- From: Jeff Sonstein <jeffs@it.rit.edu>
- Date: Wed, 5 Mar 2008 14:06:35 -0500
- To: "Sean Owen" <srowen@google.com>
- Cc: Mobile Web Best Practices Working Group WG <public-bpwg@w3.org>
On Mar 5, 2008, at 12:33 PM, Sean Owen wrote: > I think we agree then. I suppose I'm saying that XML is how you get > well-formedness and that is without question important. What I am > referring to is the fact that XHTML imposes more in its schema/DTD > beyond well-formedness, like, "<span> must appear in <p>" or things > like that. what can appear inside of what is an integral part of wellformedness [sounds of English teachers rolling in graves] > Could some of that have been removed, and would it have > hurt its goals, and would that have left more valid documents out > there, and would that have been a net win? Because right now few > people are apparently following the rules. If I'm opening any can of > worms, it's that one. IMHO it is a good "can" to open from time to time I'd bet what you were thinking of was *simplification* of the standard and I'd also bet that most of the problems encountered in test documents would not have been helped by an even simpler sub-set of XHTML > One more example from mobileOK: there are three ways to specify a > character encoding in an XHTML doc: HTTP header, XML header, <meta> > tag. here the issue is *who* is asserting things... the server may be set to say it is one thing in the response headers and the content creator may want to over-ride that in the content itself putting the XML declaration in the beginning of the file itself like this: <?xml version="1.0" encoding="utf-8"?> is frankly quite often *not* done... more often the author uses the meta tag like this: <meta http-equiv="content-type" content="application/xhtml+xml; charset=utf-8" /> to insure that the browser understands what is going on (even if the server returns bogus data in the HTTP response header) and many times the author fails to include an XML declaration as a proper preamble to the document so really what we have in this example is two methods by which the content creator can over-ride the encoding claims of the server in the response header(s) > I am sorry, this is a tangent... maybe not... this sort of discussion about XHTML-Mobile ends up meaning things like having a script node in the head or letting an element use an event attribute [two very common practices] might no longer make the checker freak out for example all of us do ham-fisted things when coding and we fail to rapidly lose old patterns of work... me no less than others I keep forgetting and end up with content-type declarations that say it is "text/html" instead of the recommended "application/ xhtml+xml" for example... oh well <grin/> I guess my point is that talking periodically about the language as theory and the language as people really practice it is A Good Thing esp in the context of a "Best Practices" document... gotta stay rooted in reality how did I ramble on so much? sorry jeffs -- "[W]hat really happened after the U.S. invasion is that what little Iraqi state existed just fell apart in our hands, like a broken vase. And then Rummy let the shards get looted. So yes, when the Bush team says rebuilding Iraq is like rebuilding Germany, it's half right. It is like rebuilding Germany, but not post-World War II Germany. It is like rebuilding medieval, pre-modern Germany - the Germany of clans and feudal fiefs, before there was a state." - Thomas Friedman - ============ Prof. Jeff Sonstein Director, MS-IT Program http://www.it.rit.edu/~jxs/ http://ariadne.iz.net/~jeffs/ http://chw.ariadne.mobi/ http://www.xvrml.net/ http://ariadne.iz.net/~jeffs/jeffs.asc http://www.it.rit.edu/~jxs/emailDisclaimer.html
Received on Wednesday, 5 March 2008 19:06:57 UTC