[Bug 4475] Namespace declarations treated as attributes

http://www.w3.org/Bugs/Public/show_bug.cgi?id=4475





------- Comment #3 from chris@w3.org  2007-04-24 13:37 -------
There has been some work in this area: Vitali et. a. "Datatype- and
namespace-aware DTDs - A minimal extension". Extreme Markup 2003

http://www.mulberrytech.com/Extreme/Proceedings/html/2003/Gessa01/EML2003Gessa01.html

although its not clear that this isn't just pouring good money after bad. In
some ways  I prefer Robin Berjon's characterisation on DTD based,
namespace-unaware validators:

  "So when they say OK they really haven't checked anything, when they say
NOT OK they might be on crack, and like all namespace-unaware things
they're a dead branch of the XML tree. But feel free to use them anyway :)"

(see link below for the reasoning of which the above is a summary)
http://tech.groups.yahoo.com/group/svg-developers/message/48562

So, rather than *add* namespace processing to a DTD system, which seems hard, I
proposed instead to filter out namespace declarations, to prevent the DTD
validator thinking they are attributes. This might help RDF, or things which
use qnames in attribute values; and it will help prevent anyone still
maintaining a DTD from trying to add as many namespace declarations as they can
think of on every element and hacking in magic prefixes, with or without
parameter entities to allow the instance to define new magic prefixes.

It won't help if the qualified names are used in element names or attribute
names, but then that takes us right back to the other bug I raised, 4776 - that
the validator should have three states, valid, invalid, and "i can't tell".
http://www.w3.org/Bugs/Public/show_bug.cgi?id=4476

Received on Tuesday, 24 April 2007 13:37:37 UTC