[Ex-plain] Extensibility

Mike Taylor mike at tecc.co.uk
Tue Mar 26 13:26:16 CET 2002


> Date: Tue, 26 Mar 2002 12:13:55 +0000 (GMT)
> From: Robert Sanderson <azaroth at liverpool.ac.uk>
> 
> > > What I thought on when Ian mentioned this was something like the
> > > "AdditionalInfo" structures that are scattered liberally around the
> > > Z39.50 spec. I find these tremendously elegant -- a place to hook in
> > Isn't that exactly the promise of XML namespaces?
> 
> Yes.

Cool.

> > Does this also work in SGML?  If not, does that fact account for
> > Rob's lack of enthusiasm for it?
> 
> I don't mind XMLNS, but I think that records returned via
> elementsets F and B shouldn't have them so as to prevent parser
> confusion.  If you know that you won't have to deal with extensions
> unless you step outside of the known world, this will make it much
> easier on client writers, I feel.

At first I thought, that sounds reasonable to me.  Then I realised it
means that anyone who wants to add a (namespaced) extension to an
Explain-- record is only allowed to do so when that record is
requested with an element-set name other than "f".  That I'm not so
keen on.  (Also, isn't there at least the implication in the standard
that "f" returns all available elements?)

> SGML had namespaces long before XML did. CONCUR allows you to do
> thinks like: <(explain)map attributeSet="BIB1"> See the Gentle
> Introduction to SGML from the TEI Guidelines (TEI P3)

Ah, OK.  So it had namespaces, but not the same namespaces that XML
has.  So if any server were to return a record with a namespaced
element or attribute, it would have to decide what sort of record it
was: XML or SGML -- a decision which sort of has no impact with the
records we've been discussing up till now.  Awkward.  I'm half
inclined to say (and I know you won't like this, Rob), badgers to SGML
-- let's just jump right into XML camp and have done.

Anyone?

BTW., just had another name idea, which I thought I may as well chuck
into the mix.  "ExplainAgain".  I think it does OK on the Tin, Google
and Pronouncability tests.  My reservation would be that it's sort of
backward-looking.

 _/|_	 _______________________________________________________________
/o ) \/  Mike Taylor   <mike at miketaylor.org.uk>   www.miketaylor.org.uk
)_v__/\  "There's a goldmine of goals there that never quite arrives"
	 -- Gary Lineker, on Andy Cole.





More information about the Ex-plain mailing list