[OPEN-ILS-GENERAL] ***SPAM*** Re: generic training server with predictable data

[Dan Scott]

Hi Don:

On Tue, 2010-06-22 at 08:23 -0400, Don Butterworth wrote:
> Hello everyone,
> 
> My name is Don Butterworth. I am the Head of Technical Services at
> Asbury Theological Seminary and have been lurking on this list for the
> last couple of months. Late last year our library migrated from
> Horizon to Symphony. After having worked with the cataloging and
> acquisitions modules for over six months, it is my personal opinion
> that our department's efficiency has been reduced by at least 20%.
> Because of that, if I can convince the "powers that be," I hope to set
> up an Evergreen test database here on campus, so that we can get a
> feel for the software, workflow, and level of IT involvement. Then,
> when the Acquisitions module is released in 2.0, thoroughly test it to
> see if it is a viable alternative to Symphony. 
> 
> I have files of OCLC MARC records, for all of last year's
> acquisitions, which I plan to import as our test database; about
> 10,000 records. If it is legal, I believe I can convince our
> administration to contribute them to an RSCEL training server, if that
> would be helpful. 

I think that would be helpful, but as I'm not a lawyer, I'm not going to
venture an opinion as to whether it would be legal for you to share OCLC
MARC records; you might want to seek independent legal advice on that.

I will note that the Evergreen source code already includes sets of MARC
records of varying types in the Open-ILS/tests/datasets directory (see
http://svn.open-ils.org/trac/ILS/browser/trunk/Open-ILS/tests/datasets/README for a description of those sets). In some cases the leader doesn't correctly identify the encoding of the records, others have similar problems that reflect the problems that we'll run into with other real-life records. I would be happy to add more records to those datasets and use those as the basis of a canonical set of predictable data, as long as contributors sign off on their right to contribute records.

I think Simon Spero was collecting a set of torture-test MARC records;
those might be useful for testing purposes as well.