Date: Thu, 23 Sep 93 13:11:17 -0700
From: jak@violet.berkeley.edu (John A. Kunze)
Message-Id: <9309232011.AA28341@violet.berkeley.edu>
To: timbl@nxoc01.cern.ch, uri@bunyip.com
Subject: Re: URN single or multiple variants (was: four-part harmony?)
Tim, I think I finally figured out what you were saying. Let me test
my understanding by restating it.
I think a key to Tim's proposal is that a URN should be fundamentally
just a name, and not also a statement about intellectual uniqueness;
that if URN1 == URN2, then the objects they name are "the same" only
if they somehow have an additional optional "sameness" property or
flag attached to them to indicate that those semantics are in effect.
Interesting idea, and definitely less constraining than the assumptions
I was working under. I do think, however, that the "sameness" property
that you wish to make optional was a sine qua non of the URN concept.
> Date: Thu, 23 Sep 93 17:15:20 MET DST
> From: timbl@nxoc01.cern.ch (Tim Berners-Lee)
>
> *Disagree*. Whether two things arethe same should *NOT* be regarded
> as a function of who you ask. It is a function of what transformations
> you are allowing which preserve "sameness". As a person I reserve the
As a user, I'd pay somebody else to think about sameness for me. That's
because I don't want to arbitrate all the nasty ambiguities that arise
when variants are derived in non-linear ways, or in irreversible ways
(e.g., lossy ways) from the "generic" object. Anyway, it's often hard
to know what the "generic" object is (e.g., which of these dusty scrolls
is THE bible anyway?).
Reminds me of Mark Twain's, Jumping Frog of Calaveras County, a short
story he wrote in English and translated back into English from a French
translation. He republished the three objects in one small book to
show how different they were. Are the three objects the same?
Maybe the first two, but the re-translation was humorous in intent.
What is the relationship of the small book to the original story?
I don't know, but I'm sure glad there are people who perform the
service of sorting it out for me.
> right to use different criteria for different purposes. I want to
> be able to give names to generic works, AND to particular translations,
> AND to particular versions. I want a richer world than you propose.
> I don't want to be constrained by your two-level system of
> "documents" and "variants". There are objects, which may be more or
> less generic. I may have many or no levels.
>
> I was proposing that the properties of a name be given in terms
> of the attributes of the document which may vary for the same
> name. This provdies a framework for actually talking about
> what we *do* mean by "sameness". It allows us to tackle the
> problem without constraining ourselves. I feel your variant
> specifier constrains the model without acually helping us much.
Hmm. Here I still see us in agreement. You want to associate certain
attributes with a name, and attributes may vary for the same URN, which
was assigned at the whim (hopefully a methodical whim) of an IdAuthority.
That's what I was saying.
> OK, now I haven't said what a variant specifier looks like or where it
> goes (only where it doesn't go). I believe there was some particularly
> focussed agreement on the needs and issues up to this point: that
>
> "The properties of a given URN should be a valid question between
> client and server and beween the URN issuer and issuee." (Tim B-L)
>
> These properties I've collectively called the variant specifier. Now to
> continue with the list of cautious premises.
>
>
> Hang on hang on hang on. A property of a URN is like "any document
> with this URN is in the same language". Your variant specifier
> spacifies a particular variant, not such a property.
I think you mean "any doc with this URN is the same *except* possibly
for language". Otherwise it sounds like you're prepared to give all
documents in English the same URN?
> (13)To paraphrase Tim, given a URN you should be able to ask some server
> to return you one or more variant specifiers, one for each variant.
> You select the variant you want, and pass it off together with the
> URN when you need to lookup the corresponding URL.
>
> **NO**. This is more like contraphrasing me. I don't believe in
> variant specifiers, so I wouln't have said that. I can imagine
> saying "Give me a URN for a postscript version of document <urn>".
> I can imagine saying "Give my a URN for a 600x300 pixel 5-colour GIF
> of document <urn>". I can imagine there being an infinite
> number of possible variants on a document, so I wouldn't ask for a list.
> But what I would get would be another, more specific, URN.
>
> I guess this is abstract, but I think it is useful and simple.
I'd be interested in seeing more details.
-John