- From: Charles McCathieNevile <chaals@opera.com>
- Date: Sun, 25 Jun 2006 14:14:56 -0400
- To: "Jim Ley" <jim@jibbering.com>, "Web APIs WG (public)" <public-webapi@w3.org>
On Sat, 24 Jun 2006 05:22:51 -0400, Jim Ley <jim@jibbering.com> wrote: > "Anne van Kesteren" <annevk@opera.com> >> http://www.w3.org/TR/2006/WD-xbl-20060619/#eventxbl introduces >> EventXBL.trusted... > if it can be suitably defined, then I think it should be on Event > directly Makes sense to me > if at all, however I have reservations of how it can be defined and > implemented. User agents will need to provide API's for their plugins > that can distinguish between the 2 cases of user and non-user provided > events - an Access Technology must be producing events that are trusted, > or it would hobble AT's. ... Yes. This doesn't just apply to access technologies - user agents generally use JS to extend functionality in the browser, and people are adding both trusted and untrusted extensions., although in the case of accessibility these are clearly very important, they matter to a lot of other users too. One approach would be to leave it open for now how things become trusted, which means User Agents would start out being conservative, but if authors made use of this would be under pressure to implement a way for users to declare various things trustworthy. The risk is that if user agents are not prepared to do so, a scenario arises where authors rely on things being trusted anyway, leaving blocs of users and of useful functionality out in the cold. cheers Chaals -- Charles McCathieNevile, Opera Software: Standards Group hablo español - je parle français - jeg lærer norsk chaals@opera.com Try Opera 9 ASAP! http://opera.com
Received on Sunday, 25 June 2006 12:15:08 UTC