[OAI-implementers] RE: [oai-alpha] Re: Fwd: MARC to DC Translator (ALPHA)
Young,Jeff
jyoung@oclc.org
Thu, 25 Jan 2001 16:44:10 -0500
It's difficult for me to judge how much of this is of interest to the
OAI-implementers group. At the very least, perhaps Diane and I have hinted
that the issues are more subtle and difficult than one would hope.
I'd like to reiterate that the client/server model isn't the most efficient
way to do conversions. If people understand this and still desire a choice
of crosswalks, though, I'd be willing to consider it.
Finally, I hope that the concern about "round trips" between MARC and DC is
more of an issue for CORC than it is for my server. CORC uses "qualified"
Dublin Core whereas OAI specifies (and my server produces) "unqualified" DC.
Basically, I've taken CORC's qualified DC and "dumbed it down" to
unqualified. From there, there's no going back to MARC. One could argue the
result is still tainted by the issues Diane raises, but, OTOH, perhaps the
problem is mitigated by the limitations inherent in unqualified Dublin Core.
Jeff
> -----Original Message-----
> From: Diane I. Hillmann [mailto:dih1@cornell.edu]
> Sent: Thursday, January 25, 2001 3:59 PM
> To: Young,Jeff
> Cc: childree@oclc.org; oai-implementers@oaisrv.nsdl.cornell.edu
> Subject: RE: [oai-alpha] Re: Fwd: MARC to DC Translator (ALPHA)
>
>
> Jeff:
>
> Multiple crosswalks might be useful, particularly in that it
> might give you
> the option of qualified or unqualified DC (not to mention all
> the other
> crosswalks out there).
>
> Eric C. will tell you that I've been arguing with him and
> everybody at OCLC
> about this issue for years. My argument is less with CORC's specific
> compromises (though I've argued with those, too) than with
> the ostensible
> reasons for them. Some of them were early attempts to "anticipate"
> qualifiers that *might* have been blessed (but weren't), but
> many more were
> made to facilitate round trips between MARC and DC. It seems
> to me if you
> want to go back and forth between a rich and a simple record
> for the same
> resource, linking (or some other strategy) makes far better
> sense than
> crosswalking again, and gives far more reliable results. This
> is one of
> those problems that is endemic with early implementations, but is not
> necessarily more valid just because it's being done in a
> large database.
>
> Eric will also tell you that I have also been vocal on MARBI against
> changes in MARC to facilitate round-tripping. So I am, at
> the very least,
> consistent.
>
> Diane
>
> At 01:46 PM 1/25/2001 -0500, you wrote:
> >(I'm moving this from oai-alpha to oai-implementers where it
> should get
> >broader attention.)
> >
> >Thanks for your feedback, Diane. One appealing option for me
> would be to
> >support multiple crosswalks with my server. I would note,
> however, that a
> >client/server model isn't the most efficient way of doing
> these things. The
> >main reason I used this approach is because the MARC to DC
> conversion that
> >CORC uses isn't easily de-coupled from the overall system
> (regrettably).
> >Client/servers are also nice because of their flexibility.
> Given a choice,
> >though, I would have preferred to write an XSLT stylesheet,
> which could be
> >easily distributed. Maybe someday.
> >
> >Having worked on CORC during it's development phase, however, I feel
> >compelled to make a defense of its crosswalk. CORC has a
> large and diverse
> >user base in Dublin Core terms, and thus compromises were
> undoubtedly made.
> >Being able to convert MARC to DC and then back to MARC is indeed a
> >challenge. I would note, however, that our staff worked
> closely with the
> >Library of Congress MARC office on virtually the entire
> crosswalk. The
> >result should be a fuller, richer record than most
> crosswalks provide. The
> >CORC staff is aware that some tweaks are necessary, and are
> working on them.
> >As Dublin Core become more prevalent, CORC will be obliged
> to accommodate
> >changes if crosswalk standards arise. Another benefit of the
> client/server
> >model is that such changes and fixes to CORC will be
> immediately reflected
> >by this server.
> >
> >I wouldn't claim that CORC's crosswalk is definitive, but I
> do think it
> >deserves consideration. Discussing the options available, as
> Diane as done,
> >should be very helpful for people faced with the task of
> converting their
> >MARC collections to DC.
> >
> >Jeff
> >
> > > -----Original Message-----
> > > From: Diane I. Hillmann [mailto:dih1@cornell.edu]
> > > Sent: Thursday, January 25, 2001 10:16 AM
> > > To: OAI alpha group
> > > Cc: oai-alpha@nsdl.CORNELL.EDU
> > > Subject: [oai-alpha] Re: Fwd: MARC to DC Translator (ALPHA)
> > >
> > >
> > > Jeff:
> > >
> > > There is, in fact, a mapping from MARC to DC at the address
> > > below, though I
> > > can well imagine that you could not find it in a web search.
> > >
> > > http://128.253.121.110/NSDLmetaWG/MARCtoDC.htm
> > >
> > > Note that this mapping is to qualified DC, though that
> shouldn't be a
> > > terrible problem.
> > >
> > > I'd be a bit leery of using a mapping based on CORC, as
> it makes some
> > > unwise compromises (IMHO) chasing the unholy grail of
> > > round-trip mapping.
> > >
> > > Diane
> > >
> > >
> > > At 08:09 PM 1/24/2001 -0500, you wrote:
> > >
> > > >--- begin forwarded text
> > > >
> > > >
> > > >Status: U
> > > >From: "Young,Jeff" <jyoung@oclc.org>
> > > >To: "OAI alpha group" <oai-alpha@nsdl.CORNELL.EDU>
> > > >Subject: [oai-alpha] MARC to DC Translator (ALPHA)
> > > >Date: Wed, 24 Jan 2001 17:49:02 -0500
> > > >List-Unsubscribe:
> <mailto:leave-oai-alpha-156R@www.nsdl.cornell.edu>
> > > >Reply-To: "OAI alpha group" <oai-alpha@nsdl.CORNELL.EDU>
> > > >
> > > >While working on an OAI repository, I had the need to
> convert MARC
> > > >communication format records to unqualified Dublin Core. I
> > > searched the web
> > > >for such a tool, without success, and decided to write one
> > > based on OCLC's
> > > >CORC system. I suspect such a service would be useful for
> > > others as well, so
> > > >I'd like to announce the availability of the server for
> > > evaluation purposes.
> > > >Its status should be considered alpha, and I can't make
> any long-term
> > > >promises about its continued availability yet.
> > > >
> > > >I've set up a form from which records can be submitted to
> > > the server at
> > > >http://purl.org/alcme/marc2dc/index.html. If you look
> > > closely, you'll see
> > > >some caveats listed on the form, as well as a reference to a
> > > file containing
> > > >a few test records I've put out on the web for you to play with.
> > > >
> > > >Until it's been scrutinized, I wouldn't recommend that you
> > > trust the results
> > > >as definitive. I would also request that you don't
> > > overburden the server by
> > > >converting your entire OPAC just yet. Just give it a poke
> > > and let me know
> > > >what you think. I'd be particularly interested in
> > > information and perhaps
> > > >even the comparison of results with any other MARC to DC
> > > converters that
> > > >people may know about.
> > > >
> > > >Thanks,
> > > >Jeff
> > > >
> > > >---
> > > >Jeffrey A. Young
> > > >Senior Consulting Systems Analyst
> > > >Office of Research, Mail Code 710
> > > >OCLC Online Computer Library Center, Inc.
> > > >6565 Frantz Road
> > > >Dublin, OH 43017-3395
> > > >www.oclc.org
> > > >
> > > >Voice: 614-764-4342
> > > >Fax: 614-764-2344
> > > >Email: jyoung@oclc.org
> > > >
> > > >
> > > >
> > > >
> > > >---
> > > >You are currently subscribed to oai-alpha as: dih1@cornell.edu
> > > >To unsubscribe send a blank email to
> > > leave-oai-alpha-156R@www.nsdl.cornell.edu
> > > >
> > > >--- end forwarded text
> > > >
> > >
> > >
> *~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*
> > > Diane I. Hillmann
> > > Metadata Specialist
> > > National Science Digital Library Project at Cornell
> > > Department of Computer Science Voice: 607/255-5691
> > > 419 Rhodes Hall Fax: 607/255-4428
> > > Ithaca, NY 14853 Email:
> > > dih1@cornell.edu
> > >
> *~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*
> > >
> > >
> > >
> > > ---
> > > You are currently subscribed to oai-alpha as: jyoung@oclc.org
> > > To unsubscribe send a blank email to
> > > leave-oai-alpha-156R@www.nsdl.cornell.edu
> > >
>
> *~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*
> Diane I. Hillmann
> Metadata Specialist
> National Science Digital Library Project at Cornell
> Department of Computer Science Voice: 607/255-5691
> 419 Rhodes Hall Fax: 607/255-4428
> Ithaca, NY 14853 Email:
> dih1@cornell.edu
> *~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*
>
>