RE: Linked Data URIs in MARC Authorities

Karen,

At one time MARC had a $g defined for URNs, but it was rarely used. Since URNs and URLs are subclasses of URI, $g was eventually abandoned in favor of the renamed and generalized $u (Uniform Resource Identifier). See the "CONTENT DESIGNATOR HISTORY" at the bottom of this page:

http://www.loc.gov/marc/authority/ad856.html 

For people who are confused about the difference between URN, URI, URL, etc., check out the W3C/IETF report explaining the "Classical view" vs. "Contemporary view"

http://tools.ietf.org/html/rfc3305#section-2.1

I can imagine someone wanting to store legacy URNs or "info" URIs in a 024 $u, but IMO *HTTP* URIs SHOULD be constrained to be Linked Data (303s or hash). This is because the things MARC authority records describe are "non-information resources".

Jeff

> -----Original Message-----
> From: Karen Coyle [mailto:kcoyle@kcoyle.net]
> Sent: Friday, October 01, 2010 6:20 PM
> To: Young,Jeff (OR)
> Cc: public-lld
> Subject: RE: Linked Data URIs in MARC Authorities
> 
> Quoting "Young,Jeff (OR)" <jyoung@oclc.org>:
> 
> 
> >
> > 024 8# $u http://example.org/foo
> >
> > I would argue that the spec for this new $u should be explicitly
> > worded to mention "Linked Data". Sensible behavior would be for it
> > to lead to content-negotiatable representations in HTML, MARCXML,
> > MADS, RDF, etc.
> 
> But isn't the identifier *just* an identifier? It could be used for
> anything where an identifier is useful -- not just linked data. Or are
> you thinking of this subfield to be *only* for LD identifiers? In that
> case, it might be useful to use a subfield other than $u, which in
> MARC has usually been used for URLs, not URIs (the 856 is specifically
> a location area field). So 035 $l or 035 $i, or something like that.
> 
> kc
> 
> >
> > Jeff
> >
> >> -----Original Message-----
> >> From: rxs@talisplatform.com [mailto:rxs@talisplatform.com] On Behalf
> Of
> >> Ross Singer
> >> Sent: Friday, October 01, 2010 4:36 PM
> >> To: Martin Malmsten
> >> Cc: Young,Jeff (OR); public-lld
> >> Subject: Re: Linked Data URIs in MARC Authorities
> >>
> >> Martin, I think it's a fine proposal.
> >>
> >> The only possible downside I can see (as opposed to using, say, the
> >> 035, for example) is that it would be in a different location
> >> depending on the kind of authority record it is
> >> (personal/corporate/meeting name, uniform title, topical,
> >> geographical, etc.).
> >>
> >> That's not necessarily a killer, but it would mean you'd need to
> look
> >> for every field until you found the URI.  Using the 035 would
> >> centralize that a bit.
> >>
> >> Martin, since $0 isn't actually considered part of MARC authority,
> >> have you seen any systems reject this (or have you just used it
> >> locally)?
> >>
> >> My guess is that systems will ignore the subfields they don't
> >> understand rather than raise an error, but I guess it will take a
> real
> >> world trial to know for sure.
> >>
> >> -Ross.
> >>
> >> On Fri, Oct 1, 2010 at 4:29 PM, Martin Malmsten
> <Martin.Malmsten@kb.se>
> >> wrote:
> >> > Jeff, Ross,
> >> >
> >> > we use $0 when exporting our bibliographic[1] records which is why
> I
> >> chose it. Again this is just testing, but it seems a likely
> candidate.
> >> >
> >> >> It seems applicable, but the context it would be used in would
> sort
> >> of
> >> >> imply the opposite meaning than what it does in bibliographic
> >> records.
> >> > I see the link as going either "sideways" to another authority
> >> record/page/resource or "upwards", e.g from our 750 to a LCSH. In
> the
> >> latter case we would ultimately want to propagate changes made to
> the
> >> LCSH into our record, making the link behave like between a bib and
> an
> >> auth.
> >> >
> >> > /martin
> >> >
> >> > On Oct 1, 2010, at 9:53 PM, Ross Singer wrote:
> >> >
> >> >> Jeff,
> >> >>
> >> >> The 1xx$0 is actually used in bib records (not authority) and is
> >> defined as:
> >> >> $0 - Authority record control number (R)
> >> >>
> >> >> http://www.loc.gov/marc/bibliographic/bd100.html
> >> >>
> >> >> It seems applicable, but the context it would be used in would
> sort
> >> of
> >> >> imply the opposite meaning than what it does in bibliographic
> >> records.
> >> >>
> >> >> -Ross.
> >> >>
> >> >> On Fri, Oct 1, 2010 at 3:47 PM, Young,Jeff (OR) <jyoung@oclc.org>
> >> wrote:
> >> >>> Martin,
> >> >>>
> >> >>> I can believe that "the 1XX identifies what the record is
> *about*"
> >> and would challenge anyone to argue otherwise.
> >> >>>
> >> >>> What is your argument for choosing $0 rather than $u? Neither
> are
> >> currently specified and $u appears to be commonly used for URIs in
> >> other fields:
> >> >>>
> >> >>> http://www.loc.gov/marc/856guide.html#other_fields
> >> >>>
> >> >>> Jeff
> >> >>>
> >> >>>> -----Original Message-----
> >> >>>> From: Martin Malmsten [mailto:Martin.Malmsten@kb.se]
> >> >>>> Sent: Friday, October 01, 2010 3:32 PM
> >> >>>> To: Young,Jeff (OR)
> >> >>>> Cc: public-lld@w3.org
> >> >>>> Subject: Re: Linked Data URIs in MARC Authorities
> >> >>>>
> >> >>>> Jeff,
> >> >>>>
> >> >>>> I understand, but would not putting a $0 in the 1XX accomplish
> >> just
> >> >>>> that since the 1XX identifies what the record is "about"? I'm
> just
> >> >>>> saying that by using $0 you could link to other things (or
> Things)
> >> from
> >> >>>> other parts of the record as well.
> >> >>>>
> >> >>>> However, we do actually use 856 with a $z in our production
> >> environment
> >> >>>> today. It works, but I do not like the amount of implicit
> >> information
> >> >>>> with this (or rather our version of this) solution.
> >> >>>>
> >> >>>> Example:
> >> >>>> 100 '1' ' ' $aStrindberg, August, $d1849-1912
> >> >>>> 856 '4' '8' $uhttp://viaf.org/viaf/54154627 $zVIAF
> >> >>>>
> >> >>>> /martin
> >> >>>>
> >> >>>> On Oct 1, 2010, at 8:54 PM, Young,Jeff (OR) wrote:
> >> >>>>
> >> >>>>> Martin,
> >> >>>>>
> >> >>>>> I think our use cases are getting mixed up. I want a place to
> >> >>>> identify the thing the Authority record (as a whole)
> represents.
> >> >>>> Linking to *other* things inside a MARC record is a harder and
> >> more
> >> >>>> controversial problem as Michael's response indicates. I'm
> hoping
> >> this
> >> >>>> is low-hanging fruit, but I admit the difference is subtle.
> >> >>>>>
> >> >>>>> Jeff
> >> >>>>>
> >> >>>>>> -----Original Message-----
> >> >>>>>> From: Martin Malmsten [mailto:Martin.Malmsten@kb.se]
> >> >>>>>> Sent: Friday, October 01, 2010 2:36 PM
> >> >>>>>> To: Young,Jeff (OR)
> >> >>>>>> Cc: public-lld@w3.org
> >> >>>>>> Subject: Re: Linked Data URIs in MARC Authorities
> >> >>>>>>
> >> >>>>>> Jeff, Karen.
> >> >>>>>>
> >> >>>>>> I prefer a subfield over a field because may I want to link
> only
> >> >>>> parts
> >> >>>>>> of the record, and not necessarily the 1XX-field, to another
> >> >>>> resource
> >> >>>>>> without having to resort to a $8-link (*shudder*).
> >> >>>>>>
> >> >>>>>> Example:
> >> >>>>>> 150 ' ' ' ' $aMödrar
> >> >>>>>> 750 ' ' '0' $aMothers $0
> >> >>>>>> http://id.loc.gov/authorities/sh85087526#concept
> >> >>>>>>
> >> >>>>>> /martin
> >> >>>>>>
> >> >>>>>> On Oct 1, 2010, at 6:46 PM, Young,Jeff (OR) wrote:
> >> >>>>>>
> >> >>>>>>> How about this:
> >> >>>>>>>
> >> >>>>>>> 856 4# $u http://example.org/foo
> >> >>>>>>>
> >> >>>>>>> Here's the documentation for the field:
> >> >>>>>>>
> >> >>>>>>> http://www.loc.gov/marc/authority/ad856.html
> >> >>>>>>> http://www.loc.gov/marc/856guide.html
> >> >>>>>>>
> >> >>>>>>> Jeff
> >> >>>>>>>
> >> >>>>>>> From: Martin Malmsten [mailto:Martin.Malmsten@kb.se]
> >> >>>>>>> Sent: Friday, October 01, 2010 12:26 PM
> >> >>>>>>> To: Young,Jeff (OR)
> >> >>>>>>> Cc: public-lld@w3.org
> >> >>>>>>> Subject: Re: Linked Data URIs in MARC Authorities
> >> >>>>>>>
> >> >>>>>>> I'm considering/testing $0 in the 1XX fields, analogues to
> $0
> >> in
> >> >>>> the
> >> >>>>>> bib record. The idea is that a DbPedia/Freebase/VIAF URI
> could
> >> >>>>>> authorise an authority record. "Global headings change"
> becomes
> >> a
> >> >>>> fun
> >> >>>>>> challenge with LD URIs within the record :)
> >> >>>>>>>
> >> >>>>>>> Sent from my iPhone
> >> >>>>>>>
> >> >>>>>>> On 1 okt 2010, at 18:00, "Young,Jeff (OR)" <jyoung@oclc.org>
> >> wrote:
> >> >>>>>>>
> >> >>>>>>> If somebody wanted to put a Linked Data RWO URI in a MARC
> >> Authority
> >> >>>>>> record, where would it plausibly go?
> >> >>>>>>>
> >> >>>>>>> Jeff
> >> >>>>>>>
> >> >>>>>>> ---
> >> >>>>>>> Jeffrey A. Young
> >> >>>>>>> Software Architect
> >> >>>>>>> OCLC Research, Mail Code 410
> >> >>>>>>> OCLC Online Computer Library Center, Inc.
> >> >>>>>>> 6565 Kilgour Place
> >> >>>>>>> Dublin, OH 43017-3395
> >> >>>>>>> www.oclc.org
> >> >>>>>>>
> >> >>>>>>> Voice: 614-764-4342
> >> >>>>>>> Voice: 800-848-5878, ext. 4342
> >> >>>>>>> Fax: 614-718-7477
> >> >>>>>>> Email: jyoung@oclc.org
> >> >>>>>>>
> >> >>>>>>
> >> >>>>>> -------------------------------------------------------------
> ---
> >> -
> >> >>>>>> Martin Malmsten (martin.malmsten@kb.se) - Senior Developer
> >> >>>>>> National Library of Sweden / National cooperation dept. /
> LIBRIS
> >> >>>>>> http://libris.kb.se
> >> >>>>>>
> >> >>>>>>
> >> >>>>>
> >> >>>>>
> >> >>>>
> >> >>>> ---------------------------------------------------------------
> --
> >> >>>> Martin Malmsten (martin.malmsten@kb.se) - Senior Developer
> >> >>>> National Library of Sweden / National cooperation dept. /
> LIBRIS
> >> >>>> http://libris.kb.se
> >> >>>>
> >> >>>>
> >> >>>
> >> >>>
> >> >>>
> >> >>>
> >> >>> Please consider the environment before printing this email.
> >> >>>
> >> >>> Find out more about Talis at http://www.talis.com/
> >> >>> shared innovation(tm)
> >> >>>
> >> >>> Any views or personal opinions expressed within this email may
> not
> >> be those of Talis Information Ltd or its employees. The content of
> this
> >> email message and any files that may be attached are confidential,
> and
> >> for the usage of the intended recipient only. If you are not the
> >> intended recipient, then please return this message to the sender
> and
> >> delete it. Any use of this e-mail by an unauthorised recipient is
> >> prohibited.
> >> >>>
> >> >>> Talis Information Ltd is a member of the Talis Group of
> companies
> >> and is registered in England No 3638278 with its registered office
> at
> >> Knights Court, Solihull Parkway, Birmingham Business Park, B37 7YB.
> >> >>>
> >> >
> >> > -----------------------------------------------------------------
> >> > Martin Malmsten (martin.malmsten@kb.se) - Senior Developer
> >> > National Library of Sweden / National cooperation dept. / LIBRIS
> >> > http://libris.kb.se
> >> >
> >> >
> >> >
> >
> >
> >
> >
> 
> 
> 
> --
> Karen Coyle
> kcoyle@kcoyle.net http://kcoyle.net
> ph: 1-510-540-7596
> m: 1-510-435-8234
> skype: kcoylenet
> 

Received on Saturday, 2 October 2010 13:35:29 UTC