- From: Sander Tekelenburg <st@isoc.nl>
- Date: Wed, 15 Aug 2007 17:09:12 +0200
- To: <public-html@w3.org>
At 01:26 -0400 UTC, on 2007-08-15, Gregory J. Rosmaita wrote: > aloha, sander -- ben millard taught me an invaluable moinmoin > wiki trick -- to obtain the actual document source that is > served to your UA, save the source code and search for a > text string you know to be a header, and there will be an > ID defined for it I prefer to be lazy ;) I simply switch to a User Style Sheet that renders ids within the document: [id]:after {content: " #" attr(id) " "; border: solid thin black; margin-left: 1em; font-size: 0.8em; font-weight: normal; color: black; background: #fed} I don't know how well this approach can be made to work for non-visual browsing though. [...] > as for keeping the ToC current, we could either appoint someone > to do it (harvest new pages and mount them on the ToC at least > once a day), rely on those who create pages to add them to the > ToC, or write some sort of script that scrapes for diffs at > regular intervals to add anything marked as "New" or anything > that is moved With (potentiall) some 400 editors, I think TOC generation should be automated. So the latter seems ideal, but takes work to implement (reliably). If we're going to switch to another wiki anyway, it seems more logical to me to pick one that generates an up to date TOC by itself. Dokuwiki for example does. (See "[admin] W3C Wiki migration" thread.) -- Sander Tekelenburg The Web Repair Initiative: <http://webrepair.org/>
Received on Wednesday, 15 August 2007 15:14:27 UTC