Ladies and gentlemen, word reaches me that Microsoft are being asses again.
This time, they’re holding up the W3C process by suddenly bringing up patent issues. The HTML5 spec is rapidly changing, and if Microsoft are going to stick with reviewing for patents instead of fully committing to the standard, saying “okay, we’re going to support this and license whatever patents we need to”, that’s their decision to make.
But what’s holding Microsoft back now is the <canvas> element, which has been in the spec for a while, which is reliably implemented in three browsers on all major platforms, which is probably the most stable part of the spec (on account of being widely implemented). And which was reviewed in the previous patent review when the charter was first accepted.
Thanks to Shawn Medero (working group member and personal friend) for alerting me again to their suspect practices. With luck, he’ll write something up with novelties like sourcing and shit.
Update: Shawn posted a detailed comment to this post which consist of his distilled thoughts, rather than my reactionary thoughts to his immediate reactions to an ongoing chat. As usual, I’m “calling things like I sees ‘em”, but you’ll probably find Shawn’s comment to be the more sober material.
Just to clarify my thoughts on this.
<canvas>being out of scope of the HTML WG charter was brought up by Chris Wilson on Nov.10th 2007 at our face-to-face meeting in Boston.Then again today during our conference call and it was addressed in the post-call IRC discussion here:
http://krijnhoetmer.nl/irc-logs/html-wg/20071116#l-561
As you mention,
<canvas>is rather well implemented (even on Windows) and probably the most interoperable part of the spec but that doesn’t necessarily mean it is in the scope of the charter.So… the charter:
http://www.w3.org/2007/03/HTML-WG-charter.html#deliverables
There are three deliverables that seem to cover:
I think every major GUI toolkit has something like
<canvas>(though they don’t have something as extreme as SVG)… so its inclusion here makes sense. HTML 5 does set out to clearly define all sorts of error handling for parsers but it also attempts to make it more of an application API.What I find ’suspect’:
<canvas>was always going to be part of the HTML 5 inherited by the W3C’s HTML WG.On the flip-side, Chris has well stated that MS doesn’t view his co-chair position as the best use of his time:
http://blogs.msdn.com/cwilso/archive/2007/01/10/you-me-and-the-w3c-aka-reinventing-html.aspx
So who knows… I’m not accusing anyone nor do I have specific evidence to do so. Nor do I really want these things - I (and everyone else should) want Microsoft at the table of one the most important specs the W3C maintains. If you are Microsoft, given Eolas, you have to review review review the patent situations… it is just not clear to me how they didn’t do this already. And the timing stinks because we are just starting to gain momentum in the working group (with real world deliverables anyway…)
By Shawn Medero · 2007.11.16 19:51
It is also worth point out that I have nothing at stake with regards to lawsuits and the HTML 5 spec… so my observations should be taken with that context in mind. (I’m just an individual contributor and not representing a corporate entity at this time.)
By Shawn Medero · 2007.11.16 20:59