Stephen,
This is very interesting. Thanks for sharing. As you rightly point out,
“embracing Linked Data does not need to be a stark all-or-nothing
proposition.”
The wisc.edu prototype looks fine, basically an extremely sized-down
version of http://data.bnf.fr/ (which includes over 2 million authors as
of January).
My question then becomes: is this what (library) users want?
The 2 basic use cases for libraries (that involve collections) are:
1) User wants to find a specific item (howsoever item is defined).
2) User wants to find a manageable amount of information about a
specific topic.
Linked Data more obviously supports the second use case (as well as a
host of others, about which more below), but is LD the optimal (and most
economical) means for presenting library users with “all” they want to
know, or have access to, about a specific topic? I don’t know, but I
doubt it.
Many of the LD4L Use Cases
(https://wiki.duraspace.org/display/ld4l/LD4L+Use+Cases) seem to me to
overlook the obvious primary use cases for (library) users listed above
and strike me as either:
1) contrived: driven by what the data makes possible rather than what
people actually do
2) not unique to an LD ecosystem: many of these can be addressed to a
greater or lesser degree without Linked Data (with capital letters).
Finally, I want to re-iterate that Linked Data is great and Linked Open
Data is even better (at least from the user’s standpoint), but the main
creators of metadata (no longer libraries, but vendors, publishers,
etc.) whose livelihoods depend (in part) on placing a price on metadata,
are very loath to give it away for free. Again, my main beef with RDA
and BIBFRAME (and this is what I meant by the ambiguous “top-heavy” in
an earlier post) is that too much effort has gone, is going, and will
continue to go into articulating, refining, disseminating, etc. the
standards, models, etc. and too little has gone, or will ever be
available to go, into creation of the (meta)data that necessarily will
be the underpinning of the entire apparatus.
For example, in the course of typing this, I didn’t create any
bibliographic metadata at all. Now back to work.
Thanks,
Jeff (for the sake of disambiguation: https://www.youtube.com/watch?
v=YKIqmTeFlzI)
|