Bernhard Eversberg <ev_at_biblio.tu-bs.de> wrote:
> Alexander's advice: "Oh, and build a goddamn search engine."
> is a bit unspecific.
In that posting it was, but linked all the other stuff preceding it,
including our talks about how Google does it. So let me be specific
before I take off ;
Build a search engine / spider infra-structure that harvest for
knowledge, for *librarians*, not for the general public. Build one
that finds stuff that might have librarian value over content value.
Think of how Google Scholar tracks citations, well, track subjects,
rate articles and books by author cred, harvest from more specific
sources (but not limited to academia), find correlatives to WikiPedia
articles, assign pools of subject experts to select resources (to
various lists), share these lists easily, build a Wiki / CMS over it,
embed smart cataloging directly in it (or use Coins and harvest them),
track patron usage to subject matter, and on and on. You don't need me
to come up with suggestion, the library value sky is the limit.
So *don't* create yet another MARC db search engine, create a
spidering library-values one. You need a tool that embrace both
library values (identity management, epistemology, subject-centric
browse and search, selection, subject experts, restoration and
cleanup, well-linked, ontologically sane and structured, faceted, and
so on) and the future of digital and physical content. You need a new
infra-structure. Build a new friggin' infra-structure around a search
engine / spider cluster that's not based on MARC, but on something
funky, even FRBR at the core of a distributed system. Collaborate on
it; every major library around the world only needs to assign a
pittance each, or dedicate a little or more (depending on how serious
you are) developer / manager / designer time. You've done it before,
and you definitely should do it again. Make an engine (and
infra-structure) the world would love to use.
The problem is that of all the search engines of the world there is no
alternative that embraces library values (although OpenLibrary and
LibraryThing both are getting closer all the time, and OCLC is, well,
another story). You need to embed your values into the fabric of the
future. It's both needed and missed, and if not done, probably render
libraries another few step less relevant to that very future.
Ok, holidays. Toot!
Alex
--
Project Wrangler, SOA, Information Alchemist, UX, RESTafarian, Topic Maps
--- http://shelter.nu/blog/ ----------------------------------------------
------------------ http://www.google.com/profiles/alexander.johannesen ---
Received on Wed Jul 07 2010 - 08:26:16 EDT