I might be a bit biased towards using DOI or a similar Handle System
based service for maintaining persistent identifiers ;)
My main interest right now in this matter is serial identifiers and
metadata for serials. I work for a legal publisher that publishes
almost everything through at least three different media (paper, cd-rom,
the web, news services, etc.).
To the extent that public authorities are going digital, we'll also
need to be able to refer to and link to e.g. a court ruling or a clause
in a law, using public persistent identifiers.
Why DOI? DOI is based on a solid framework for describing those things
and events that we might want persistent identifiers for:
http://www.indecs.org
and allows for resolution to abstracts, library records, or
"appropriate copy" (assisted by e.g. OpenURL or SFX):
http://www.dlib.org/dlib/september01/caplan/09caplan.html
http://www.la-hq.org.uk/directory/record/r200112/article2.html
Try e.g.
http://www.crossref.org/mr/index.html
http://demo.exlibrisgroup.com:8888/demo?id=doi:10.1006/pmed.2001.0888
If you'd like a few more (a lot actually) perspectives, start here:
http://info.lib.uh.edu/sepb/lbcat.htm
http://info.lib.uh.edu/sepb/sepb.html
Kind regards
Peter Ring
Magnus Informatik A/S
A Wolters Kluwer Company
-----Original Message-----
From: Caroline Arms [mailto:[log in to unmask]]
Sent: 26. februar 2003 15:19
To: [log in to unmask]
Subject: Re: Persistent identifiers
Peter Ring's list of references covers many of those I would have
provided. None of these are necessarily objective and unbiased. You
have to look at the issue from many perspectives.
http://www.dlib.org/dlib/january03/paskin/01paskin.html
This highlights the issues relating to the issue of when things are the
same and should share an identifier.
http://www.arl.org/newsltr/194/identifier.html
This 1997 piece by Cliff Lynch ends with 5 questions to consider in
relation to identifier schemes for things you might have in libraries --
still worthwhile questions.
http://www.uddi.org/
Identifiers are a key part of the Universal Description, Discovery and
Integration (UDDI) protocol
FYI, the Library of Congress uses CNRI's handle server so that the catalog
records it distributes for items it holds and can make available to the
public have persistent links to online presentations of the items. Those
catalog records have their own identifiers (Library of Congress Control
Numbers or LCCNs). The distinction between identifiers for digital items
and identifiers for records that describe them can be very important,
since in many circumstances the relationship is not one-to-one.
Caroline Arms [log in to unmask]
Office of Strategic Initiatives
Library of Congress
On Wed, 26 Feb 2003, Peter Ring wrote:
> OK, so this is my hobby horse ;) A few more references.
>
>
> ISO/TC 46/SC9/WG3 International Standard Text Code (ISTC)
>
> http://www.nlc-bnc.ca/iso/tc46sc9/wg3.htm
> http://www.nlc-bnc.ca/iso/tc46sc9/docs/sc9n308.pdf
>
>
> National Library of Australia papers and schemes:
>
> http://www.nla.gov.au/initiatives/persistence.html
>
>
> Another survey of papers on identifiers:
>
> http://www.doi.org/publications.html#identifiers
>
>
> Some more papers on the relation between identification,
> metadata, and rights management:
>
> http://www.dlib.org/dlib/july98/rust/07rust.html
> http://www.doi.org/topics/drm_paskin_20030113_b1.pdf
>
>
> Kind regards
>
> Peter Ring
> Magnus Informatik A/S
> A Wolters Kluwer Company
>
>
> -----Original Message-----
> From: Maewyn Cumming [mailto:[log in to unmask]]
> Sent: 26. februar 2003 10:47
> To: [log in to unmask]
> Subject: Persistent identifiers
>
>
> Has anyone done any work on the different systems for persistent unique
> identifiers? I need to make recommendations regarding the best way forward
> for the UK government and would love to see some independent research, or
> hear about people's experience.
>
|