I agree with the article- replacing <time> with <data> seems particularly unwise, given that the HTML5 improvements are supposed to allow content providers to provide semantic sense to their documents. <data> might just as well be <stuff and things> for all the specificity it brings.
All that aside, I'm baffled that people are still amending and discussing the HTML5 spec. Criticise browser manufacturers all you want for implementing their own standards, but they don't really have a lot of choice when the standards bodies spend all their time dithering about and arguing over tiny semantics.
The HTML5 spec isn't going to be finalised until 2014. Insanity. It'll be out of date before it's even finished. HTML5 and CSS3 should have been finished long ago, and we should be talking about the next iteration of these standards by now.
Yeah, I don't see how to use <data> semantically at all. Even if it's within a microformat itemscope (which isn't required), there doesn't seem to be a standard way for the user agent to get a schema for the itemtype which would tell it that, e.g., <data class="dtstart"> contains a date. For an HTML standard, they're wasting too much time on features that can only be useful in a client/server hell of document-specific js everywhere.
Bureaucracy is the price we pay for standardization. You imply that there is an easier way that could have standardized these technologies "long ago", I'd like to hear it.
Bureaucracies don't have to be slow. Why not publish the draft and say you're going to finalize six months after that? Right now the idea of a W3C certified HTML5 standard is useless- browsers are already implementing what we have now, and people are writing code against it.
If that doesn't work for you, how about the idea of splitting this stuff up? Web browsers are now rapidly iterating through versions, why not do the same with web standards? Small changes each time. Uncouple stuff like the <canvas> tag from HTML5 as a whole. I'm not saying that it's the ideal solution, but it might make the web standards practise at least slightly relevant.
I have to agree with the article author. Sure, dates and times are really hard to get right. But to say "you know, times are really a kind of data" is a cop-out. Relevant:
What happens if we, as developers, just keep on using it?
I personally think the time element is important, and very very useful. How likely is it that the browser vendors will drop support for a new element that's already being used?
My point being, can we simply keep on using it in order to force it back into the spec?
This is a horrible idea. Detecting things like publish date is incredibly difficult. I can totally understand the aversion to wanting to generalize tags but this one is, IMHO, appropriate.
10 comments
[ 3.0 ms ] story [ 35.2 ms ] threadAll that aside, I'm baffled that people are still amending and discussing the HTML5 spec. Criticise browser manufacturers all you want for implementing their own standards, but they don't really have a lot of choice when the standards bodies spend all their time dithering about and arguing over tiny semantics.
The HTML5 spec isn't going to be finalised until 2014. Insanity. It'll be out of date before it's even finished. HTML5 and CSS3 should have been finished long ago, and we should be talking about the next iteration of these standards by now.
If that doesn't work for you, how about the idea of splitting this stuff up? Web browsers are now rapidly iterating through versions, why not do the same with web standards? Small changes each time. Uncouple stuff like the <canvas> tag from HTML5 as a whole. I'm not saying that it's the ideal solution, but it might make the web standards practise at least slightly relevant.
http://www.joelonsoftware.com/articles/fog0000000018.html
I personally think the time element is important, and very very useful. How likely is it that the browser vendors will drop support for a new element that's already being used?
My point being, can we simply keep on using it in order to force it back into the spec?
IE6 happens, I guess