>The difficulty that I see with adding the contents of the library
catalogs is the page rank.
This is a valid point. I see a number of ways this can be overcome.
First, people DO want to link to library catalogs. Indeed, do a site:
or link: check on any of the major libraries, and you'll see it
happening every day. From blogs, class bibliographies and every other
type of page out there, people link to OPACs. People find it natural.
Best of all, every time they do, it works!
Unfortunately, twenty minutes later the session dies and it
fails—"psyche!" But people learn, and they stop trying. They learn
that libraries aren't like everything else on the web. They learn, as
we all agree, that "OPACs suck." Except we see the "OPAC." They see
the library itself.
Second, there *are* patterns within the links. It would be interesting
to do a straightforward PR run through a standard OPAC, but I suspect
that PR would be distributed in decent, non-random ways, particularly
if the "in" is also non-random. Someone links to the LC's copy of
Nabokov's Pnin. PR flows to other Nabokov books. Even subject headings
are not random—some are way more prevalent, and these tend to be of
greater general interest. That is not to say that, as a former SEO
expert, I wouldn't think long and hard about record-to-record link
structure. There's a lot of upward potential. LibraryThing isn't very
well-optimized, and it's about the 15,000th most popular site on the
web. Still, the English site alone has 4.5 million pages indexed.
My pizza point appears to be misunderstood. The fact is, I can search
for ANYTHING that isn't a library book. I can search for Indian food,
or dog aromatherapists, or ... well you get the idea. "Anything" is a
superset of "any book." OCLC has just over a billion holdings. Google
stopped counting at 8 billion pages in 2005. It later claimed three
times Yahoo's 20 billion. It's not getting smaller.
Tim
On 4/23/07, K.G. Schneider <kgs_at_bluehighways.com> wrote:
> > I suppose "IEEE Conference 1998 Georgia Tech" might get one closer,
> > but I honestly don't think exposing our vanilla OPACs are going to be
> > the path to much joy. There needs to be a different approach to this
> > problem because merely opening our catalogs up to Google reeks of
> > polishing the turd.
> >
> > -Ross.
>
> I was just about to make similar points (though perhaps without that
> interesting vanilla/turd oxymoron), specifically that, first, there aren't
> 117,467 copies of In Cold Blood in Google (to use the number of libraries in
> the United States, let alone the world), and if there were, that wouldn't
> necessarily be a good thing; but beyond that, there are additional concerns
> with all these mini-me catalogs, primarily the high cost of installing,
> maintaining, upgrading local software, and the high cost of creating and
> maintaining local metadata... "cost" in both cases meaning both what is
> literally expended in terms of library resources (money, and people, though
> of course people are money) and in terms of a loss of balance in where we
> focus our efforts.
>
> In other words, if we all shared one OPAC for book-like things (the way we
> do for Amazon, or for that matter, LibraryThing), maybe we'd spend less time
> twiddling with every live-long field in a record or upgrading our local
> installation of Innosirsi Voylibris, and focus more on pushing print into
> digital, making discovery easier, building nicer facilities, and procuring
> more content for people to use.
>
> Karen G. Schneider
> kgs_at_bluehighways.com
>
Received on Mon Apr 23 2007 - 18:54:35 EDT