Dr Sperberg-McQueen,

At 11:45 PM 4/26/2009, you wrote:
>On 21 Apr 2009, at 16:48 , Wendell Piez wrote:
>>I take it the recipient of the data knows that by requiring XML with
>>only single quotes used as attribute value delimiters, they are
>>asking for XML-that-is-not-XML?
>
>Well, that's kind of harsh, isn't it?  XML does allow single quotes to
>be used, so <a x='y'/> is every bit as much a conforming XML document
>as <a x="y"/>.

Yes, I suppose it's a bit harsh, but only a bit. Maybe not 
XML-that-is-not-XML, but hyper-XML.

>If memory serves, ISO 8879 referred to rules of this kind as
>"application conventions".  So they have always been part of the
>story of generic markup.

Indeed, and I recognize the usefulness of application conventions (I 
hope my Balisage paper gets accepted); in fact I advocate them all the time.

But there's also a big difference between such a convention with 
respect to application semantics (where indeed such conventions are 
the grist of the mill) and such a convention that does nothing for 
semantics, but only restricts the tools that can be used.

I also acknowledge that there may be good reasons for such 
conventions (desperate hackers included), which is why I try to be 
nice (I do!). But (as you know) they should be well-controlled and 
carefully scoped, or they will bite you. The very fact that the OP 
had to ask XSL-List how to get a serializer to impose a non-standard 
syntactic restriction is an indication of something.

>Frankly, if the naming and design rules for prominent XML
>vocabularies can require the use of specific namespace prefixes
>for specific namespaces, and expect to elicit conformance instead
>of incredulous laughter, well, then I guess application conventions
>are alive and well and living among us. (I confess that my response
>was incredulous laughter, but that didn't persuade them to change
>the rules.)

Don't get me started on the confusion sown by namespaces into this 
area. (Which isn't meant as a criticism of namespaces, as I don't 
have an alternative to suggest. Just an observation.)

>There may well be SAX serializers which accept an invocation-time
>parameter to prefer single quotes when emitting attribute values
>(I've never looked so I don't know).  If there aren't, it really
>shouldn't be too very hard to write one, to pair it with a SAX
>parser, and to put together a filter that will normalize any
>XML input by emitting it with single quotes only around attributes.

Or, as I said to the OP, write yourself a hyper-XML-syntax serializer 
that does what you want using the XSLT output method="text".

Start with Evan Lenz's XML-to-string converter 
(http://www.xmlportfolio.com/xml-to-string/), which will save you 
some work. Isn't this monstrous? (See, I can be pragmatic too. :-)

Cheers,
Wendell



======================================================================
Wendell Piez                            mailto:wapiez@m...
Mulberry Technologies, Inc.                http://www.mulberrytech.com
17 West Jefferson Street                    Direct Phone: 301/315-9635
Suite 207                                          Phone: 301/315-9631
Rockville, MD  20850                                 Fax: 301/315-8285
----------------------------------------------------------------------
   Mulberry Technologies: A Consultancy Specializing in SGML and XML
======================================================================


--~------------------------------------------------------------------
XSL-List info and archive:  http://www.mulberrytech.com/xsl/xsl-list
To unsubscribe, go to: http://lists.mulberrytech.com/xsl-list/
or e-mail: <mailto:xsl-list-unsubscribe@l...>
--~--