- From: Gavin Nicol <gtn@ebt.com>
- Date: Thu, 8 Aug 1996 13:25:29 GMT
- To: marc@ckm.ucsf.edu
- CC: www-html@w3.org, www-style@w3.org
>Who *ever* claimed that the evolution of HTML proceeded in the "best way" >by any measure? The formal DTD came along after the usage curve had >began to grow asymptotic. Nobody has ever claimed this to my knowledge, nor do I say that anyone has. >Some publishing applications don't require the burden of mastering another >SGML application, and for those cases HTML should provide an optional entry >point to generic structural markup. HTML is a great glue language. >Complex collections will of course be marked up behind-the-server according >to much richer, application-specific DTDs, such as TEI or EAD. But the >cleanest scheme to preserve the most of that rich structural information >from documents marked up in an arbitrary DTD during the down-translation >to a delivery/presentation markup format as accessible and generic as HTML >is to let the attributes do the talking. Here I disagree. Attributes are fine, but you can do the same thing without reliance upon them. I know, I wrote an SGML->HTML conversion server... >In the future, will HTML become the DTD of Last Resort? I hope not. As I noted, it is a great, and becoming better glue language. For many people, it is enough, for others, it is hopelessly inadequate, which is the point I, and others have been trying to make for a very long time...
Received on Thursday, 8 August 1996 09:27:31 UTC