Hans,

I can't think of a reason to do your schema in XML Schema other than,
because it is a product of the W3C, it is currently more widely known and
accepted than RELAX NG, even though, in my opinion, RELAX NG is more elegant
and builds more solidly on good theory, such as regular tree languages, for
example. [1] John Cowan's recent review of RELAX NG and XML Schema on
xml-dev is quite conclusive. [2] Two deficits for RELAX NG: it does not have
mature identity constraints (in short, keys) other than ID/IDREF, and it
doesn't have straightforward type assignment, though both these features
will no doubt be added in version 2.0, probably as independent, modular
specs. Also, I have only heard a handful of complaints about RELAX NG which
have been superficial at best (chief of which, it has not been marketed
effectively). I think RELAX NG is well-designed and based on solid computer
science. [3]

Mike

[1] http://citeseer.nj.nec.com/murata00taxonomy.html
[2] http://lists.xml.org/archives/xml-dev/200205/msg01290.html
[3] http://www.thaiopensource.com/relaxng/design.html

> -----Original Message-----
> From: Hans Fugal [mailto:[EMAIL PROTECTED]]On Behalf Of Hans
> Fugal
> Sent: Thursday, June 13, 2002 1:55 PM
> To: Michael Fitzgerald
> Subject: Re: [gdmxml] XML Schema
>
>
> Mike,
>
> Thanks for the information. I have taken a quick look at RELAX NG (or
> relaxng as I will lazily call it) and it looks very promising. I am not
> tied to XML Schema by any requirement, and as I have barely begun the
> schema itself it would be no extra effort to use relaxng if I switch
> now.
>
> According to what I've read[1] relaxng is more than up to the task. I
> am very impressed with JARV and the overall presentation of the
> specification and web site.  I am seriously considering using relaxng
> instead of xml schema. Can any of you think of a reason why I should
> not? Of course there is nothing stopping us from doing it in both schema
> languages, but that's out of the current scope of my project; I only
> want to do one for now.
>
> Hans :)
>
> 1. http://www.imc.org/ietf-xml-use/mail-archive/msg00217.html
>
> * Michael Fitzgerald ([EMAIL PROTECTED]) [Wed, 12 Jun 2002 at 17:11 -0700]
> <quote>
> > Hans et al.,
> >
> > I know you are probably constrained to define you GDM
> schema/schemata in XML
> > Schema. Are you using a tool to do this or are you doing it by hand?
> >
> > Are you familiar with RELAX NG? [1] It is an alternative schema
> language for
> > XML produced under the auspices of OASIS [2] which is highly addictive.
> > James Clark, technical lead of the working group at the W3C
> that produced
> > XML 1.0, is chair of the RELAX NG technical committee.
> >
> > I have produced a number of schemata in both XML Schema and
> RELAX NG which
> > validate against the same instance. It is an interesting exercise. In my
> > opinion, RELAX NG wins the contest hands down.
> >
> > There is a tool, DTDinst [3], that can generate a RELAX NG schema from a
> > DTD. Jing is a readily available validator written in Java. [4]
> >
> > I am not advocating that you drop XML Schema in favor of RELAX
> NG. However,
> > it would be nice to offer alternative schemata -- all in good time, of
> > course.
> >
> > Mike
> >
> > P.S. I am on the RELAX NG technical committee, yes, but I
> promote it only on
> > the basis of its virtues.
> >
> > [1] http://www.oasis-open.org/committees/relax-ng/
> > [2] http://www.oasis-open.org/
> > [3] http://www.thaiopensource.com/dtdinst/
> > [4] http://www.thaiopensource.com/relaxng/jing.html
> >
> >
> > _______________________________________________
> > gdmxml mailing list
> > [EMAIL PROTECTED]
> > http://fugal.net/cgi-bin/mailman/listinfo/gdmxml
> </quote>
>
> --
> "Everybody is talking about the weather but nobody does anything
> about it."
>         -- Mark Twain
>


_______________________________________________
gdmxml mailing list
[EMAIL PROTECTED]
http://fugal.net/cgi-bin/mailman/listinfo/gdmxml

Reply via email to