- From: Shane McCarron <shane@aptest.com>
- Date: Tue, 21 Nov 2006 11:19:30 -0600
- To: David Dorward <david@dorward.me.uk>
- CC: www-html@w3.org, "www-html-editor@w3.org" <www-html-editor@w3.org>
David, First, I want to apologize for the delay in responding to your question. I see that some others have chimed in, and that's great! However, since I was one of the people who put C1 and C14 into the spec, I guess I should comment on this and the nature of Appendix C in general. Remember that appendix C is *informative*, not *normative*. It does not specify any requirements at all. It is a collection of suggestions based upon real world experience. Think of it as hints for creating well-formed, valid XML that should work in HTML user agents. If you have content that uses the features of XHTML described in the Appendix, using those features in the manner described should give you the best success rate in the real world. Appendix C *is* referenced from section 5.1, but this does not make the contents of Appendix C requirements. Rather, it says that you can use the text/html document type when serving XHTML 1.0 documents, and if you do following the guidelines in Appendix C will increase the possibility that your document will be processed correctly by HTML user agents - to the extent that HTML user agents ever processing anything correctly, anyway ;-). The XML PI suggestion in C1, for example, is there because certain broken user agents do stupid things when they see a PI. An old user agent on the Mac, for example, would lose its little mind. Another on Windows goes into "quirks" mode if it sees an XML processing instruction. Yet another does content sniffing, so if there are too many PIs (or even whitespace!) at the front of a document before the DOCTYPE it has no idea how to process the document. It is up to you whether these risks are something you and your users are willing to live with or not. To answer your real question though, the purpose of C14 is to demonstrate a mechanism that, if you choose to use it, can provide a pointer to your internal stylesheets that a generic XML processor will correctly interpret. If you care about having your content processed as generic XML, then you can take advantage of this feature and still continue to use internal style sheets in your documents. I would agree that this begs the question "what do I do about external stylesheets?" It is unfortunate that we did not touch on that, but the xml-stylesheet PI is certainly the correct way to reference external stylesheets as well if you need your documents to be processed by a generic XML processor. You are correct that these two suggestions are at odds with one another. This isn't intentional so much as it is a consequence of attempting to serve two very different audiences. As a content author, you need to decide which audience you are dealing with. You are also correct that having this advice in Appendix C is somewhat confusing. It is really there because it is not normative, and also because it is a hint as to how to bridge the gap between XML processors and HTML processors. Thanks for taking the time to present the issue. I think this merits an errata just to clear up the purpose of Appendix C in general. I have copied www-html-editor on this so that it will bet into the HTML Working Group issue tracking system and addressed. Thanks again! -- Shane P. McCarron Phone: +1 763 786-8160 x120 Managing Director Fax: +1 763 786-8180 ApTest Minnesota Inet: shane@aptest.com
Received on Tuesday, 21 November 2006 17:19:51 UTC