Re: [URN] Agenda for Washington Meeting -- Questions on URN

["Sam X. Sun" <[e-mail]>]

Hi, Ron,

Thanks for your response. 

----------
> From: Ron Daniel Jr. <[e-mail]>
> 
> At 03:23 AM 12/3/97 -0500, Sam X. Sun wrote:
> >
> >1.	What is the definition for URN addressed by the working group? Are we
> >defining it as a particular implementation limited to "urn:" URL scheme,
or
> >a specification that should embrace all persistent, location independent
> >global naming scheme?
> 
> IMHO, it is not possible to come up with a specification that embraces
> all persistent, location-independent, global naming schemes. The
> URN work covers a subset of them, and uses the "urn:" prefix to
> let people and software know if they are dealing with something
> in that subset.
> 

First of all, what I meant by implementation should really be "Syntax
definition". It's true that we may never be able to address all possible
syntax definitions that ever come up. But limiting URN to one particular
syntax definition has the effect of rejecting other URN syntax based on
other URL/URI schemes. While "urn:" scheme has its nice feature of easy 
input from standard English Keyboard, "hdl:" (handle system) scheme has its
advantage of requiring less reserved/excluded characters and allowing
native characters to be used without hex encoding, and the "pdi:"
(persistent document identifier) scheme has been used for document
identification for years.

I agree with you that there might be no one specification or syntax that
could embrace all persistent, location-independent, global naming schemes.
But this doesn't mean we have to pick one and reject all others. RFC1737
(URN requirement) does define URN to be limited to "urn:" scheme. But "URN
Syntax" draft (draft-ietf-urn-syntax-05.txt, section 2) starts to define
all URN have the syntax of:  URN ::= "urn:" NID ":" NSS. And this is where
get me confused.

> Second, this is not tied to one particular "implementation". The
> URN syntax document specifies a syntax, but there is no tie to a
> particular resolution mechanism. Instead, clients may use any mechanism
> for resolving the identifier that they want. NAPTR and THTTP are
> provided in case people wnat to use them, but the Handle system could
> just as easily be used.
> 

Again, I'm sorry for the confusion made by the word "implementation". It
should be "syntax definition".  On the other hand, the implementations
mentioned above are constraint to the syntax defined by the "urn:" scheme.
Thus if the syntax has any limitation, so would be the implementations that
implement it.

In summary, my point is that "urn:" syntax, as it defined now, serves well
as one URN "syntax definition", or one URN name space. But URN as a general
concept for persistent naming scheme, may not necessarily be restricted to
one URI/URL scheme. Besides "urn:", we now have seen "hdl:" and "pdi:"
claiming to be under the URN framework. And there could very well be other
URI/URL schemes to be proposed to serve the same purpose, but with
different syntaxes, and with their own set of implementations.


Thanks much for your response...
Sam
[e-mail]