[OAI-general] Re: proposed collaboration: google + open citation linking

Stevan Harnad harnad@coglit.ecs.soton.ac.uk
Wed, 6 Jun 2001 20:22:36 +0100 (BST)


On Wed, 6 Jun 2001, Terry Winograd wrote:

> sh> But to date, it is totally unnecessary. Those preprints and
> sh> postprints are all SELF-CERTIFIED as such by their authors, and
> sh> with virtually no exceptions what they say is true. Why? Because
> sh> this is an esoteric literature, written BY researchers, FOR
> sh> researchers -- for their peers (the same peers who do the peer
> sh> review). And there is simply no motivation to cheat. (What would be
> sh> the motivation?)
>
>tw> Yes, I agree, and this is why the physics archives have worked. I
>tw> don't think the approach will work as well in areas where there are
>tw> commercial interests (e.g., computing technologies and medical
>tw> cures) or ideological differences (e.g., the archaeological history
>tw> of Palestine) and when a mass audience is present (people who
>tw> aren't researchers but want to find a medical cure or buy a piece
>tw> of technology). The motivation to cheat will rear its ugly head..

To cheat by pretending one's paper has appeared in a refereed journal
(which?) when it hasn't? How often do you think that might happen?
And what prevents the same person (with the same commercial
interests) from direct emailing of the bogus journal article, with a
bogus journal front-piece to the target audience. Where is the
edvantage of dropping it in an eprint archive?

Terry, this is all very new stuff. All kinds of possibilities enter
the imagination. It's important to sort the plausible ones from the
implausible ones before letting such thought-experiments govern our
behavioral strategies.

>tw> Getting a mass search engine like Google involved ups the ante. I
>tw> think that having a separate portal for "scientific-scholarly"
>tw> resources that don't have mass visibility may be better. This might
>tw> piggyback off of the Google database if that can be done
>tw> effectively.

Maybe that's a way to do it, though I'm not sure what it entails. The
possibility I had (in my imagination) was simply a google button that
then restricts the google search only to the registered OAI-compliant
eprint archives, with "refereed" flag on...

>tw> How about the simplest possible implementation - a search page that
>tw> calls google and automatically adds the equivalent of "and ''oams'
>tw> in meta tags" to the query. Anyone can get a page included by
>tw> putting this in the meta information, but as you say, we need to
>tw> count on people not to cheat. It doesn't enforce compliance with a
>tw> standard, but does let them declare that something is
>tw> scholarly/scientific. It makes it a lot easier to get people to buy
>tw> in, rather than persuading them to implement complex meta-data.  -t

An unscreened "oams" tag that anyone could use would really invite
abuse and would be self-defeating. It takes much more to set up an
OAI-compliant eprint archive, screen it, and register it with OAI than
simply to attach oams to all files. Besides, all the OAI-compliant
archives are interoperable, and google could trigger on their other
important tags, like "refereed" and "journal name".

It seems to me the registered set of OAI metadata tags is the critical
element here, not an arbitrary oams tag attachable to any file
(except if there were a way to get the "oams" tag authenticated -- but
that would be just re-inventing the OAI without the
interoperability!).

Cheers, Stevan
--------------------------------------------------------------------
Stevan Harnad                     harnad@cogsci.soton.ac.uk
Professor of Cognitive Science    harnad@princeton.edu
Department of Electronics and     phone: +44 23-80 592-582
             Computer Science     fax:   +44 23-80 592-865
University of Southampton         http://www.cogsci.soton.ac.uk/~harnad/
Highfield, Southampton            http://www.princeton.edu/~harnad/
SO17 1BJ UNITED KINGDOM