Very well reasoned and thoughtful responses, Richard. Thanks! I'm very glad
I set this thread rolling and that we're getting some great ideas out there.
A couple of thoughts:
The rationale for using a community standard like LCSH, as you said, is
sharability and authority control. LCSH, while sometimes obscure, allows you
to build complex terms in structured ways that others will (theoretically)
mimic when they catalog similar things. Ideally, we should all have CMS and
DAMS that will allow us to both enter a term and enter its source, i.e.
Steel Industry--Pennsylvania--Pittsburgh. [LCSH]
Steel factories [local term]
My Uncle Steve's workplace [folksonomy] ;-)
The search screen could allow you to browse (the power of LCSH, in my
opinion) or do keyword searches. I like the idea of identifying ofness and
aboutness. Any suggestions how to indicate this in a data structure? Library
cataloging doesn't really let you, in my experience, so there isn't an easy
translation there. Maybe something in VRA?
As to labor intensiveness, there's a lot to be said for laying a ground work
and doing the basics as a batch process. Set up some ground rules: for
example, all objects from the Asian Art department identified as Chinese
furniture get subject heading xyz, and write a small program to find those
records and add the heading. Then, if you have the resources, have someone
look at each record and add something more detailed. I'm cooking up a scheme
to pair a curatorial intern and a library school intern--the curatorial grad
student says "this is a ...." the lib student works out an authorized
heading. Would take some curatorial QC, but could be effective, I think.
Looking forward to working this out and catching up with the libraries and
archives (my background, before I moved over to the digital imaging side)!
Deborah Wythe
Brooklyn Museum
Head of Digital Collections and Services
718 501 6311
----Original Message Follows----
From: "Richard Urban" <[email protected]>
Reply-To: [email protected]
To: [email protected]
Subject: RE: STEVE & folksonomies / was subject & keyword searching in CMS
and DAMS
Date: Wed, 23 Nov 2005 13:17:55 -0600
Will, et al.
First off...this is a great thread. We need to have more of these kinds of
discussion here!
<snip>
A few questions:
"Does anyone have an opinion about the value, in the networked information
world, of the hierarchical LC subject format I described above ("Steel
Industry--Pennsylvania--Pittsburgh.")? Are others using this format (and
why) or are you using single terms, more like keywords?"
The problem that I see in these discussions is that those not steeped in the
cataloging tradition don't often see the LCSH as a larger social system of
collaboratively creating a common set of terms. There are, no doubt,
challenges with using LCSH that derive from what LCSH is. (And I'm going
out on a limb here. LCSH isn't covered in my cataloging class until next
week....corrections welcome) LCSH subject headings aren't just made up willy
nilly, they're based on the concept of "literary warrant" or that the terms
used are actually represented in the body of materials being described.
For bibliographic texts there's a leading organization and a large group of
users, following a common format that debate the addition/deletion and
change of terms based on the bibliographic materials they see. I'm not
exactly sure how visual materials feed into this process, but the bulk of
LCSH is likely to be based on texts, rather than images. It often looks like
madness, but there is method to it.
The question seems to suggest whether we can/should develop a "visual
literary warrant" for describing the "ofness" and "aboutness" of the
materials we're describing. Things like Cataloging Cultural Objects (CCO)
are an important step towards that goal because they provide guidance and
some liberal constraints on what kinds of controlled vocabularies are used
for subject description. LCSH is not a magic bullet, but an appropriate
controlled vocabulary is going to offer some advantages over "keywords".
Namely:
1. Interoperability
We're living in a networked environment. Many of us will be sharing our
records outside the contexts of our local environment (if not today, in the
future) and using common vocabularies will make the job of aggregating and
providing access much easier. A set of keywords, or even a folksonomy
created by a specific community may not always make sense outside of that
community (and I'd say this goes both ways...LCSH doesn't always make sense
outside of it's community, useful as it can be). We're already seeing the
interoperability challenges even when controlled vocabularies are used. See
for example some of the work being done through National Science Digital
Library OAI Best Practices/Shareable Metadata Best Practices
http://oai-best.comm.nsdl.org/cgi-bin/wiki.pl?PublicTOC
2. Clustering/Collocation
Using a limited, controlled set of terms means it will be easier to
establish relationships between individual items and put like things next to
each other. This is more difficult when uncontrolled sets of keywords are
used.
3. Management
The other benefits of things like LCSH is that (in the right contexts), is
the possibility to do authority control. If terms change over time it's
easier to update, change and transform them if they've been properly
applied. How many of us are hamstrung on sharing information about our
collections because of the amount of "clean-up" of uncontrolled
terms/descriptions.
So what about folksonomies and keywords?
I think the development of folksonomies is very exciting. It allows us to
leverage large social groups to do what we don't always have the staffing to
do and it puts description in the hands of end-users.
And as some have suggested it's not an either-or proposition. I like to
think about it like the Congress. The Senate (in theory) is supposed to be
a slower-moving more deliberative body, whereas the House has it's finger on
the pulse of the people. Call it "bicameral cataloging" that takes the
benefits from persistent standards-based description and more flexible
approaches such as folksonomies.
In response to Will's question about where folksononmic terms live, I can't
say. However I'd say don't try to mix them with other terms. Being able to
identify where terms come from will be important to the long-term usefulness
and management of the records we create today. Maybe they get indexed
together in a searching service, but in a record they should be clearly
distinguishable. It will be interesting to see from STEVE and other
projects whether this kind of social tagging is also useful to staff
functions, which may suggest it's useful to include in collections records.
Looking forward to where this discussion goes...
Richard Urban
Graduate School of Library and Information Science
University of Illinois at Urbana-Champaign
[email protected]
---
You are currently subscribed to mcn_mcn-l as: [email protected]
To unsubscribe send a blank email to
[email protected]
---
You are currently subscribed to mcn_mcn-l as: [email protected]
To unsubscribe send a blank email to
[email protected]
