A couple of times in the past I’ve highlighted the NGC4LIB mailing list as a place of great discussion.  I thought it had gone a little quiet over the last few weeks.  Then last week Tim Spalding of LibraryThing fame woke the list up with, what on the surface seemed, a simple enquiry.

Does anyone know of examples of a fully-spiderable OPAC?

It’s my contention that libraries would do well in Google and even Google Local if they were spiderable. I’ve seen the Lamson Library catalog do very well—tops in Google, even without mentioning Plymouth State, but it gets a LOT of push from its association with WpOPAC.

But I need some examples. Anyone?

In clarification of his question, Tim went on to use the analogy of being able to search for ‘Pizza Portland, ME.‘ which works for finding what he wants in his locality, whereas [because OPACs are not spiderable] ‘Book Title, mytown’ doesn’t.  I think Tim is misrepresenting the problem - the library equivalent of his pizza search should be ‘Library Portland, ME‘, as [from a search engine point of view] pizza is mapped to places that sell pizza.  Try searching for ‘“ham and mushroom” Portland, ME‘ and see what you get, a few pizza joints but mixed in with a lot of other stuff.

The problem is that search engines are very good at identifying a page that references the thing you are looking for, but less good at finding the thing itself.  I agree with Tim that if OPACs were search engine spider friendly you may well get better hits than we currently do, but there again how many OPAC interfaces are mapped directly to an address so that the search engine can do its geo-location magic - very few.

Later on in the NGC4LIB thread, the OCLC Find-in-a-library service was mentioned as being the answer - but only the answer if the library funds that service by being a member.  There are many stories even in the US of the ‘find’ being a University in the next State.  Try it outside the US, and even with a modicum of geographic knowledge you could get the paranoid feeling that the results are there to taunt you instead of being helpful.

All very well to criticize what we have now, but far more valuable is the search for a solution, which is what I believe is behind Tim’s question.  The problem I believe is that there is not a library equivalent to Amazon or Wikipeadia to point your searches at.  How many of us, let alone the world beyond libraries, link to a book reference on a library page from their blog. No we all point at Amazon etc.

To be fair, unlike Talis [which has had static URLs for book references in its Web OPAC since 1995], most OPACs are not linkable anyway.

What we need are a small number of places on the net which have not only references to all the books held by libraries, but also an easy low-cost [preferably zero cost] way of linking in with individual library holdings.

Casey Durfee, in his contribution to the thread (which by now had morphed to have the title Spiderable OPACs and the elephant in the library lobby - the elephant in this case being OCLC) said:

There are a lot of reasons to be wary of a “one catalog to rule them all” solution being sold by a single vendor no matter how appealing the idea might be. [my emphasis]

There seems to be some sympathy with the idea of a centralized [spiderable] place to search for all bibliographic references, but warning bells ringing around the commercial and/or political ramifications of seeding that with any particular body.  Some people chipped in with “sounds like OCLC to me” comments, but were soon answered by those who are not, and cannot afford to be, members of OCLC. 

For something like this to work there are three hurdles to overcome:

• Firstly, and most obviously, cost.  For the contributing libraries the only differentiation between those that have and those that have not [got their holdings represented] should be their willingness to contribute.  Any cost hurdles almost by definition lead to an information visibility apartheid - those that cannot afford to contribute, disappear.

  Albeit UK only at the moment, our own Talis Source example ably demonstrates that if you remove the costs of contributing to a resource that can deliver value to library users [holdings contributed to the Talis Library Platform, via Talis Source, are publicly searchable in Talis Cenote] libraries are motivated to openly share their data.

• Secondly, fear of loss of data ownership.  There has been much concern over the years around evil people making money out of freely shared data.  The answer is licensing.  Creative Commons has shown that you can share your creative works for the benefit of all whilst still precluding the possibility of others making money from your efforts.

  The Talis Community License is our contribution to the community to deliver the same protection to aggregations of data that may not individually be protectable in the same way.

• Finally, cataloging practice.  Getting international agreement about the correct form of catalogue record to describe an individual book has up until now been akin to trying to herd cats [click the link, it is amusing, even if the IT company it is advertising is being a bit optimistic I think!].

  This is not surprising as a main job of a cataloger is to catalog a book in their institution, and not necessarily to help a user on a different continent find it.

Have I got any answers to the issues I’ve referenced?  I only wish I had, but what is gratifying is that people are engaging in the debate and starting to dismiss some of the ‘obvious’ solutions.  I believe the answers lay with Openness of data, licenses, and minds.  I also believe that the enormous amounts of money changing hands for looking after and allowing access to metadata is a model that will become untenable in the future.

Overcome some of these issues, learn from our experience in the UK, and we may end up with “One catalogue to enable them all“

(Photo taken by generalnoir displayed in Flickr)

.

Technorati Tags: Talis, Library, Open Data, Catalog, NGC4LIB, OCLC, Creative Commons

Share This