- From: Andrew Fedoniouk <news@terrainformatica.com>
- Date: Tue, 20 Dec 2005 00:14:12 -0800
- To: "Daniel Glazman" <daniel.glazman@disruptive-innovations.com>, <www-style@w3.org>
----- Original Message ----- From: "Daniel Glazman" <daniel.glazman@disruptive-innovations.com> > >> Some thoughts about: >> http://www.w3.org/TR/2005/WD-css3-layout-20051215/#tabbed >> >> Seems like CSS indeed is entering "behavioral style" area. >> >> Tabbed panels switch is a typical UI element/component and it appearead >> first >> as far as I remember in IBM's CUA documents. > > Andrew, > > See > > http://daniel.glazman.free.fr/weblog/archived/2003_01_05_glazblogarc.html#87183885 > http://daniel.glazman.free.fr/weblog/targetExample.html#general > and http://www.w3.org/Style/Examples/007/target.html > Thanks, Daniel, I saw them while ago, but this is still not a solution - it describe change of state but is silent about how. I understand that strictly speaking "behaviors" are out of current CSS scope. But as someone (Orion Adrian?) mentioned here, selectors as an entity are separate from CSS - at least they can be reused for assigning behaviors/rules. In any case it should be some mechanism allowing to specify transition of states. We've started speaking about :checked, :current, :selected, etc. without any idea who, when and how can set them. Any ideas of how such DSTL (dom state transition language) should look like? My personal experience with htmlayout where each input control and their subparts are plain styleable DOM elements shows that it can be surprisingly simple language. Sort of SQL with triggers. WHEN selector IS GETTING/LOOSING state SET ... and so on... Andrew Fedoniouk. http://terrainformatica.com
Received on Tuesday, 20 December 2005 08:14:29 UTC