[Metadatalibrarians] Metadatalibrarians Digest, Vol 50, Issue 9

[Joe Altimus]

I'd like to work with others to develop some training tools for using XSLT
to do data processing for crosswalks, analysis, etc. I can contribute
several years of experience with XSLT, as well as a background as a
cataloger and data analyst.

One thing I don't know is what specific skills are commonly needed. There is
such a mix of systems,  services, and metadata formats around now. It would
help if others could say what specific things they want to do with XSLT.

XSL is a powerful programming language. Just like Java or other languages,
there are often many different ways to do the same thing, some of which may
be more efficient or better suited to a particular task. Greta, it can
certainly be used to split a subject heading at the dashes (much more work
to correctly tag the subdivisions, though, depending on what the source data
is).

Joe Altimus
Arizona State University Libraries

On Fri, Aug 15, 2008 at 8:45 AM, Greta de Groat <gdegroat at stanford.edu>wrote:

> metadatalibrarians-request at lists.monarchos.com wrote:
>
>> ...
>>
>> I took this course http://www.amigos.org/?q=node/698 through Amigos and
>> it was great.  I learned a lot that I've been able to apply directly.
>> Also, lately I've been using the XSLT Cookbook 2.0 as a reference.
>>
>>
>>
> Does anyone offer something similar online?  I'm also finding that the XSLT
> instructional materials and training seem to be geared towards other
> purposes and it's hard to tease out the bits that are useful for our
> purposes.  Which brings up the larger training issue.  In chatting with
> colleagues, i'm finding that that there are lots of times when skills like
> this come in handy, but it's difficult to find training geared towards our
> specific needs.  It's not like we're going to go back to school to learn how
> to be programmers at this point in our careers, but we do need some degree
> of programming skills.  Look at the job advertisements for metadata
> librarians or digital librarians.   A lot of us who have been working as
> traditional librarians feel completely shut out of these jobs because we
> don't have the specific skills asked for.  Speaking from a cataloger's
> viewpoint, we're hearing folks like Karen Calhoun (as well as the training
> coming out from the Catalogers' Learning Workshop) that our skills will be
> needed to manage digital projects and do mapping and that sort of thing, and
> that we'll be working with programmers.  But realistically, programmers
> aren't always available.  It may be easy for a cataloger to transition to
> creating metadata records or managing others who are creating them, but we
> may also need to know how to create valid files of records, troubleshooting
> XSLT tranformations, Perl scripting, OAI protocols, mashups, MySQL, XQuery,
> SRU, and it looks like RDF is coming on fast.   And when it comes up, we
> need to know it right now.   Where can we get this kind of "just in time"
> training geared toward the specific needs we have?  The W3 tutorials only go
> so far.  And asking questions on the blogs that programmers frequent often
> gets you answers that you don't have the technical expertise to even
> understand, much less implement.
>
>>
>> I'd really like to have a "library" of XSLT scripts used by metadata
>> librarians, so we can share/pool our scripts, re-use bits and pieces as
>> needed, and share tips.
>>
>>
>>
> I think this is a great idea.  Where could such a thing be housed?  It
> would also be nice to have a place to ask questions.  For example, the
> question i asked the other day about splitting LCSH headings.  Can i split a
> field at the dashes and separately tag each piece using XSLT? (ok, it's
> probably too much to ask to have the pieces tagged *correctly*, but even
> getting them split would be a step forward).
>
>>
>> I'd like to second/third/fourth MarcEdit. It's a great tool. It was very
>> helpful for me to be able to see the before & after of the
>> transformations.
>>
>>
>>
> This is a great tool, but again, if you don't have troubleshooting skills,
> you have no way of knowing why your transform failed.and what to do about
> it.  Was there a problem of some sort with the style sheet?  Do you have a
> namespace problem?  Was the source file invalid?  Was it invalid because it
> was invalid Dublin Core or MODS or whatever or because it was invalid XML?
> I'm sure there are other problems that i can't even imagine.
>
> Greta de Groat
> Stanford University Libraries
> _______________________________________________
> Metadatalibrarians mailing list
> Metadatalibrarians at lists.monarchos.com
> http://lists.monarchos.com/listinfo.cgi/metadatalibrarians-monarchos.com
>