[XML-DEV Mailing List Archive Home] [By Thread] [By Date] [Recent Entries] [Reply To This Message]

RE: URN vs URL

• To: 'Chiusano Joseph' <chiusano_joseph@b...>
• Subject: RE: URN vs URL
• From: "Bullard, Claude L (Len)" <clbullar@i...>
• Date: Thu, 31 Jul 2003 09:26:21 -0500
• Cc: Andy Greener <andy@g...>, Alaric B Snell <alaric@a...>, xml-dev@l...

Yep.  Yahoo.

Possibly because they click once, figure out it's not doing anything
useful, and don't do it again.  They self-educate.  
A 'bot might have a harder time with that.  So next we 
add some RDF at the end of it, and it has these abstract 
resource names in it, and on to the ultimate turtle.

Clarity remains an issue.  What happens if people using these 
do attempt to associate semantics with them, eg, code?  I've seen 
the MS means of doing that with stylesheets which work ok.  I've 
seen RDF that eventually resolves to a splinter of Java code.  It seems 
to me that at least in the wild, we are getting way beyond syntax 
disambiguation.

So clarity:  if they don't intend it to dereference, don't 
put a hypertext control value in there.  If they reserve the 
right to do that now or in the future, stub the nub.

len


From: Chiusano Joseph [mailto:chiusano_joseph@b...]

Ah, you're down in 'bama.

<Quote>
Did you explain the dereferencing problems of using http 
style URNs in namespace names?  
</Quote>

I actually am not working on this draft - my previous employer (LMI) is
(I would have worked on it had I still be there). But the authors are on
this list, and so you might consider your voice heard. But as a
double-ensurance, I will make sure that when I review the next version
that this type of verbiage is included - and if not, I will recommend
that it be included.

[not trying to resurrect a perma-thread here] 
As a quick point related to this: I have used URL-based namespace
identifiers in all of the E-Grants (grants.gov) schemas that I have
written. I am (pleasantly) surprised that at no point was there any
expression of confusion (from any agency involved) regarding the issue
at hand here - perhaps in some ways it's also a matter of educating
folks.

• Prev by Date: Re: Defining non-WXS datatypes
• Next by Date: Re: How did XML Blueberry get its name?
• Previous by thread: Re: URN vs URL
• Next by thread: DOM to SAX
• Index(es):
  • Date
  • Thread