- From: Douglas Rand <drand@sgi.com>
- Date: Tue, 07 Oct 1997 13:21:02 -0400
- To: "Chris Wilson (PSD)" <cwilso@MICROSOFT.com>
- CC: "'David Perrell'" <davidp@earthlink.net>, www-style@w3.org
Chris Wilson (PSD) wrote: > ship product - and we call each other on the phone, and make a blood > pact to forget legacies and both follow the same very strict DTD (stop > laughing, darn you!). But Chris, I *needed* a good laugh. I really did. :) > Would we be able to convince everyone to upgrade? Never. Not in a million years. Between HTML being edited about 70% of the time on text editors and the plethora of other editors which probably don't force a strict DTD you'd be sunk. >.. > browsers in use, and is therefore a de facto standard by a landslide. Hear, hear. Let us not forget folks that standards committees generally do *not* dictate this stuff. The general job description has more to do with codifying and regularizing existing behavior than in dictating compliance. HTML, de facto, does not follow the W3C DTD in a meaningful way. My comment is that we have to be extremely careful to maintain legacy behavior where it makes sense. Let me give you an example from my current round of work. PRE is a block tag and the DTD states that it cannot hold other blocks. <!ELEMENT PRE - - (%text)* -(%pre.exclusion)> But can I do that in our (almost unknown) browser? Not on your life. I have people doing things like: <PRE> <P>.... <P>... </PRE> Which is to say that they use it as a container to get "preformatted" text. So instead of %text, I have %body.content. I can shout till I'm blue in the face that this isn't right (and it isn't) but I will not get my users to rewrite their pages because it looks "fine in Netscape dammit." Doug -- Doug Rand drand@sgi.com Silicon Graphics/SSO http://reality.sgi.com/drand Disclaimer: These are my views, SGI's views are in 3D
Received on Tuesday, 7 October 1997 13:21:27 UTC