On Thu, Nov 12, 2009 at 5:27 PM, Jon Phipps <jonphipps_at_gmail.com> wrote:
> I think that lcsh has a particular problem in that the data was starting to
> become widely used, as linked data, when the original site was abruptly shut
> down. See the comments:
> http://lcsh.info/comments1.html
>
> The current id.loc.gov site took quite a while to bring up and in the
> interim a tremendous amount of credibility was lost, particularly
> internationally. People who had started using the data were left in the
> cold, and I for one would be very reluctant to trust any aspect of my system
> to that particular dependency again.
>
Well, certainly the handling of lcsh.info was clumsy and definitely
left libris.kb.se in the lurch (who were probably the biggest
stakeholders when lcsh.info went down), but I'm not sure the lack of
huge uptake of id.loc.gov is because of nervousness about the
credibility of the service.
In fact, right now the service is _far_ more dependable than it was
when it was running on a machine that I and 4 other people with root
access share for our various pet projects. That has been hacked in
the past.
That it is provided by the Library of Congress gives me considerably
more confidence that it's a service that won't disappear tomorrow
because a machine goes down or bill forgets to get paid or somebody
gets hit by a car while their riding their bike home from work or
something.
lcsh.info was, basically, a lesson about how this stuff is going to
work on the web.
We're in better shape now: LC is behind the support, it's mirrored at
lcsubjects.org, the data is available and it could easily be recreated
in a matter of days should the current stopgaps fail.
Rather than being an issue of credibility, I would say the biggest
reason that id.loc.gov is getting relatively little use is because the
communities that it's designed for aren't using it: libraries.
LCSH authorities aren't terribly interesting to non-library
communities by themselves. They have simpler or more appropriate
domain-specific thesauri to describe their data. What is interesting
to non-library consumers, however, are the resources we've described
with these subjects. Then when these subjects are related to their
subjects, we have a rosetta stone of sorts.
I realize all of this is preaching to choir by replying to Jon.
The problem is of the chicken or egg variety. There is no way that
libraries could effectively model their records as linked data without
big, centralized initiatives such as id.loc.gov or viaf.org to provide
common identifiers to common data points. However, until the
libraries actually start thinking seriously about linked data, these
are going to be sleepy, niche pockets of the linked data cloud with
very little traffic.
-Ross.
Received on Thu Nov 12 2009 - 22:03:34 EST