W3C home > Mailing lists > Public > uri@w3.org > May 2000

Re: [Fwd: Re: Approval of initial Dublin Core InteroperabiityQualifiers]

From: Martin J. Duerst <duerst@w3.org>
Date: Tue, 09 May 2000 16:11:10 +0900
Message-Id: <>
To: Ray Denenberg <rden@loc.gov>, W3C URI List <uri@w3.org>
At 00/05/01 10:46 -0400, Ray Denenberg wrote:
>"Martin J. Duerst" wrote:
> > Could somebody from the DC community explain what an
> > 'encoding scheme' is in DC? Or give a pointer?
>I'll be happy to be enlightened by someone more DC-mainstream;  pending that,
>here's my explanation (in the context of the current discussion):  Say 
>that a DC
>metadata element needs to identify a resource, for example, say it's the
>'relation' element, where you indicate both a relationship (e.g. 
>'replaces') and
>what resource the resource-under-description bears this relationship to. 
>You have
>to identify the resource, so you need a resource identifier. Quote from DC:
>(begin quote)
>Qualifiers for 'Relation':
>    Element Refinements:
>         Is Version Of
>       Has Version

>    Encoding Scheme:
>       URI
>(end quote)
>So for example say you want to identify the resource by "isbn", this 
>presumes that
>there will be a URI scheme for 'isbn'.

No, it presumes that there will be an URI for each item that can
be identified by an isbn number.

>I'm not sure I should  respond to this assertion, on this (public) list,
>so I'll save it for the AC meeting.

Good point.

> > >What's
> > >needed is a coherent document (preferably, normative), that ties all 
> of the
> > >relevant information together, perhaps pointing to the appropriate RFCs.
> >
> > What should such a document say? Should it be updated every time
> > a new URI scheme is defined? Isn't it dangerous to try and say
> > everything in one single place?
>I would be happy (or at least happier) if there were a document that 
>listed the
>relevant RFCs (and permitted the reader to determine by inference that a 
>given RFC
>is  irrelevant, out-of-date, still experimental, etc.) and told you which 
>RFC to
>look at to solve a given problem.  But it should be an *authoritative* 
>(not just someone's private web page, or some *informal* W3C page).

As for status of RFCs, there is an RFC or some list that's
continually updated to document this. Of course, this is
the 'official' status, which is sometimes different from
the actual status.

> > > -- we'll never even begin to
> > >discover what URI schemes are going to be necessary, useful, and/or
> > >popular;
> >
> > Well, this is a feature, not a problem. If we knew exactly which
> > URI schemes would be needed/popular in a few years, we would
> > already have invented them.
>Our views differ here.  Your's seems to be: determine which shemes are 
>going to be
>the winners, then register them;

No, not exactly. Before you start using a scheme, PLEASE register it!
Don't use unregistered schemes.

>the reason for the intertia is that we haven't
>quite figured out which. Our  view is: there are many potential schemes; 
>we won't
>know which are going to be winners without some shakedown process, including
>registering some that ultimately won't survive, so let's get started 
>schemes, see which are deployed, which become popular, and which ones vendors
>support -- the reason for the intertia is the confusion over how to 
>register the

No, the main reason is that there is no police that hands
out fines if you use an unregistered scheme, and that many
people think that somebody else could do the work.

> > .>....registration of "isbn:"  hasn't
> > >even been explored, apparently because it is assumed that it simply 
> cannot be
> > >registered, for legal reasons.
> >
> > What would these reasons be? I would be interested to know.
>I'm sure you know as much as I do about this. The IETF can't simply write 
>an RFC
>registering ISBN as a URI scheme because of the proprietary aspects of 
>ISBN, even
>though there is an RFC that describes this hypothetical scheme. It is up 
>to the
>ISBN Agency to decide that they want ISBN registered, and (according to Leslie
>dagle, in response to my original posting to the DC list) they're 
>exploring it.
>Problem is, they seem to have been "exploring" it for a very long time.  I 
>know what the legal/business consideration are.

Would be interesting to know more. I'm not a lawyer, but below are a
few guesses.

Is the problem that the string 'ISBN' is protected e.g. as a trade mark?
Trademark law doesn't forbid you to speak about things that have trade

I'm also sure that many library systems use ISBNs internally in some form,
and pass them over the network. Defining an URI for them is just an extension
of this practice, to a 'somewhat larger' network. It would be strange
if the ISBN agency could forbid (or would want to forbid) the use of
ISBN numbers on computers or networks.

The IETF has similar examples of third-party registrations, for MIME types.
If a third party needs an identifier for a format produced by a product,
they can have it registered. There is no assumption of ownership by
this registration. If the product vendor later gets aware of the registration,
they can take it over. Something similar could be done for ISBN, the
registration will only say that it identifies whatever the ISBN agency
decides it does.

It may get more complicated when discussing what isbn: or urn:isbn is
supposed to return, but as far as I understand, at least for the
registrations and for DC, it would be enough to say that it identifies
the relevant book as the resource, but doesn't (at least for the moment)
lead to an entity.

Regards,   Martin.
Received on Tuesday, 9 May 2000 07:42:57 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:25:01 UTC