xml-dev - Re: Is XDEV useful? (was re: XDEV proposals)

[Peter Murray-Rust <peter@ursus.demon.co.uk>]

At 14:10 22/11/97 -0500, David Megginson wrote:
>Peter Murray-Rust writes:
>
> > Thanks very much David,
> > 	You pose - but do not answer - a question :-). 
>
>Perhaps as I move into my mid-30's I'm assuming the modesty and

Well, so am I (but not in the decimal system :-) Age is unimportant on
XML-DEV (except as datatype, of course :-)

[...]
>
>Each entity can have its own DOCTYPE declaration -- simply start a new
>invocation of your parser.
>
This raises a common problem I have. If I have an 'include file' (e.g. a 
chapter) I can 'include' by the following mechanisms:
	- declare it as an entity and use &chapter1; In this case it should not
have any doctypes, or other header info
	- reference it by XML-LINK="SIMPL" HREF="chapter1.xml"
	- use your NOTATION trick

The advantage of the last two is that they are standalone XML files and can
be validated independently, and so I'm leaning towards them in general.
They also have the merit that you load the TOC and then look at whatever
chapters you want. This takes less memory and is faster

The advantage of the first is you have a single object in memory which can
be searched (e.g. Xpointers). 

Any comments?

>
>The simplest approach to AF does not require an architectural engine
>at all; instead, simply look at attribute values instead of element
>type names; i.e., instead of
>
>  IF element_name = "FOO" DO
>    do_a_foo()
>  END
>
>try
>
>  IF attribute_name("MYARCH") = "FOO" DO
>    do_a_foo()
>  END

Oh dear! Like the man who didn't realise he had been using prose all his
life. This is exactly what I do for most of my stuff at present :-) It's
advantage is that it makes the DTD much more forgiving :-)

>
>
> > >Here, again, architectural forms will help.  As long as you use a DTD,
> > >and the DTD implements a "food information" base architecture, the
> > >supermarket will be able to incorporate your molecular information
> > >automatically.
> > 
> > Ah - but this is the problem. I have no idea who will use my information
> > and that is why I think that AFs are limited in my area. In Java classes,
> > for example, I can use the Date class without the authors knowing I exist.
> > I hope that others can use my Molecule class/element in the same way. 
>
>How could they possibly use your information automatically if you
>weren't using some kind of shared standard?  How would they know what
>information applied to what food, for example, unless you had somehow
>encoded that information in advance for them?

No :-) I produce something I think other people would value and just
produce it with (hopefully) good documentation. Thus I have a class
RealSquareMatrix 
in JUMBO. I may make an <!ELEMENT> out of it. I would then document it with
what I felt were ReallyUseful properties of RealSquareMatrices. If people
want to use it, they're welcome.  This is the way that we use java.* and
other classes.

So, if I produce <MOLECULE> I will document what it is, what its components
are, and then offer Molecule.java as something that will
display()/validate() it. For example, a Molecule can have Atoms but not
Bonds, but not Bonds without Atoms. If a food manufacturer reads my
documentation, they can decide for themselves whether it's useful. [I have
had interest by those involved in submission of drugs - e.g. pharmaceutical
companies and regulatory agencies.]
The users then have to satisfy themselves whether <MOLECULE> is robust,
future-proofed, etc.

In the same way I shall take <MATHML> on trust. I shall create MATHML
objects (possibly with TeX or symbolic algebra) and use them for chemistry.
The original authors of MathML need never know what I am doing (although I
have actually met some and am very excited about what they are doing).
>
	P.

Peter Murray-Rust, Director Virtual School of Molecular Sciences, domestic
net connection
VSMS http://www.nottingham.ac.uk/vsms, Virtual Hyperglossary
http://www.venus.co.uk/vhg

xml-dev: A list for W3C XML Developers. To post, mailto:xml-dev@ic.ac.uk
Archived as: http://www.lists.ic.ac.uk/hypermail/xml-dev/
To (un)subscribe, mailto:majordomo@ic.ac.uk the following message;
(un)subscribe xml-dev
To subscribe to the digests, mailto:majordomo@ic.ac.uk the following message;
subscribe xml-dev-digest
List coordinator, Henry Rzepa (mailto:rzepa@ic.ac.uk)