- From: Tantek Celik <tantek@cs.stanford.edu>
- Date: Wed, 16 Aug 2000 04:05:18 -0700
- To: webmaster@richinstyle.com
- CC: "www-style@w3.org" <www-style@w3.org>
From: Matthew Brealey <webmaster@richinstyle.com> Date: Wed, Aug 16, 2000, 4:45 AM > True. But these subtleties aren't really highlighted in the current > draft. Elucidating examples for each would be nice. This is general feedback that has been given in this thread, and one I have noted as an issue. I will try to have more illustrative examples in the next draft. >> > Incidentally, the changes I propose are entirely compatible with current >> > implementations; AFAICT, only user-select: none is currently being used. >> >> Interesting. I wasn't even aware of any current implementations. Which >> implementation are you referring to? > > That of Mozilla: > > http://lxr.mozilla.org/seamonkey/source/layout/base/public/nsStyleConsts.h#80 > > (note that inclusion here is not conclusive evidence of support; see the > rest of the source if you're interested.) Unfortunately, I am not permitted to view that source, so I will take your word for it. >> > * I am starting to err on the side of automatic menu generation to cope >> > with the issues (e.g., different ways of doing menus, system-specific >> > menus; keyboard selection issues, etc.) >> >> Interesting. > > Really? > > But I do think so. Why? Two reasons: > > creating controls from a primitive element inevitably means that people > will be forced to use things that aren't platform-familiar, which is > *bad* *bad* *bad*. One would not dream of hand-coding system controls > when creating an application, and to suggest that a web page or (horror > of horrors) a 'web application' should do so is bizarre. Let me see if I can summarize this issue. The issue is the need to set a symbolic value on a set of property instead of just one property to achieve a "platform-familiar" user interface element. Is that correct? > there are quite a few issues that don't seem to be satisfactorily > resolved. > > See http://www.iarchitect.com/mshame.htm for more UI issues. I agree that that site documents a lot of good UI issues, but I'm not sure which in particular apply to the CSS3 UI draft. > PS. I am not happy with the whole selection/focus situation, but am too > lazy to think about it at present. Will probably consider at some future > date. I think we need more implementation experience to work out the remaining problems. I think the solution is getting better with each iteration though. Thanks, Tantek ---------------------------------------------------------------------------- Continue as planned. http://www.microsoft.com/mac/ie/
Received on Wednesday, 16 August 2000 09:07:11 UTC