Mace Ojala: find out why libraries are not Open Data

The Open Data -idea and the movement have been active now for several years. Libraries, and their bibliographical metadata seem like the perfect data to make available in a wider variety of ways; it’s systematically created and maintained, (supposedly) coherent and high quality, well documented, the concept of Open Data is quite easy to understand, the data is propably outside of copyright or the copyrights belong to libraries, it’s creation is typically tax-funded and is of wide interest, Open Data is politically endorsed, there are step-by-step guides how to do it, most if not all of the data is already available online via OPACs and possibly Z39.50 APIs too, libraries claim they are an “open platform” and easily accessible, technological innovation in libraries seems to be quite low… and finally, very few people seem to outright oppose libraries going Open Data.

A handful of libraries have taken the initiative to go their own way and go Open Data. However most libraries haven’t. During the Cycling for libraries -unconference I intend to find out why not.

I believe i can identify top 5 reason why libraries are still not Open Data. Also i hope i can figure out some counterarguments, and perhaps formulate counter-counterarguments to them, together with some other participants of Cycling for libraries.

Mace Ojala

Other Posts

There are 5 comments. Add yours.

  1. Jörg

    till quoted German National Library’s SRU price list. There is a law about the German National Library which requires a “Kostenordnung” (fee regulation statement) http://www.dnb.de/SharedDocs/Downloads/DE/DNB/service/kostenordnung.pdf?__blob=publicationFile

    In this “Kostenordnung” nothing is stated about the cost of data itself but the service costs for users. Surprisingly, the SRU service fees are not mentioned in the “Kostenordnung”.

    As far as I know, library service centers here in Germany offer membership fees for libraries not based on data volume, but on data services. Most popular is shared cataloging online service (as opposed to catalog pooling), where libraries are charged for online access to centralized services, but not for the data they use. The reason is simple: the centralized service is a license product and the service center is charged for the number of users. The fees are charged not-for-profit, they do not exceed the service center costs.

    Having “closed data” or charging externals for data as a revenue stream is very unusual in german libraries since they are tax funded and non-profit organizations.

  2. Thanks for your comment Till. At least you have a price for the records from German National Library, as a pretty well defined product, Dublin Core is free and so on.

    I cannot imagine how other libraries than the National Library would respond if a citizen/colleagues would approach them and ask for access to the data either as a dump, or via harvesting. We have laws that might force libraries to give that data somehow (via a process called tietopyyntö), but those are untested as of yet.

    I would like to know how big a market it is in Finland when libraries transfer data from one another. I know we are paying BTJ to produce the data, also for the national bibliography Fennica. Most libraries don’t do cataloguing themselves, but buy stuff from BTJ or from the academic union catalogue Linda or maybe some of the large public library consortias (that process is broken at the moment, i’m afraid). And i would see no problem in buying BTJ for producing the metadata, if we would just keep the product (ie. bibliographical metadata) accessible.

  3. I don’t see a business case in closed data for single libraries, too.
    But some library data aggregators (like the well known big players, national libraries and for example my own employer) keep telling me, that their (traditional) business model doesn’t allow them to do open data, because selling data is part of their traditional revenue stream. Remembering some numbers I have seen, they may have a point here… It seems, part of my job relies on selling closed data.
    On the other hand, I don’t believe in the future of that business model. We need something different.
    In fact, I think, we have already part of a different business model: What we actually “produce” are services on top of library data. The problem is, that we keep billing the data, not the real costs of the services.
    I think, we bill the data, because we always did it. And because there is the believe, that it is easier to sell a “good” (= records) to libraries (because they can have them in a storage system, count them, delete them, beautify them…) than a service. Records are countable on top, that’s a good naive measure to fix a price… But it’s just hilarious when you look at the price list for use of the SRU service of the German National Library: http://www.dnb.de/EN/Service/DigitaleDienste/SRU/sru_node.html (13 Euro to retrieve 100 records, at least they give a big discount per record if you want more…, I wonder if they deliver them in a nice jewel box or so).
    With some new services we try to bill the actual costs in a transparent price model (basic setup and maintenance fee, all extras billed by effort, no matter how big you are or how many records you retrieve). Surprisingly enough (for some) it works. If you communicate it clearly, libraries even seem to like it.
    If that works, why not open the data? They will still need the services anyway…

  4. During Cycling for libraries i spoke with a number of people, trying to search for arguments how their organizations justify saying “no” to Open Data. Unfortunately i failed to find any formulated arguments against Open Data. Surely several people were familiar with the concept already, and when i explained it to those who weren’t, everybody thought it’s a great idea, and said naturally libraries should do it straight away.

    However none of the people were able to quote an official argument why their organisations weren’t committed to Open Data. Ignorance, unfamiliarity, fear, resistance to change, lack of resources, outdated library directors, elitism, technological infantilism, simple hesitation and other such typical reasons were speculated.

    Like for many years in library conferences, at IFLA WLIC 2012 in Helsinki, the concept of Linked Open Data (“LOD”) was presented in several sessions. Part of this discourse are the copyrights and licensing for library’s bibliograhic metadata. Perhaps there is a businesscase in bibliographic metadata that libraries want to protect by hindering access to bibliograhic data by licencing and technological means. Individual metadata records are not protected copyright, but a database is and that copyright belongs to it’s maintainer; f.ex. a library who might want to guard it’s investment in this intellectual property.

    Personally i of course think these protectionist measures are hurting libraries, the Internet and the society. I doubt there is a businesscase in library metadata, especially since libraries seem reluctant to sell it to a willing purchasers (f.ex. me).

    If you have arguments against making library data more accessible, i would be very happy hear about them!

Leave a Reply

Your email address will not be published. Required fields are marked *

Categories

Archives