- From: Jukka K. Korpela <jkorpela@cs.tut.fi>
- Date: Fri, 25 Jul 2003 00:24:48 +0300 (EEST)
- To: www-style@w3.org
On Wed, 23 Jul 2003, Todd Fahrner wrote: > > So how should we understand the words "Developers are encouraged to > > use it > > as a default style sheet in their implementations."? > > It is naive, but obviously enough so that I don't think there's much > danger of it being taken to heart. Well, maybe the risk that developers actually implement what the specifications say is not too high, especially if the specification contains an informative appendix that contains a normative statement that is not to be taken as what it actually says. There's nothing naive in the _style_ of the statement; but the appendix as a whole is self-contradictory. So I guess is the implied naivity is in the content. The style sheet is first claimed to reflect actual browser behavior, after careful study; yet it contains strange features like odd ideas of abbr and acronym appearance that surely don't correspond to the behavior of any widespread browser. Then it is presented as something _recommended for actual use as browser's default style sheet_, as such. It's difficult to take this at face value, but do I read your lips correctly when I interpret it as a joke? > Less naively, I think it represents > an assertion that cross-UA renderings for the screen media type (at > least) *should* be more alike than not in the absence of author or user > styles. Assertion? If you say "should", you are making a recommendation of some kind. But should they be alike each other, or some of the sample style sheets in CSS specifications? Besides, why _should_ they be alike? I think the sample style sheet, if retained, should be one of the following: a) an illustration of what a style sheet _might_ be b) an approximation of what style sheet corresponds to a typical browser style sheet in some (which?) situation, as judged by actual browser behavior c) a recommendation of what a browser style sheet should be, with suitable reservations. The sample style sheets have so far presumably tried to be all of these, failing to do any of the jobs. Maybe the best approach would be to remove the sample style sheet and start creating a separate recommendation on a default browser style sheet for graphic browsers. It could actually do some good, if done well. -- Jukka "Yucca" Korpela, http://www.cs.tut.fi/~jkorpela/
Received on Thursday, 24 July 2003 17:24:54 UTC