- From: David Woolley <david@djwhome.demon.co.uk>
- Date: Tue, 13 May 2003 07:22:05 +0100 (BST)
- To: www-html@w3.org
> Unfortunately, many content developers do not have access to, or are not > allowed to alter, the master style sheet(s) of a site. They may not even see > a document's <head> element. This is typical of many large corporations. That's a travesty of the whole original concept of HTML: to allow the individuals creating the information to author their own pages, by making the language simple enough. HTML, although it has degraded into another desk top publishing languages, was supposed to differ from the existing ones by being so simple that anyone could use it and so universal that any hardware or OS could display it usefully. I see XHTML2 as attempting to regain that ideal with those wanting a poor man's PDF continuing to use HTML 4.01 Transitional > Often, a developer must resort to a little style attribute here and there. Two possibilities here: - either they are violating a house style policy and the commercial factors that resulted in central control of styling would dictate that they should not be allowed to; - the content management system and/or master style sheet should allow them to do so. One approach is to make sure that the master style sheet covers the semantic requirements that force this need for styling. This is the best, because it prevents the capricious use of styling. Another is for the content management to accept some form of simplified styling language (if you must, inline style attributes, although you then have to validate it to remove all the styling that would compromise the house style rules) and merge them into the head element of the compounded document. > The style attribute does no harm. Although I rarely use, I do see a need for > it to remain in the specification. The style attribute is responsible for the heavy bloat of most commercial web pages, and although it doesn't prevent <span class="red">, at least makes it a little easier to replace presentational detail markup with semantic markup.
Received on Tuesday, 13 May 2003 02:22:09 UTC