[
Lists Home |
Date Index |
Thread Index
]
I think there is one work which says that W3C XML Schema is nothing other
than a sublanguage of regular tree grammars -- it says it is a subset
which has determinism -- a feature which might be useful for quite a few
applications. So no one is arguing that W3C XML Schema's principles are
flawed. I think the only argument is whether W3C XML Schema specification
is understandable to developers, users, and students (for some professors
etc in this list), and also whether in the complicated mess, they have had
some designs wrong. I am mentioning a reference, please refer to
http://citeseer.nj.nec.com/murata00taxonomy.html
Dare, if you do not mind, can I ask you a few things, Just say technical
features --
(a) Do you like W3C XML Schema?
(b) I presume it will be yes, so what are the features which you like
most, and how do you think each feature will be used?
(c) Have you read RELAX NG specification?
(d) I presume again it might be yes, if so, which is easier to understand?
(e) Any comments you wish to add..
I think that will help.
I will tell you my answers:
(a and b) I like the fact that W3C XML Schema is deterministic. But
determinism, they have combined single type tree grammar with
1-unambiguity, I do not like that. I like the fact that there should be
some constraint specification for data modeling, however, I think
constraint specification is open, so even though there might appear
solutions, I do not think there are well thought out solutions so far.
(c and d) I have read RELAX NG. RELAX NG can be understood, whereas W3C
XML Schema cannot be understood. Several features of W3C XML Schema affect
each other, they are not easily understood..
cheers and regards - murali.
|
