- From: Anne van Kesteren <annevk@opera.com>
- Date: Tue, 05 Jan 2010 14:30:52 +0100
- To: "Silvia Pfeiffer" <silviapfeiffer1@gmail.com>, "Scheppe, Kai-Dietrich" <k.scheppe@telekom.de>
- Cc: "Aryeh Gregor" <Simetrical+w3c@gmail.com>, robert@ocallahan.org, Philip Jägenstedt <philipj@opera.com>, "Maciej Stachowiak" <mjs@apple.com>, "Tab Atkins Jr." <jackalmage@gmail.com>, "Edward O'Connor" <hober0@gmail.com>, "Jeremy Keith" <jeremy@adactio.com>, HTMLwg <public-html@w3.org>
On Tue, 05 Jan 2010 14:08:29 +0100, Silvia Pfeiffer <silviapfeiffer1@gmail.com> wrote: > I'd actually like to chip in here - I was rather offended by the > remark that authors cannot be trusted. While there are many authors > out there that really don't know what they are doing, there are many > that do. Standards are not written for those that don't know what they > are doing - those people don't care one way or another anyway. But we > write these standard for those that know what they are doing to enable > them to give good markup to the browser, which it turn enables the > browsers to do the right thing. Actually, standards are written for both audiences. That is why we have e.g. defined graceful error handling in CSS and HTML. One of the reasons Selectors does not have a parent selector is because of big performance impact it would have when used without taking that into account. And experience with Selectors (especially the more complex Level 3) shows that most authors do not take performance aspects into account. Only a couple. We have to take into account that people will do things incorrectly. It is pretty clear from content surveys that authors will do things incorrectly often in greater numbers than they do things correctly. -- Anne van Kesteren http://annevankesteren.nl/
Received on Tuesday, 5 January 2010 13:32:03 UTC