Microsoft v Google: Handbags Drawn Over HTTP Standards
from: TechCrunchIT
Anyone who thinks that web authors being able to specify that browsers should interpret their pages in a standards compliant manner (this is what MS is enabling IE to do with this) is a bad thing has perhaps not done enough cross-platform web development to understand why it’s necessary.
This is not about a standards war, it’s about MS fixing their broken browsers in the only way that will work.
5 Comments
I don’t think it’s a question/complaint that the attribute is proposed it’s a matter of calling it a “standard” as you’ve done.
“…until it has gone through a formal standards body…”
http://friendfeed.com/e/5c13fd2a-9a76-006e-aa2c-966c9d0afa4d/I-m-amused-by-the-hypocrisy-of-GOOG-employees/
No, I haven’t. You must have misunderstood me, and I think you’re also misunderstanding the original topic.
The additional attribute is not a standard, and it’s not being pushed as one by Microsoft. (You do know that your quoted excerpt in the comment above is from the Google employee, right?)
The content-type header and MIME type handling are the “standards” in question. Currently IE and other browsers do NOT handle them according to the standards.
This specific issue is a fix to IE that will allow it to correctly process the content-type declarations and MIME types, while preserving the current behavior (ie: “not breaking”) pages that do not contain the attribute.
This is all about an effort by Microsoft to adhere to and support standards, and a smaller part of a much larger recent positive trend within the company. But of course there will always be people who will refuse to believe anything remotely positive regarding MS, so they will attempt to spin this into the exact opposite of what it is. Too bad.
If they wanted to adhere to standards (with this) and provide usability MS should either:
Implement something like Firefox:
http://developer.mozilla.org/en/docs/How_Mozilla_determines_MIME_Types
Or
look for assistance from the W3C and if they decide this is the proper method it should be a standard
Again they’re forcing webmasters to code uniquely for IE:
Content-type: text/plain; “No, I really meant it”
And at the same time asking other browsers to support it without–again–the help of a governing body.
I think that if you do some research on this issue you will find that:
1. ALL the major browsers have some degree of trouble with MIME types, especially when it comes to the XHTML standard.
2. They also ALL use some variation of content sniffing. The article you linked to even references the fact that Mozilla had followed IE’s lead in implementing this, so I’m not sure what you were trying to show there. (Did you read that article?)
3. There is actually a valid, pragmatic reason that most of the major browsers (IE included) do not strictly enforce this part of the standard. And it isn’t because they hate standards. It’s because (today) many pages would break if they did.
Also, you need to understand that this is not MS coming up with a different (or incorrect, inconsistent, etc.) implementation of an existing standard, which they have clearly been guilty of in the past, and I have criticized them for.
This is the exact opposite of that: it’s an attempt to allow compliance with the existing standard without breaking the many existing pages that would be broken if they switched to a strict implementation of the content-type standard (which, again no other browser does either).
The only thing you might legitimately criticize them for on this issue would be that they’re putting out a proposed solution to a common (across all major browsers) problem that COULD eventually become a standard IF adopted by the other browser manufacturers.
This is an important, fundamental difference when compared with their past history with web standards, and is a clear example of the shift that the company is making in this regard. Hopefully you can see that.
It’s all about respecting the standard and finding the best way to comply with it - NOT ignoring or implementing it incorrectly as they’ve done in the past.
If you still want to complain about this manner of “leading” or forging new ground as it relates to something that could eventually become a standard, then in order to be consistent you must have a huge issue with every AJAX web app you use today (GMail, etc.).
I say that because all (>90%) AJAX apps today make use of XmlHttpRequest. As you probably are well aware, this is an API that began its life (and existed for years) as a proprietary Microsoft implementation in IE before the other browsers adopted it. Of course, today it’s a W3C “standard”, although it hasn’t been ratified yet (still in “draft” status).
But I guess since that’s such a “bad” way of evolving a standard, it would have been better if everyone had waited until it’s an official standard before using it (so we’d still be waiting right now), and furthermore should have rejected it because it came from MS in the way that it did, eh?
PS - to more specifically address your last couple points:
1. They aren’t “forcing” webmasters to do anything. The whole point of this approach is that webmasters can continue as they do today without anything changing, but they now also have an additional option they can utilize IF they want to actually adhere more closely to this particular standard.
2. AFAIK, they aren’t asking any other browsers to support it or push for it as a standard right now, although it would be a good idea for them to do so, and they probably eventually will (see my earlier XmlHttpRequest example).
That is how good standards evolve, and I think it’s an even better situation today, since they will be more open to being flexible in their implementation as the “standard” evolves, if it goes that way.