From: Robert Hilliker <[log in to unmask]>
To: [log in to unmask]
Date: Fri, 4 Jan 2013 11:32:15 -0500
Subject:

Stevan,

While I share your concern that the primary focus needs to be on
enhancing open access to scholarly research--and that it can be easy
to let bells-and-whistles distract us from that core mission--I
frankly don't trust Google Scholar to "solve" all our
access-and-discovery issues and believe that work on cross-repository
discovery tools is actually an important piece of ensuring that OA has
the largest possible impact on society.

Given my own experiences working with the folks at Google to ensure
our repository was "properly" indexed in Google Scholar, I can tell
you that, while it has had massive benefits for us in terms of
increased traffic, it has also exposed limitations and blind spots in
Google's policies.  If anything, Google Scholar's indexing is more
opaque than their WWW indexing, particular their inclusion/exclusion
requirements, but also their relative weighting of OA versions of
content as opposed to subscriber-only and/or PPV versions.

As just a small example, our repository, which hosts only research
outputs, has over 8,000 items in it, yet even after months of
back-and-forth with the team at Google they still only index some
4,700 of those in Google Scholar.  Why?  Because they purposefully
exclude datasets, video of conference presentations (though they're
happy to take the "proceedings" versions), and so on.

Moreover, why, once a user comes to our site from Google Scholar,
should they not be presented with the option of seeing OA content from
outside Columbia?  If your concern is "empty" repositories, then why
shouldn't we seek to leverage the work (and success) of others and, at
the same time, provide better access to OA scholarship to the broader
community?  Cross-discovery could enrich small collections and provide
additional, local incentives to OA for scholars at campuses where
there is not the same groundswell of support for OA.

Further, as initiatives like ORCID (Open Researcher and Contributor
ID) begin to get off the ground, there are opportunities for
repositories to play a key role in ensuring that these consortial
efforts help us to further the goals of the OA movement by enhancing
the accessibility of OA content and not just that of commercial
publishers and content providers.  As the transformation of the PIRUS
(Publisher and Institutional Repository Usage Statistics) Project into
IRUS demonstrates, publishers and other commercial content providers
continue to (by-and-large) be driven by bottom-line considerations
(the PIRUS2 report makes this very clear).  Therefore, it is in the
interests of the OA movement as a whole to ensure a robust--and
open--ecosystem of discovery paths exist for OA content.

Sincerely,

Rob

------------------------------
Robert Hilliker, PhD, MLIS
Digital Repository Manager
Center for Digital Research and Scholarship
Columbia University
International Affairs Building
New York, NY 10027
E-mail: [log in to unmask]
Web site: http://academiccommons.columbia.edu
Twitter: @ResearchatCU

On Jan 2, 2013, at 8:09 PM, Stevan Harnad wrote:

CHEER-LEADING, CHALLENGES AND REALITY

What is missing and needed is not "awesome repositories cross-search tools."

What is missing and needed is OA repository deposits, and OA deposit mandates.

The repositories are mostly empty.

And Google Scholar finds what OA content there is -- wherever it is on
the web -- incomparably better than "awesome repositories cross-search
tools."

Here is just a sample vanity search on a relatively uncommon name (try
your own):

Awesome repositories cross-search tool: Harnad 140 hits
Google Scholar: Harnad 15,900 hits (author:Harnad: 1,010 hits)