Hi Osma,
Thanks very much for watching the presentation, and for sharing the link to your own.
I haven’t gotten all the way through it yet, but my early impression is that it, like many (all?) of the “linked data for libraries” presentations I have seen, it is based on an unstated assumption, which I will phrase as a question:
Does library bibliographic metadata have value outside the silos in which it currently resides?
If yes, why did dozens (hundreds?) of US (and other?) libraries give millions (if not hundreds of millions) of dollars’ worth of bibliographic metadata to Google as part of the Google Books project?
If no, then the assertion that pushing library bibliographic metadata out onto the open Web as linked data is worthwhile, is false.
Put differently, this can be stated as something like the ought-is fallacy. (I thank Adam Chandler for pointing me here: http://www.txstate.edu/philosophy/resources/fallacy-definitions/Ought-Is.html):
Library bibliographic metadata has been created over decades by thousands of highly-trained professionals, so inevitably it must have value in the larger ecosystem of the open Web.
But does it?
I will watch your presentation more carefully and respond off-list, because I think my comments will fall outside the scope of strictly BIBFRAME-related discussion.
Thanks again,
Jeff
P.S. You characterize MARC as “an awkward and archaic foreign language.” I feel the same way about Finnish. ;)
________________________________________________________________
Hi Jeff,
Thank you for sharing this entertaining and thought-provoking presentation!
You make valid points about the relative importance of library metadata
and cataloging vs. the scale of the web. However, even knowing the
facts, I have a somewhat different view on this universe. I come from a
Semantic Web / Linked Data background and to me, MARC is an awkward and
archaic foreign language. If library data is to have a future, it has to
be opened up in a form that non-librarians can understand.
Our main fear, as the national library of a small country, is that
libraries like ours are losing the battle for relevance on the web
against multinational data giants such as Google and Facebook. In the
old world, people came to libraries for their books. Now they go to web
search engines for their data, and libraries are being sidelined. If we
don't improve our data offerings soon (and to be honest, we're pretty
late to the game), then nobody is going to need us in the future,
resources will be cut, not just cataloging but all library departments
will shrink.
Investing in open data and Linked Data offers a possible way out, a way
of regaining some relevance in the world of web data. To me, BIBFRAME is
a step away from MARC and roughly in the right direction, though still
way too restricted to the library world view. Sure it has its problems,
both technical (too complex, not reusing other web vocabularies, still
too much oriented around strings and not things) and social (developed
mostly behind closed doors by a small group of people and
organizations). However, trying to improve it is more helpful than
rallying people against it.
I gave a DCMI webinar recently about data models for library data that
people on this list might be interested in. It's available on YouTube as
well on the DCMI channel:
https://youtu.be/Ua9Wd90h97Q
-Osma
27.03.2017, 19:33, Jeff Edmunds kirjoitti:
> The presentation to which I referred several weeks ago ("Life after MARC: The Future of Discovery") is now available on YouTube:
>
> https://www.youtube.com/watch?v=gg-I0Eu5zAQ
>
> It touches both on the relationships between MARC and BIBFRAME (in a very basic way--the audience was not all catalogers and technologists) and on how linked data may or may not affect the library metadata ecosystem from the perspective of a large academic research library.
>
> Thanks,
>
> Jeff
>
|