OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.


Help: OASIS Mailing Lists Help | MarkMail Help

codelist message

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]

Subject: Re: [codelist] code list validation

From: "G. Ken Holman" <gkholman@CraneSoftwrights.com>
> So my comment was:  you stated "I'm not talking
> about validation" but you didn't say what you
> *were* talking about ... can you elaborate,
> please, on the LoC interests in pointing to code lists from W3C schema?

Simply a means within the schema language to point to a codelist.  Just

It doesn't have to be from an xsd element/attribute. (But I hadn't
previously been aware that you could stick an attribute from a foreigh
namespace into an xsd schema. I should have known that, though.)   I'm not
talking about standardizing validation techniques. Yes, of course we're
interested in validation, but my guess is that validation will take care of
itself - that having a standard syntax to point to a codelist, along with a
codelist standard, will encourage the development of open source code for
codelist validation.

For any schema there are application specific semantics that a w3c-schema
validator cannot be expected to know about.  So software developed to
process a given schema has to do more than, say, run it by xmlSpy, it has to
deal with each semantic element.  The more of these that are standardized,
the easier the task.

> still need stylesheets (or some other
> transformation technology) to transform a W3C
> Schema with foreign attributes pointing to a
> genericode enumeration into a pure W3C Schema
> with W3C Schema enumerations in place of the genericode enumeration.

I don't think so. I'm guessing that the existence of even the "foreign"
attribute will help, as I mentioned above. But I would be pleased to run
this issue by the bibliographic community  (specifically the MODS listserv,
see below); however, being new to this process (OASIS standardization) I'm
not sure if I can legitimately do that -- are there confidentiality issues?
Could someone advise me?

 Take for example the LC schema we call MODS:

There are many MODS records out in the world, and there are several MODS
validation/processing software packages. In the schema, in several cases
there is a controlled list, not with inline enumerated values but instead
with a comment saying that "the value of this element (or attribute) is from
the list at http://.....)."   If there were a way for this comment to
instead be expressed syntactically in the schema  (not necessarily via the
xsd namespace, but even from an external namespace)  my guess is that would
make the software development easier.  There is a large and active MODS
listserv and (as I mentioned above) I would be pleased to run this question
by the list, the question of how useful a standard attribute of this type
would be.


[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]