office 2007 save in "compatible mode" by default?
Re:Google.com?! (Score:2, Informative)
That reminds me of this article on CSS sprites
http://www.websiteoptimization.com/speed/tweak/css-sprites/ [websiteoptimization.com]
The idea is that you groupt a bunch of small images into one large one and use background-position to select the right one.