- From: Robin Berjon <robin@berjon.com>
- Date: Wed, 25 Apr 2012 21:03:02 +0200
- To: Larry Masinter <masinter@adobe.com>
- Cc: "www-tag@w3.org" <www-tag@w3.org>
On Apr 25, 2012, at 17:48 , Larry Masinter wrote: > I see in twitter-land some flap about opera implementing webkit prefixes. It's looking like Mozilla will be doing it too. While I deplore this outcome, given the currently deployed content I can't say that I blame them — it's not as if they have much of a choice. > If browsers implemented both unprefixed and prefixed values, then site designers could just switch to unprefixed and there would be a transition plan. Finding a transition plan is easy. The more interesting part is coping with the mess in an interoperable fashion that doesn't create more harm. I think it's possible, we just need rules to process prefixes that aren't your own in a predictable manner, that supports the transition. It needs to be done relatively fast though. > Even if someone wants to implement someone else's prefix, isn't there some way of illustrating it as "non-conforming" ? Not in any way that anyone would care about. > What are the URIs for vendor prefix names? Are vendor prefixes like XML namespaces? How are they different? They have no URI. They are not like XML namespaces. The difference is the absence of indirection (there is no locally mapped prefix for prefixes) and (assuming namespaces that use some sane form of URI) no distributed authority. -- Robin Berjon - http://berjon.com/ - @robinberjon
Received on Wednesday, 25 April 2012 19:03:28 UTC