Comments on: HTML5^H: What it means to developers, standardistas, and browser vendors

By: Cyprien Noel

Cyprien Noel — Wed, 16 Feb 2011 21:47:41 +0000

I agree with previous comments, the kind of app that is built today makes it unmanageable for developers to deal directly with yet another set of browser versions. I hope more big players will start writing sophisticated platforms like GWT to abstract us away from those problems.

By: Marc DIethelm

Marc DIethelm — Mon, 14 Feb 2011 11:21:52 +0000

I currently tend to think, let WhatWG develop the cutting-edge ‘living standard’ and APIs in concert with progressive browser vendors and ‘alpha developers’. And let W3C sort out the versioning so we can have certain baselines that we know implementers either support or don’t.

By: Joeri

Joeri — Mon, 24 Jan 2011 12:20:20 +0000

I don’t agree that the release of new browsers makes things easier. It’s actually the end-of-life of old browsers that makes things easier. IE6 has become a non-issue for me since its corporate marketshare became small enough (among our customers). So, now I can use PNG, yay. But IE7 still lives on, and in relation to HTML5 it’s no better than IE6. Even after IE7 dies, IE8 won’t bring many new things to the table. It won’t be until IE8’s marketshare becomes small enough that coding to HTML5 will be simple. Until then we need fallback code, and it’s going to be painful.

Like the other poster said, we need the javascript wizards to build high-quality wrapper libraries that smooth out the HTML5 transition. There are already a few choices, like excanvas and svgweb, but many of the HTML5 features don’t yet have high-quality fallback libraries, and even when they do, it’s very difficult for newbie authors to figure out which fix-up libraries they need to include.

Let’s be careful about advocating HTML5 when in reality it’s still very hard to use most of it if you’re not on mobile.

By: Jakub Nesetril

Jakub Nesetril — Sun, 23 Jan 2011 14:44:03 +0000

The problem with versions of HTML (as I see it) is plainly summarized in your own sentence:

“Just as an iOS developer can see what 4.3 beta, we should have something similar.”

Yes – but 4.3 beta is NOT a version of a standard – it’s a version of an implementation. And thus the iOS developer can freely inspect what’s IMPLEMENTED in that version.

With interoperability, this is simply not viable. At any point in time, the implementations differ. Pushing out a standard and slapping a version number on it doesn’t bear any meaning until implementations actually deliver working code.

The only way this could be interesting to developers is if browser vendors somehow pledged / had an obligation to ship code that passess all potential test suites developed along with the standard. Currently, that’s far from realistic.

By: Ms2ger

Ms2ger — Fri, 21 Jan 2011 12:34:02 +0000

And for those who want to help, the HTML WG testing task force (http://lists.w3.org/Archives/Public/public-html-testsuite/) welcomes everyone to submit tests. Some are already online at http://w3c-test.org/. For CSS, there’s http://test.csswg.org/ and for other web specifications, you can always contact its editor.

In any case, tests really are essential for today’s standards, not only to know how interoperably they are implemented, but also to help browsers improve their implementations and fix their (unavoidable) bugs.

By: Jonathan Nieto

Jonathan Nieto — Thu, 20 Jan 2011 21:05:41 +0000

With the release of the new browsers and its clear adhesion to the standards, all will be much easier for everyone.

Intereting article!

By: Craig Webb

Craig Webb — Thu, 20 Jan 2011 19:36:40 +0000

Joe developer here. I have been following you and other JS Whizzes for months. I built a website on the “It’s ready to go now” HTNL5 ticket. The problem that I found is that the out-in-front developers do not back-up their wonderful technology explorations with documentation for the plodding work that is IE work-a-rounds. The long-tail of documentation isn’t there, and when it is, it is written by PC/IE people instead.

The solution that I have found is in incessantly following people like you (and the people who follow you via conferences) on Twitter. The #Fail on that is that JS experts often do not follow us joe-developers back, so when I post a question on Twitter, it goes into the ether, unanswered. What we need are forums to ask and get help, and these forums need to be touted within #conferences and on Twitter, so we know here to go.

Us Joe-developers are *grateful* for your out-lying leadership and we are happy to hack along, adding our own contributions, but as you march ahead, keep an ear on the band behind you.