U Buffalo's new native XML catalog?
By ksclarke, Section Mindful Link Propagation
Posted on Thu Nov 13th, 2003 at 09:18:21 AM EST
Mark Ludwig at U Buffalo has done some interesting work on moving their OPAC to a native XML document store. The demo/test version is available for previewing/testing. What follows in the "extended copy" are snippets from a Nov. 7th, 2003 meeting (whose notes were on the Web). It seems like feedback on the project has been good and that people are offering valuable suggestions.
|"It [the new catalog] starts with the NOTIS-based catalog (though any catalog of MARC records would serve) and takes data from it. The ULLOG program looks at current activity on NOTIS and updates every night; the ULHAS program deals with circulation activity. Data flows from BISON through those programs and are formulated into three sub-files:CATPAGE, CIRC-STAT, and PATRON, and from there into the TextML documentbase and into indexes on a Windows 2000 server. The server allows us to provide various web services, e.g. user interfaces. The services go through a web server to the users' PCs.
The indexes include author browse, title browse, subject browse, keyword, language, location, year. The keyword index covers call number, location, and status information.
The ULHAS program provides patron empowerment, allowing patrons to see what they have charged out, what is overdue or been recalled, what they have on Hold, etc. This data can be updated twice a day. The display shows: "Current as of" date and time so the patron knows how up-to-date the information is.
The web catalog not only displays URLs but also shows what type of link each one is: table of contents, book review; publisher description; full text, etc. Thumbnail pictures of book covers are provided when available, one for each ISBN."
Though the catalog is still in its preliminary stages, it looks very promising. Here is the schedule from that same meeting's notes:
He has provided the user interface which contains features such as:
- bookmarklets for search re-direction
- design finalization & integration
- report forms
- date & language
- security for patron has & renewal request
- design finalization
- comprehensive holdings data is being tested
- current issues bug
- help & documentation
- proxy script
V. 2.0: will have:
Version 1.0 Benefits:
- 24x7 availability
- current indexes
- patron empowerment
- book covers
- google-like interface
- pure 'native' XML technology
This is interesting to me because I've always thought of cataloging (in its current descriptive form at least) as text-centric. In cataloging, records are touched infrequently, while ACID databases are usually used for heavy write activity. The one question using documents raises, for me (and for which databases are useful), is authority control. It will be interesting to see how they handle the issue. Authority control was mentioned in the meeting's notes so it is something they are already considering.
This system is also interesting because it is lightning fast. I do not like the frames much, but it will be an interesting project to watch as they push the boundaries of what we think of as a library catalog...