Dustin, I totally agree with you that the current system is whack. Glacial adoption of W3C standards has a real cost on developers and users, and ends up with rounded corners taking 16 lines of CSS, and gradient fills taking 8.
I don't think that we'll be better off just by making the current process "go faster," though.
There needs to be some proper vetting of features. You'd want to avoid the tyranny of the monied interests. It can't just be Apple/Microsoft/Google making the decisions about what goes in and what doesn't -- even though they're going to shoulder the cost on the user-agent side to implement the feature.
It might just be as simple as leaving the current ecosystem as it is, looking at the adoption rate and success of a given construct, and folding those into the "living" spec (without the prefix, of course) -- and just do this process on a regular, fairly frequent interval, like every 6 months.
I disagree about the proper vetting of features being a necessity. As long as open-source rendering engines like Webkit or Gecko are pushing forward and providing ideas and implementations of those ideas, we can let web developers vote with their websites.
Everyone understands that advanced features of CSS can't be relied upon—this is now part of the collective front-end developer consciousness.
Get rid of the W3C and let Mozilla and the Webkit project define the web moving forward; they've done a great job so far.
> Get rid of the W3C and let Mozilla and the Webkit project define the web moving forward; they've done a great job so far.
The standardization of required codecs for the `<video>` element is indeed a wonderful accomplishment by the non-W3C HTML5 WHATWG.
Seriously, a third-party neutral organization will always be needed. Who will you appeal to when the main companies start having different ideas about something? Another thing that organizations like W3C or IETF do is to provide some kind of shelter from patents and similar problems.
We can discuss about the efficacy of such shelters and about the ability to mediate between different options, but the web is definitely under a better situation than wireless telephony, a field where company-run consortiums decide standards and "settle" disputes.
Who will you appeal to when the main companies start having different ideas about something?
No one. If the browsers refuse to agree on something (codecs being a good example), there's nothing that anyone can do about it. Just say "no standard is possible in this area" and move on.
OTOH, if two or more major browsers agree on the syntax and semantics of a feature, let's issue it as a standard (and de-prefix it) ASAP.
Speaking of `<video>`, i came across a site today that used <embed src="foo.wav">. As long as video and audio formats are loosely specified, we could've used the existing syntax and existing browser infrastructure (you still can't disable HTML5 audio/video in Chrome[1] for e.g. data cap reasons)
The irony is that <embed> was never a standard (unlike <object>); if you believe the W3C fans then you must be evil to use it. But it works, and has always worked, better than any of the alternatives.
I hear with html5 they've given up and declared it part of the standard.
The "they" that "gave up" is a different group of people than the "they" who thought it was evil, by the way.
The current group working on the contemporary HTML living standard just describes reality, by and large, so there was never really any doubt that "embed" would be part of the standard when we started back in 2004.
I think you've hit it with watching the adoption rate. It's not as if Microsoft or Apple or Google can force these features on to people. Web developers have to actually use them.
I fail to see how "monied interests" can affect much unless developers who are very much not part of these interests actually use these features. I think the way things are going (i.e., engines add features, developers use them or don't, the ones that get used get approved by de facto adoption, etc.) is perfect and fits the best with the pace at which things are moving in this industry.
I don't think, as many do, that it will lead to the proliferation of a bunch of confusing, bloated features being shoved into the spec (as loose as that word can be applied here). We haven't seen much of that yet, and I really don't think we will.
If you don't mind - I'm missing something here I think - does the W3C have any legal power whatsoever to force vendors who want to ignore them and develop proprietary closed features to fall into line? I don't believe they do, other than just some bad PR.....
I don't think that we'll be better off just by making the current process "go faster," though.
There needs to be some proper vetting of features. You'd want to avoid the tyranny of the monied interests. It can't just be Apple/Microsoft/Google making the decisions about what goes in and what doesn't -- even though they're going to shoulder the cost on the user-agent side to implement the feature.
It might just be as simple as leaving the current ecosystem as it is, looking at the adoption rate and success of a given construct, and folding those into the "living" spec (without the prefix, of course) -- and just do this process on a regular, fairly frequent interval, like every 6 months.