Friday 10 September 2010

Loss of information from the record slip on a library book

Since the OU library has gone over to 'self service' books don't have the return date stamped on them. This is a loss of information, because it used to be that when I took a book out I could look on slip in the front to get some idea of how much interest there has been at the university in that subject. (I used to get perverse satisfaction when I found that I was the only person to have borrowed the book in, for example, the last 20 years!)

To be honest, this was probably only 'of interest', I don't know that I ever made explicit use of it, but it is something that's gone. Well, I guess the information is there still, electronically, and if I had access I could even get more information from.

Confidentiality would mean I wouldn't be allowed to know who had borrowed it, I assume, although that info presumably exists on the system*, but even information like what departments the people who borrowed it were in would be useful.

My real point, though, is that while that info is probably there I can't readily access it, unlike a glance at the slip.

*Do I recall that some libraries deliberately don't keep that level of detail? I seem to remember that some destroy it so that the authorities cannot come demanding to see it? I think something along those lines was recounted in The Virtual Revolution about the library at Santa Cruz, but I also think I heard something about it more recently.

3 comments:

Stephen said...

Going back even further, in an academic context I used to appreciate the borrower names on a book's lending card. You could quickly work out a degree of relevancy by examining who had previously borrowed the book. Of course today that would be considered a gross invasion of privacy.

David Chapman said...

For some reason blogger deleted this comment, so, with the author's permission, I'm trying to add it myself (if anyone knows why blogger removed it, please let me know!)

Owen has left a new comment on your post "Loss of information from the record slip on a libr...":

In the US libraries generally remove records of who borrowed which books quite promptly in order to avoid being put in a position of having to reveal this to authorities under the patriot act, or other legislation. Although strictly it isn't necessary to remove all the data, as you could preserve anonymous transaction information, deletion of all the relevant data is often the easier course of action. If you are interested, this is a short article from 2003 that looks at the issue http://www.bibliomining.com/nicholson/nicholsonpdfs/preshred.pdf

In the UK, my personal experience is that libraries have been a bit more relaxed, although the Data Protection act says such personal data should be kept 'no longer than necessary' (this isn't defined.)

More recently there has been some work to look at how libraries could share interesting 'usage' data without publishing confidential information. Dave Pattern and the University of Huddersfield have led the way on this, publishing circulation transaction data for reuse - see http://library.hud.ac.uk/data/usagedata/_readme.html

A JISC funded project called 'MOSAIC' (http://library.hud.ac.uk/wikis/mosaic/index.php/Main_Page) investigated how such data might be exploited, which resulted in a number of demonstrations - http://devcsi.ukoln.ac.uk/demonstrator/2009/10/22/jisc-mosiac-project-competition-winners/

More recently the University of Lincoln have followed the lead of Huddersfield, but unfortunately (in my opinion) we haven't yet seen other libraries take the same step.

Further work by Dave Pattern and the Huddersfield team has shown correlation between the introduction of a 'recommender' service based on their transaction data and an increase in diversity of materials borrowed from the library as well as further correlations between library use and increased final grades.

This is an underexploited area, but one which is generating some interest in the HE library sector - so I hope we will eventually see more institutions follow the Huddersfield/Lincoln example, and publish anonymised data for both local use, and public reuse.



Posted by Owen to Intropy at 13 September 2010 10:09

David Chapman said...

Thanks for that, Owen, I'm glad I've finally got that comment!

The OU library seems to have been quite forward-thinking when it comes to exploiting new technologies. It would be nice if they were able to try something like this out.