Yahoo, LII, etc.

Jerry Kuntz jkuntz at rcls.org
Wed Nov 19 14:08:10 EST 1997


Karen G. Schneider wrote:

[snip]
> More staff, better indexing, more money for the
> brainiacs to develop better tools for automating organization and
> retrieval.  Now let's merge the RCLS kid's database with LII for a
> children's room (similar to Yahooligans).  Another ggrant rolls in, and we
> buy up a few more small collections and give 'em the big standardized
> reorg.  The records also go into Intercat, of course.
> 
> As for the kids, Yahooligans was created to make money for Yahoo; Surfwatch
> features a setting which can restrict Surfwatch to the Yahooligans
> database, and lame it is, but the concept is interesting, if we are talking
> about customizing a database so different users have different resources
> presented differently.
> 
[snip]

What Karen is referring to in the above vis a vis RCLS is an LTSA grant
that we received to create a children's search engine. We'll be meeting
within our system tomorrow to determine the database criteria and layout
(we've already decided to go with SWISH-E as the engine). The target
size--for the purpose of this grant--is 5000 sites. Not a great amount,
but compared to other children's search engines, nothing to sniff at.
Sites will be added and indexed through some nice web-based maintenance
mechanisms that Roy and his colleagues have developed for SWISH-E. (Some
of the grant money is also going to professional graphic design, so we
won't be inflicting the web community with more of our chintzy homemade
graphics!) If you have a strong opinion as to what sort of structure we
should be using--Dublin Core metadata, LC Children's Subject Headings,
etc.--we'd love to hear. Immediately.
In the context of dicussions about Yahooligans and Yahoo (and other
selective guides), yes I think librarians can build a better mousetrap.
In the context of the massive robot-indexed search engines, I'd like to
hear how librarians think they could improve upon the commercial
services. One idea I had (but it's probably not original) is to create a
robot-indexed search engine built off of Alta Vista's or Infoseek's top
25,000 most frequently linked URLs; i.e. something like a citation
index.

			Jerry Kuntz
			Electronic Resources Consultant
			Ramapo Catskill Library System
			jkuntz at rcls.org

P.S. Beyond the extent of the grant period (1 yr.), I'd imagine we'd be
very receptive to merging, collaborating, etc. But for the short term,
we acting on our own and on our own initiative.


More information about the Web4lib mailing list