Working on HTML5.1 | Hacker News Reader

Working on HTML5.1

w3.org

71 pointsdavecardwell23 comments

Has this diverged from WHATWG's "living" HTML spec or is it just a snapshot of it?

The WHATWG and W3C are only vaguely related. https://wiki.whatwg.org/wiki/FAQ#WHATWG_and_the_W3C_HTML_WG

They basically hate each other.

I think the WHATWG would have a nice time finishing off CSS3.

(Seriously, the W3C started writing drafts in 1999; is it ever going to be finished? Why did they start on CSS4 before finishing CSS3?)

I'm guessing it's yet another hostile fork: https://annevankesteren.nl/2016/01/film-at-11

As far as I understand, they (the W3C) do this every few months in an attempt to wrest defacto control of HTML back from WHATWG. They copy the WHATWG spec, strip out the licensing and authorship information, and publish as "their" HTML spec.

Then a few months passes, their copy falls behind the actively edited WHATWG one, so they have to fork it again. And so on, and so on.

It's really pretty pathetic.

This message by Hixie is very insightful if you like to delve into the details of WHATWG/W3C politics: http://lists.w3.org/Archives/Public/www-archive/2014Apr/0034...

Thanks; that long message was fascinating to a programmer who doesn't really follow this stuff.

That's a really interesting look at the politics between those two. Applies to some stuff I'm involved in too. Useful link, thanks very much for sharing!

The W3C prefer their specifications to be 'finished' at a certain point, and the WHATWG simply add their changes to the living spec.

So for the most part, changes get made in the WHATWG spec, they get copied to the W3C one, the W3C then stops copying additions at a certain point (say, when HTML 5.0 is considered 'complete') and then moves onto version 5.1. So version 5.1 is basically all the WHATWG additions that didn't make it into 5.0 before the spec was 'finalised', along with various changes made because the two groups don't get along with each other too well.

Can somebody in the know please summarize the major changes from 5.0 to 5.1 for every day developers? I found the changelog a little too detailed to read and grep.

It would be nice if people claiming to support web standards (Mozilla) would actually implement HTML5 already (new HTML5 input types).

Agreed. The whole concept of browsers only supporting some parts of the HTML and CSS specifications but not others (and this support differing quite significantly between browsers) is ridiculous. How about they focus on getting on the support down for major recommendations before trying to spend all that effort on supporting drafts with about two years to go before they're ready for production?

Changelog available here: https://w3c.github.io/html/changes.html

The goal isn’t perfection (which is after all the enemy of good)

Who has this much arrogance?

This isn't the 1980s there aren't a few hundred developers playing around on spare university machines here.

This spec is one of the most expensive documents in the world.

I'd be suprised if tens of billions of dollars hasn't been lost because of just HTML 5 nevermind its predecessors.

HTML 5 affects - the cost of computers: having to buy better ones as it hogs more ram and cpu - the accessibility of the internet as swathes of poorer users are cut off from websites that don't support their mobile phones or won't load on their internet speeds - the accesibility of millions more who can't browse on internet enabled devices (internet enabled TVs, unupgradeable phones, etc) - the accessibility of websites as screenreaders stop working - the huge cost of development as web developers have to skill-up - the stresses on developers and businesses who don't want to have to re-develop their work again and again and again and again - the monopolies that run in the internet as development of alternative browsers for - the backwards compatibility conundrum: we've nearly got rid of the pervasiveness of of the HTML4 only browsers - the cost to the environment (this is serious: mobile devices use some pretty horrible rare metals, they use energy and they aren't easy to dispose, so if you have to upgrade it's not good) - the list goes on ...

So the goal should pretty much be perfection, because at the costs involved good simply isn't good enough

If that is too much, then modularise the specs, delegate the work into manageable specs and if you've not got a spare billion lying around to develop it, then find it because this spec needs to be awesome.

Failing that, since HTML started: the Linux community realised that X11 isn't good enough; Microsoft canned its UI framework (or three of them?), Apple and Google brought in new Mobile UI frameworks and also, J2ME and Symbian largely disappeared: it might be time for us to consider deprecating HTML and finding a more appropriate body capable of creating a better web language.

Is there 6.0 in the works also somewhere? Given the language's importance in a future universally-connected world, it would not surprise me.

Kinda. WHATWG (who do the actual work on HTML) have dropped the version number and made HTML a living standard: https://html.spec.whatwg.org/multipage/

And this quote by Hixie: "You know what will be around ten years from now by looking at what is implemented in two browsers today. If it’s implemented in two browsers today, I can almost guarantee it’ll still be around in ten years. If it’s not, all bets are off. This has nothing to do with “living standard” vs. versioned specs, though. HTML4 has all kinds of features that aren’t in HTML anymore — for example, <object declare> and <a coords>." (http://html5doctor.com/interview-with-ian-hickson-html-edito...)

Will we still be saying "HTML5" in ten years?

The WHATWG thinks we shouldn't really be referring to it as HTML 5 now, just HTML.

https://html.spec.whatwg.org/multipage/introduction.html#is-...?

https://wiki.whatwg.org/wiki/FAQ#What_is_HTML5.3F

I think even the buzzword is a misnomer. Hint: What did people call <canvas> before "HTML5" took hold?

Is it a joke?

"HTML5 was released in 2014 as the result of a concerted effort by the W3C HTML Working Group." - Sure looks like one.

I still get along pretty well with html 3.0

From the document: HTML is a very large specification.

My translation: HTML is an evolutionary dead end, and because it's so large it won't die in our lifetime.

Reply on news.ycombinator.com