Subject: Re: need help -- ISSNs
Paskin, Norman (n.paskin@doi.org)
Date: Fri, 29 Oct 1999 15:39:37 +0100
Message-Id: <97A4BBFAC1B9D211B2620008C71EF881F7FDB5@ELSOXFS12305> From: "Paskin, Norman" <n.paskin@doi.org> To: "'arl-ejournal@arl.org'" <arl-ejournal@arl.org> Subject: Re: need help -- ISSNs Date: Fri, 29 Oct 1999 15:39:37 +0100
On 29 Oct 1999, Arthur Smith <apsmith@aps.org> wrote:
>
> On Fri, 29 Oct 1999, Joel Diedhiou <jdiedhiou@yahoo.com> wrote:
> >
> > Does somobody have any explanation what difference between
> > electronic ISSN and its print versioncounterpart is? How do
> > professionals manage to assign electronic journals their ISSN?
> > Why is there a difference between the ISSN assigned to an
> > electronic journal the one assigned to its print version
> > counterpart.
>
> Well, it comes somehow from the definition of the ISSN - suffice to say
> it is more tied to the physical medium than to the "content". Though
> note that the there are ways in which what might be called "content"
> differs from one medium to another (the arrangement of "bits" or "dots"
> certainly differs) and ancillary pieces of the representation, such as
> HTML pages, links, etc, will also differ. So in fact we are supposed to
> have a separate ISSN for print, online, and CD-ROM distribution. This
> sort of manifestation vs. work identification issue has also bogged
> down the DOI effort. In my opinion it should be up to the publisher to
> determine whether a new ISSN or identifier is needed, not the body that
> gives them out.
>
> In practice what we do is use the electronic and CD-ROM ISSN's for
> purchasing etc., but wherever an ISSN is used for content identification
> (PII, SICI, other metadata, etc.) we use the original print ISSN of the
> journal. In practice it's not really a problem except for the hassle of
> obtaining and tracking extra ISSN's.
I would like to comment on Arthur Smith's response to this ARL list
question. He mentions that "This sort of manifestation vs. work
identification issue has also bogged down the DOI effort". This is
misleading - I would say that the DOI effort encountered this problem,
sought and found a solution for it, and is immensely enriched for the
experience. Far from being bogged down it has helped us to reach a
solution which is capable of deailing with exactly the sort of issues
posed in the original ISSN question to the list.
In brief the solution adopted is a generic data model approach which is
capable of expressing entities of any form (including abstract entities
such as works, physical entities such as books, etc) and the
relationships between them (e.g. two journals or ISSNs with different
delivery media are nevertheless related by virtue of having the "same"
content). See for example http://www.dlib.org/dlib/may99/05paskin.html
Those interested in more on this should consult http://www.indecs.org/
and also see details of the forthcoming conference on "Names Numbers
and Networks" at the Library of Congress on Nov 15 (details at the same
site). The data modelling work has also benefitted from the IFLA study
on Functional Requirements of Bibliographic Records led by Tom Delsey
which dealt with some of these questions in a bibliographic/library
catalogue context.
It remains true that deciding what is the "same" and "different" is an
issue which has to be decided on a pragmatic basis, perhaps by the
publisher -- e.g. is changing one comma enough to make it a "different"
work (this is the issue of granularity - also dealt with in the indecs
study).
Norman Paskin
International DOI Foundation
<n.paskin@doi.org>
This archive was generated by hypermail 2a16 : Mon Dec 20 1999 - 18:02:17 EST