-----BEGIN PGP SIGNED MESSAGE-----
Am 05.08.2014 12:13, schrieb Juha Hakala:
> Sadly especially libraries took their part in blurring that semantics,
> but that is a different story.
>> Can you elaborate a bit what you mean by this?
[I noticed that you talk about "semantic identifiers" in the sense
that one can extract (meaningful) information by analyzing the
form of the identifier given. In contrast to that I was talking about
the semantics of the identifier (system), meaning what the "things"
are which are denoted by the identifiers and when they should be
considered equivalent (with respect to that identifier). In the ISBN
example the "semantics of the identifier" IMHO can be expressed in
short by the statement "ISBNs pertain FRBR manifestations"]
So if your question hasn't become mmot by that clarification, here is
<http://lccn.loc.gov/2012532574> is a two-volume set with three ISBN:
One for each volume and one additional for the set.
LC's record based on AACR tradition represents the set as a whole, thus their
recording of the two additional volume ISBNs must be questioned: If you
take one of these ISBNs and communicate it to your local bookseller the
outcome will be quite different depending on which of the three numbers
you happened to select. That is because ISBN semantics has been deliberately
broken by the library.
Or rather the libraries: All is working well again within an universe
where individual representations of volumes do not exist at all, like
in anglo cataloguing. But problems like above arise again when AACR2
records meet e.g. RAK records from Germany, where usually there exist
three individual bib records exactly corresponding to the set and the
individual volumes linked to it (thus each of the three records
corresponds 1:1 to one of the three ISBNs in our example).
However having aligned entities does not automatically prevent libraries
from wreaking havoc: Since cataloguing is based on the volumes (not
an hypothetical set container) the set ISBN might well enter any of
the individual volume records. And vice versa.
(I just checked the german and austrian union catalogues, indeed all
of them correctly associate the three ISBNs with their respective bib
records, thus my remark has to remain a bit speculative)
There are more well-known examples with respect to ISBNs representing
different print and eBook formats, these are parenthetically resolved
by RDA: Even communities not (or not yet) cataloguing with RDA seem
to become "FRBR aware" and start shaping their entities closer towards
"manifestations" (my subjective impression, I wouln't be able to
Just the other day in a personal communication Caren Koyle pointed
me to the fact that /public libraries/ may have a different agenda,
(based on the same MARC records) they seem to pursue an approach which
could be described as "expression level cataloguing", by "enriching"
their records with identifiers of resources which others would recognize
as only related ones. (I hope I got that right)
>> Other problem with ISBN semantics is that in some countries publishers would
>> want to assign the same ISBN to all versions of a book. Libraries are opposed to
>> such practice.
I could imagine that publishers in those countries will act that way:
When they have internal consensus they simply do it (assuming that
most sales are domestic and libraries won't boycott them anyway).
Curious question: How can the broader community prevent subcommunities
splitting off by introducing their own rules of the game?
>> International ISBN agency was closely involved with the registration of the URN
>> namespace. The registration request was written by myself and the (past)
>> chairperson of the International ISBN Agency. I work with the current
>> chairperson to revise the existing ISBN namespace registration (revision is
>> required due to the introduction of ISBN-13).
(So what happened to <
http://tools.ietf.org/html/draft-ietf-urnbis-rfc3187bis-isbn-urn-03 > after it
expired on April 22, 2013?)
>> In my opinion, International ISBN Agency has definitely endorsed URN:ISBN scheme.
All right. I derived my allegation from the fact that I was never able
to find any mention of URIs or schemes on http://isbn-international.org/
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1
Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/
-----END PGP SIGNATURE-----