inkdroid

for something completely different ….

This morning Clay and I were chatting about Library of Congress Subject Headings and SKOS a bit. At one point we found ourselves musing about how much reuse there is of topical subdivisions in topical headings in the LC authority file. You know how it is. Anyhow, I remembered that I’d used marcdb to import all of Simon Spiro’s authority data–so I fired up psql and wrote a query:

SELECT subfields.value AS subdivision, count(*) AS total
FROM subfields, data_fields
WHERE subfields.code = 'x'
  AND subfields.data_field_id = data_fields.id
  AND data_fields.tag = '150'
GROUP BY subfields.value
ORDER BY total DESC;

And a few seconds later…

 subdivision                          | total
--------------------------------------+-------
 Law and legislation                  |  3342
 Religious aspects                    |  2500
 Buddhism, [Christianity, etc.]       |   898
 History                              |   847
 Equipment and supplies               |   571
 Taxation                             |   566
 Baptists, [Catholic Church, etc.]    |   476
 Diseases                             |   450
 Research                             |   422
 Campaigns                            |   378
 Awards                               |   342
 Finance                              |   284
 Study and teaching                   |   284
 Surgery                              |   275
 Employees                            |   269
 Spectra                              |   261
 Computer programs                    |   259
 Labor unions                         |   218
 Testing                              |   207
 Diagnosis                            |   194
 Isotopes                             |   190
 Complications                        |   183
 Physiological effect                 |   172
 Programming                          |   163

There’s nothin’ like the smell of strong set theory in the morning. Although something seems a bit fishy about [Christianity, etc.] and [Catholic Church, etc.]… If you want to try similar stuff and don’t want to wait hours for marcdb to import all the data and you use postgres, here’s the full database dump which you ought to be able to import:

  % createdb authorities
  % wget http://inkdroid.org/data/authorities.sql.bz2
  % bunzip2 authorities.sql.bz2
  % psql authorities < authorities.sql

In case you missed it the Object-Reuse-and-Exchange (ORE) folks are having a get together at Johns Hopkins University (Baltimore, MD) on March 3, 2008. It’s free to register, but space is limited. The Compound information objects whitepaper, May 2007 Technical Committee notes and the more recent Interoperability for the Discovery, Use, and Re-Use of Units of Scholarly Communication provide a good taste of what the beta ORE specs are likely to look like.

The ORE group isn’t small, and includes individuals from quite different organizations. So any consensus that can be garnered I think will be quite powerful. Personally I’ve been really pleased to see how much the ORE work is leaning on web architecture: notably resolvable HTTP URIs, content-negotiation, linked-data and named graphs. Also interesting in the recent announcement is that the initial specs will use RFC 4287 for encoding the data model. Who knows, perhaps the spec will rely on archive feeds as discussed recently on the code4lib discussion list.

I’m particularly interested to see what flavor of URIs are used to identify the compound objects:

The protocol-based URI of the Resource Map identifies an aggregation of resources (components of a compound object) and their boundary-type inter-relationships. While this URI is clearly not the identifier of the compound object itself, it does provide an access point to the Resource Map and its representations that list all the resources of the compound object. For many practical purposes, this protocol-based URI may be a handy mechanism to reference the compound object because of the tight dependency of the visibility of the compound object in web space on the Resource Map (i.e., in ORE terms, a compound object exists in web space if and only if there exists a Resource Map describing it).

We note, however, two subtle points regarding the use of the URI of the Resource Map to reference the compound object. First, doing so is inconsistent with the web architecture and URI guidelines that are explicit in their suggestion that a URI should identify a single resource. Strictly interpreted, then, the use of the URI of the Resource Map to identify both the Resource Map and the compound object that it describes is incorrect. Second, some existing information systems already use dedicated URIs for the identification of compound information objects “as a whole.” For example, many scholarly publishers use DOIs whereas the Fedora and aDORe repositories have adopted identifiers of the info URI scheme. These identifiers are explicitly distinct from the URI of the Resource Map. from: Interoperability for the Discovery, Use, and Re-Use of Units of Scholarly Communication

I understand the ORE group is intentionally not aligning themselves too closely with the semantic web community. However I think they need to consider whether compound information objects are WWW information resources or not:

By design a URI identifies one resource. We do not limit the scope of what might be a resource. The term “resource” is used in a general sense for whatever might be identified by a URI. It is conventional on the hypertext Web to describe Web pages, images, product catalogs, etc. as “resources”. The distinguishing characteristic of these resources is that all of their essential characteristics can be conveyed in a message. We identify this set as “information resources.” (from Architecture of the World Wide Web vol. 1).

I’m not totally convinced that the resource map can’t serve as a suitable representation for the compound information object–however for the sake of argument lets say I am. It seems to me that the URI for the compound information object identifies the concept of a particular compound information object, which lies in various pieces on the network. However this doesn’t preclude the use of HTTP URLs to identify the compound objects. Indeed the What HTTP URIs identify and Cool URIs for the Semantic Web provide specific guidance on how to serve up these non-information resources. Of course philosophical arguments around httpRange-14 have raged for a while. But the Linking Open Data project is using the hash URI and 303 redirect very effectively. There has even been some work on a sitemap extension to enable crawling. As a practical matter using URLs to identify compound information objects will encourage their use because they will naturally find their ways into publications, blogs, other compound objects. Using non-resolvable or quasi-resolvable info-uris or dois will mean people just won’t create the links–and when they do they will create links that can’t be easily verified and evolved over time with standard web tools. The OAI-ORE effort represents a giant leap forward for the digital library community into the web. Here’s to hoping they land safely–we need this stuff.