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

[Leslie Daigle <[e-mail]>]

Sam,

I'm afraid I don't have the resources right now to address your reply
in the detail it deserves -- I will definitely track down the references
you've supplied.

I think others have already covered much ground on this topic, and
I'd like to add only a couple of points (in part to allay other concerns
that have been aired re. this thread):

On Thu, 4 Dec 1997, Sam Sun wrote:
> It is my believe that a Handle is a URN, and can be presented in syntax of
> either "hdl:" or "urn:hdl:". When used under "urn:hdl:" syntax, it is
> subject to the syntax rules defined by the URN Syntax draft
> (ftp://ds.internic.net/internet-drafts/draft-ietf-urn-syntax-05.txt). And
> when presented using "hdl:" syntax, it follows the syntax defined by the
> Handle Syntax draft
> (ftp://ds.internic.net/internet-drafts/draft-sun-handle-system-00.txt).

As I have said all along, this seems reasonable.

Now, have a look at draft-ietf-urn-nid-req-02.txt, and come to the
meeting with your comments on where you think Handles fit in to this
structure/process for registering URN namespaces!  That document is
_not_ finished, and definitely requires input from people who have
namespaces that they think belong in URN-land!


Now, for some personal opinion matter, which _might_ help clarify
where some of this confusion has erupted from:

It's worth noting that Once Upon A Time Long-Since Obscured, there
were URLs and URNs, and URLs were all prefixed by the four characters
"URL:" and URNs were to be prefixed by "URN:".  

	URL:<url-scheme><url-syntax>
	URN:<nooneknewwhat>

However, URLs were implemented long before URNs, and many schemes 
appeared, and gosh, why bother with the URL: as it was redundant?  So, 
things got abbreviated to <url-scheme><url-syntax>.  

When it came time to discuss URNs seriously, there was further discussion
of this, but "URN:" was felt to be an important part of the URN syntax
in order to distinguish these URIs from all the URL schemes that exist
today (and will exist tomorrow).  So, in it went.  

Note that the URN: was meant to distinguish the type of URI, and the
subsequent chunk of the identifier is to designate the namespace (or
scheme, if you use the URL analogy).

	URN:<namespaceid>:<namespacespecificstring>

More time has gone by, and in the name of convenience,  we talk now
of "URI schemes", and that "URN:" is just one of those schemes.  It's
been pushed down a level.

At this point, that's probably the easiest route forward, so that's fine.
But, it makes it look like there is One True Way to do URNs, whereas
there are A Thousand Blooming Flowers of URL schemes, and I'd argue
that's because they aren't actually being compared at the same level.

As I said, much of the above is personal opinion and recollection of
history, others are free to differ, but it provides a perspective
that might help clarify how limiting we feel this work is _not_.

Leslie.

----------------------------------------------------------------------------

  "_Be_                                           Leslie Daigle
             where  you                           
                          _are_."                 Bunyip Information Systems
                                                  [phone]
                      -- ThinkingCat              [e-mail]
----------------------------------------------------------------------------