- From: Ben 'Cerbera' Millard <cerbera@projectcerbera.com>
- Date: Sat, 4 Aug 2007 23:58:59 +0100
- To: "Philip Taylor" <philip@zaynar.demon.co.uk>, "Lachlan Hunt" <lachlan.hunt@lachy.id.au>
- Cc: "HTMLWG" <public-html@w3.org>
Philip Taylor wrote: > Ben 'Cerbera' Millard wrote: >> Lachlan Hunt wrote: >>> [2 other points] >>> * It is very widely used in reality. >> Do you have research to back up the third point? > http://canvex.lazyilluminati.com/survey/2007-07-17/analyse.cgi/attr/for > http://canvex.lazyilluminati.com/survey/2007-07-17/analyse.cgi/attr/headers > From looking at about eight thousand pages (details on the index page) > [...] So when <label> is used, two thirds of the time for="" was present. That matches what I find. But I was interested in a different point: * In my professional experience, <label for> is not "very widely used in reality" compared to how often labelling text does not use it. * Sites I have handed over to clients with correct <label for> sometimes get broken by them later on [2]. (When <label> is present with a for="" attribute it can still be broken.) * In my everyday browsing I rarely find clickable labels (<label for>). I commonly find unclickable labels (either <td> or not marked up at all). As I don't have figures for this, feel free to be skeptical. Your study, although very useful, doesn't have numbers for this particular subject. [1] <http://canvex.lazyilluminati.com/survey/2007-07-17/analyse.cgi/index> [2] <https://umbel.co.uk/Buy.aspx?id=151> -- Ben 'Cerbera' Millard Collections of Interesting Data Tables <http://sitesurgeon.co.uk/!dev/tables/readme.html>
Received on Saturday, 4 August 2007 22:59:18 UTC