On 10/15/05, Eric Scheid <[EMAIL PROTECTED]> wrote: > > On 15/10/05 4:23 AM, "Tim Bray" <[EMAIL PROTECTED]> wrote: > > > +0 no opinion on custom-XML vs microformat (real opinion: it will > > make no difference whatsoever). > > I'm -1 on non-XML formats for special purpose machine-intended information. > > Firstly I hold a small fear that if the document format is xhtml then all > sorts of other xhtml stuff will get added to the page so it looks pretty, > navigates well with other pages, etc. That is, make it look like an html > page fully integrated in the site. >
I actually hope that just this will happen. The light of day is good for your data. There is not reason for that to interfere with parsing. The cloud of dirty old HTML casts a shadow on XHTML but XHTML is XML. It *just looks like* HTML. I know that we all know that, but it can be hard not to keep thinking of it as something less parsable. > More importantly, we don't get to specify a media type for the specific > document format, we get stuck with the media format for xhtml (of which > there is a continuing dispute). We can't rely on @rel alone, because there > might be other introspection formats out there (eg. RSD). > > <link rel="introspection" type="application/rsd+xml" ... /> > <link rel="introspection" type="application/atomserv+xml" ... /> > <link rel="introspection" type="application/xhtml+xml" ... /> > > Of these, which might you feel comfortable as thinking are definitely the > introspection document for the Atom Publishing Protocol, and not a > micro-format introspection document for some other web log publishing > protocol. > This on the other hand is a thinker. Browser integration... hmmm. - Luke
