Is this thing on?

Second, the Web browser community isn't making much progress on standards relevant to the Web because it's spending all of its time fighting the larger Mobile Web community.

Is the Mobile Web Community larger than the "Web"? It all sounds quite theatrical. And isn't Opera conquering the mobile market to bring desktop technology to mobile phones? At least I have the impression...

Other than that, it really is well-written. Capitalist thinking, but still well-written. The part you quoted fits quite good in what I think about web technologies.

I don't think though, that the approach of the WHAT-WG is ideal, either.

Khm... Mozilla is doing what Netscape/AOL tells them to do. This is the way it was five years ago and the same is now. FF is like a girl from red district (: I don't like them.

I have read this post before. I have to say it's a very, very interesting reading, since I myself try to contribute to W3C mailing lists and the WHATWG one. However, I haven't yet "felt"/noticed what's being talked about in the posts that criticize W3C (I'm rather new and not very active). Thus, I can't agree nor disagree.

Being a "web standards follower" (hopefully not a zealot, but I leave others decide ) it's confusing: if the W3C is not better, if this is not what we should all follow, then what? Microsoft? Opera? Mozilla? Apple? Talking to someone who contributes a lot more than me on the W3C mailing lists (and AFAIK also went to conferences) says it's even worse (how much worse can it be?).

If what David Baron says is 100% accurate (the "fight" between working groups, mobile versus desktop, etc)... then it's really bad for the Web as a whole.

W3C is not perfect, that's for sure. However, I believe it's better than blindly following Microsoft. I am not for blindly following the W3C either, hence I contribute to WHATWG too.

Well, it's pretty much right that certain groups do not "cooperate". Not sure about the "fighting" part, though (never experienced anything like that). There are wars over CSS vs. XSL, MathML vs. others, HTML vs. XHTML and there are questions like "ActiveX is evil, now what is XPCOM?".

But I think David Baron's suggestion of a capitalist-like "see who wins" poses a great threat. It's great as long as things go the way you like, but it's not as good if they do not (and they will do both). While we're at it, why not follow M$'s example? After all, they've got the most market share. They are the industry leader. Are they? You see, it's not black and white...

I personally like XHTML2. If there was any chance M$ cranked this into their browser (properly) I would instantly switch over to XHTML2, even if Mozilla/KHTML/WebKit would not understand it (provided Opera does, of course). WHAT-WG is working against XHTML2 implementation, because it is not "backward compatible" (they, too aren't always, though). If we'd always use this as an excuse we would never go anywhere far... ("this CD does not fit my cassete cartridge, it is not backward compatible", imagine that).

WHAT-WG try to specify error processing. That is an extra set of rules to be applied to a browser. In (real) XHTML, the Browser simply fails, period. No guesswork required.

I like other things about WHAT-WG, though, so I would in no case think of abondoning it.

About standards and what to follow... there is a related blog post by hallvors about specs violation, real world circumstances and error-following.

I don't think there is a definite answer, but imho it would be sufficient if noone used ActiveX, VBScript, XUL, XPCOM, (XBL?) and the like. ActiveX and VBScript are met (rather) seldom these days, but XUL becomes more and more a problem for Opera (and all that because the Mozilla guys allow crome stuff in web content).

_Grey_:

Now that you remembered me, yes. I have seen a huge thread about MathML versus "the others" (it was more like "go kill MathML!!! i hate it!!!"). The thread reached hilarious levels .

Backwards compatibility is not that important if in less than 2 years major UAs implement the new technology (say XHTML2). The problem here is that MSIE does not implement anything new. So backwards compatibility becomes important: one cannot ignore IE given its market share.

I'm for XBL2. This is no longer proprietary to Mozilla. XBL2 would be a great addition to web applications.

I like Web Forms 2 too.

@robodesign:

I'm also for Web Forms 2. One of the things I like about WHAT-WG. Second is canvas.

Not sure about XBL, though (that's why I put it in brackets), but for now I think it should not be used at all (or very responsible, at most). I must admit that I don't entirely understand XBL, so I can't say anything special about it (apart from noting that I dislike -moz-binding and behaviour [IE] properties in CSS files...) until Peter Paul Koch takes the time to review it

I'm largely on White Lynx's side when it comes to MathML (he's posted in this forum and on some W3C-like Forum I got links to). It's likely that he started the thread you mentioned . Don't remember such a thread, though. I've seen posts of him quite aggressive, but I've also seen a thread were he politely discusses with some of the heads behind MathML on this forum (if my memory is correct).

The WHAT-WG-side seems safe so far. So until they screw things up we should follow them

@_Grey_:

Yes, canvas is also a good thing coming from WHATWG.

I myself understand XBL2 and probably this is why I like it. Maybe you saw I have reviewed the spec (my name on the mailing list is Mihai). I do not wait for other people to review a spec to be able to say I like it or not.

As for the thread I was talking ... it wasn't White Lynx . His name was Juan and the CSS-based approach was one of the alternatives to MathML he liked. White Lynx posted some replies. He was polite in my opinion. I stopped following the thread after about 20 emails, hehe . I even posted myself 1-2 replies (more than enough for me).

Honestly, I am not 100% sure on which side I am. Why? If I would've been, I would've had made my statement clearly on the thread. I haven't used any of the approaches. I tend to believe that working math for the Web with CSS seems more reasonable. MathML is better for processing.

Both approaches should be available and working. Each with his own needs. I can't see the CSS-based approach excluding/making obsolete/useless the MathML approach (nor vice-versa). IIRC someone even said these are orthogonal approaches.

wow