Because I was looking to see what people are planning to do to replace or evolve their current “library management” or backend type systems, I attended a number of sessions from which a pattern seemed to emerge.  System designs had shared pools of bib and eresource data, with local information for Circulation, Acquisitions, etc. linking in or getting loaded in.  So OCLC Worldcat Local is based on the OCLC database.  But when Andrew Pace spoke on “OCLC Web scale computing” where he talked about adding Circulation, Acq and ERM, he mentioned a “Knowledgebase” to help with eresource management.  That made me wonder, is OCLC going into the Knowledgebase business that I think has been dominated by Ex Libris SFX and Serials Solutions?

Meanwhile, when Oren Beit-Arie of Ex Libris presented their URM Universal Resource Management  framework, it also contained a box for a shared data pool of bibliographic and eresource Knowledgebase data.  He called this DaaS  Data as a Service.  My colleague Janet Fox remarked that the whole presentation was done as if OCLC did not exist.  That led me to speculate, do they plan to use some separate database and does this become like RLIN as a database of big academic libraries?  This was not addressed in the session but made me wonder.  Oren did comment that any data in that data store would be owned by the libraries involved and would not be claimed by Ex Libris.  OCLC really shot themselves in the foot with the flap over record use policy even though they have stepped back from that.

Another similar database is the Serials Solutions one.  Jane Burke demonstrated the 360 Resource Manager that works off their Knowledgebase at a presentation for the CARLI libraries.  And I got a demonstration of Summon which is similar to Worldcat Local.  You can upload your library’s bibs, combine them with journal article and ebook and other eresource data.  A distinguishing feature seems to be that in Summon you can limit to eresources that you own.  I assume the Knowledgebase helps with that.  Also, Summon indexes full text of articles not just metadata.  Worldcat Local plans to do that in the future.  But the end result is that Serials Solutions will also have a big data pool of mixed bib and eresource data.

I also had a discussion with Bob Molyneau of Equinox, a company that supports Evergreen open-ils.  They have a consortial enhancement for Evergreen called Fulfillment that is being done for some of the State Libraries in the Midwest.  Bob speculated there might be other uses for a consortial layer on Evergreen.  For instance, instead of complicated automatic prediction to help claim serials issues if they are not received, you might have a consortial layer service that recognized when one library had received the piece and claimed if others had not.

In the Ole presentation by John Little of Duke University, I was not sure if they planned that type of shared data pool of bib and Knowledgebase information.  The design document is due out later this month.  Certainly one of the first pieces they want to develop is for management of Eresources both owned and licensed.  And he mentioned the CUFFS ERM Knowledgebase.  OLE is a project to try to produce an open source library management system for academic institutions.  The structure of the project is based on other Kuali projects where for the first few years the development agenda will be driven by the “build partner” libraries that invest in the early development cycles.  They will contract out the development.

Meanwhile, LibLime, a company that supports the Koha open source library system, got frustrated and put up biblionet, a shared pool of bib data that is separate from OCLC.  I haven’t heard that they  have any plans for an eresource Knowledgebase at this time, but it is another example of this trend.

The name of my post, “Just say no” vs. “Yes, we can” refers to the different ways of approaching developments and a conflict in the market place right now.  For instance, Andrew Pace suggested that libraries would need to standardize some of their processes by moving to a Worldcat Local type of system.  He said that everybody hated “circ tables”.  He also suggested that for library fund codes libraries need to “work towards homogeneity”.  He seemed to be saying give up customization for cost savings of centralization.

On the other hand, at sessions about Open Source in libraries, there is open frustration with vended systems where you pay a lot of annual support only to be told “NO” when you wanted to do something.  Another system vendor was explaining that they wanted to reduce the number of options in order to make sure all new features were available to most customers and support was better.  That certainly makes some sense.  But I think it is the wrong time for vendors to be doing the “just say no” piece and disallowing customization and control by the library.  There certainly are opportunities to get rid of unnecessary complexities, on the other hand the library management system is there to support the work of the library.  Policy decisions, like circulation policy need to be made so that the library provides the services their customers need, they should not be limited by system limitations.  Many libraries may get by with a fund code of just “gift” or something simple.  A large library that receives hundreds of thousands of dollars to acquire materials from gifts of endowment money must be able to track that in a way that meets their needs .  “yes, we can” is the necessary mantra, not “just say no.”

In the past, vendors limited what could be customized to support lots of libraries.  But the technologies involved have changed.  I think younger technologists come into the library setting and say, why are you living with all these restrictions?  You could do what you want to do.  You can do what the vendor is telling you you cannot do.  I believe this is why a number of people think Open Source systems will be a better solution for the future.

But what if we end up with a lot of separate pools of data, and not one central database?  Well, another buzz word this year is “Linked Data”.  The way I think of that is, instead of allowing outside applications to dip into my database of information about my library holdings, I dump the information I want to expose in an XML file and let people harvest, crawl, pull off pieces of it as they like.  For instance, a Google Book Search engineer suggested if libraries put out their holdings information that way, it could be crawled by Google and other search engines and would lead people back to libraries from searches on the net.  If the libraries want one big central db, we could put out our information and let OCLC or someone else crawl it.  You don’t absolutely have to all be making the records on the same system to combine them in useful ways.

One of the most interesting comments I heard was at LITA Top Technology Trends.  Clifford Lynch of CNI  suggested there is an enormous mass of literature that embodies the past.  But for scholarly materials there are very new open collaborative ways to author and communicate.  Eventually it will be very problematic to link between scholarly literature of the past and future scholarly production.  Now there’s something to think about.

OCLC Webscale services
Presented by Andrew Pace this is about development of Circ, Acq and ERM add-ons to OCLC Worldcat Local.  In April 2009 they announced a “Quickstart” strategy to get libs to sign up for worldcat local.  Now he is concentrating on a Knowledgebase for ERM and Lib management systems for Acq and Circ.  He cited Fred Kilgore’s original plan to put all parts of the ILS online but said OCLC had failed to offer circ systems because they don’t do well supporting systems run out in the libraries (He got corrected from the floor later.  Acq and Serials Control were offered centrally online and were migrated off to local ACQ350, SC350 systems, maybe Andrew is too young to remember).  He was suggesting there would be new workflows and they would not preserve old workflows.  He actually made the comparison to something you would just plug in.  So Circ, Acq and ERM will become plug and play modules that run on the central OCLC servers.  He thinks this will “free” libraries and will reduce the Total Cost of Ownership.  He claimed Open source would have a lot of overhead.  He said this OCLC system would be SOA to interactw tihe student registration and course management systems.

Circ will not change, Acq is an opportunity to change adding ERM, license management.  He sees an opportunity for “cooperative intelligence” for consortial fund management, collection development, recommender systems based on circulation in various libraries.  I have a problem with this.  If you think about concerns people have about commercial sites like Google collecting a lot of information about people who use it and using that information without the knowledge or consent of those users, are we really supposed to trust OCLC with sensitive fund information and assume they will use that wisely and in our interest?  I actually asked about what if OCLC becomes a monopoly with circ, etc. and jacks up the price.  they are not doing this out of the goodness of their hearts so what would prevent them from price gauging.  Andrew said he had goodness in his heart and that the “collective” would be responsible.  I’m part of the OCLC “collective” now.  I don’t feel I have any control over it when they decide to raise their prices.  One of my colleagues, Chris Cronin, asked about whether Open Source wasn’t the way to go if this was offered as a “collective” thing.  Andrew answered that they used open source software for some of the pieces, and OCLC Developer’s network is supposed to encourage community development.  He did not say they would open source the code for these new systems.  He did emphasize that the library’s data is the library’s data and what goes in must go out also.

Next steps are a Library Advisory Council (I don’t believe there were any ARL size librareis); a Steering committee of more handson types of librarians.  3 pilot sites, Idaho community libraries, Orbis Cascade Linfield College and Pepperdine University.  A public library will be added.  they are using agile programming methodology.  System configuration is implemented as part of the Worldcat Local quickstart.  Circ will test from this month to mid 2010.  License manager due late 2009, and Print and electronic acquisitions the beginning of 2010.  He referred to his blog for more info.

In response to questions he claimed all people hate Circ tables and mentioned “finite administrative setups.”  There would be “Service Level Agreements” with OCLC that would cover some privacy issues.  For library fund coes he thought libraries should work towares “homogeniety”.  He mentioned Marshall Breedings suggestion that the Borrower db become a “Customer relationship db”.  They will have Acq in a year, selective rollout of the whole ILS replacement FY 2011 and scaled rollout after that.

I am not do not like the forced lowest denominator approach.  I think this attitude is what is driving libraries to Open Source systems.  To pay a large amount of support only to be told what you CANNOT do in a system is straining the patience of libraries.  I certainly think the next set of systems will be leaner and have fewer bells and whistles but the limitations of a vendor run system cannot be allowed to dictate what a library can do for circulation policies or use of funds.  For lots of libraries they loan for a period.  For very large academics, you loan for the whole quarter or semester and allow recall when something is needed.  This is because the primary users are doing in depth research in unique areas.  For funds some libraries might just use a few.  Some large sites may have elaborate and extensive funds for endowments where they have received gifts of money to purchase materials.  This might be regularized if it amounted to a few thousand dollars of purchases each year, but where it is hundreds of thousands of dollars of purchases a library will not be forced to stop receiving and reporting on such funds just because other libraries don’t do it that way.  That is not what I would call progress.  This vendor insistence that they know the best way for you and stuff it if you don’t agree is not something people are willing to put up with any more.  So I believe for a lot of libraries this approach will not work.  Some of the borrower and acquisitions data also just may not be appropriately stored at the level of OCLC Worldcat.  Even the circ data has privacy issues that are controlled by state legislation in some instances.

LITA NGCIG Post Integrated Library Systems:  The Open Library Environment (OLE) and the Unified Resource Management (URM) Projects

Oren Beit-Aire and Kathryn Harnish from Ex Libris spoke first.  Much of this information is available on their website.  Oren mentioned a need to accomodate escience data in the next 5 years although I’m not sure exactly where that fit in.  There are charts of the architecture, there is an URD2 which is essentially Primo, and a URM for the management of print and electronic and “new types” of data.  This represents the “decoupling” of front and back ends of the old ILS.  There is another box for the Data Sources including metadate, usage data, the Knowledgedb, Vendor Info, Tags and Reviews.  this is a shared, “community” source not local to the library, managed by ExLibris or a “community of users”.  [My colleague Janet Fox, said, they did that whole presentation without mentioning OCLC.  Are they putting up a share db in competition to OCLC?  like the old RLIN?  Or like the biblios.net that LibLime has put up?]  He called this DaaS  Data as a Service and made the point that no restrictions would be put on use and reuse of the data.

It is to be an open platform, it is SOA, supports ‘collaboration of choice’.  Kathryn went on to talking about putting ebooks in a ‘community zone’ while print would be local.  She showed a ‘cataloging lobby’ and a workform that included RDA elements.  There are forums for users to ask questions (biblios.net has that).  They want to support new uses like collaborative collection development.  Princeton, Boston College and a Belgian site are helping with development.  \

John Little spoke about the OLE project.  They use a “community source” rather than “open source” model.  They are looking for build partners who would decide what features are built in the first 2 years and the code would not be open for others to use/modify until later.  Initial design will be availabe later this month on the OLE web site.  They want a reference implementation up by mid 2011.  That will have only what the build partners agree should be in the first release.  John said they can meet this deadline because they will build on top of the Kuali Rice middleware for nervous system, system bus, enterprise workflow, enterprise notification, id management.  The will contract out the programming of the library app to workers in Bulgaria.  there will be a single project manager here.  this is based on Kuali Coesus Research Administration system.  That was developed in this manner, outsourcing the coding to India.  He also mentioned useing the CUFFS ERM Knowledgebase.  The system will not provide a discovery layer, a library would use Worldcat Local, VuFind, Blacklight, eXtensible catalog, Primo …

Build partners will decide what gets coded first.  Interested libraries include Indiana, UPenn, Duke but none are announced partners yet.  They want 5-7 partners.  Cost is 185K to 260K per year depending on number of partners.  Partners would need to put part of system in production and contribute staff resources as well.  John said according to Mellon none of their projects has failed and they have yielded 35 to 90% cost savings to institutions.  Benefits to build parners are access to emerging technologies, use monetary resources in a productive and directly influencial fashion, leverage ROI on campus for enterprise systems.  He commened his staff were enthused about working on new systems.  Secondary benefits are exposure, staff enrichment, productivity and a culture of strategic innovation.

Oren did point out that many universities don’t use Kuali so in reality you don’t get to incorporate Acquisitions into an enterprise wide system.

LibraryThing

There was a demo at the Bowker booth of Library Thing for Libraries.  There are 53 million tags.  Penntags in 3 years added only as many as get added to LT in a day.  there are 42 million books and 750K members.  Terms like “chick lit” help public libs to supplement subject headings.  For academic libs new terms and terms not in LCSH can be helpful, liek “social capital”.  linking is not restricted to ISBN, uses LCCN, OCLC no, author/title.  He also showed how you can vet reviews from your own users and you can show only yours, only ones from library catalogs or all reviews from LibraryThing.  Bowdoin College has a good example.  It is just a piece of javascript on a screen and CSS is used ot configure it.  There are excellent stats on what is used and you can import those stats to Excel.  All reviews are moderated, looked at.  You have to look at your own or pay them to do it.  There is a Facebook App so people can feature what they reviewed at their library and link back to the library as they share lists with others.

Ingram Lightning Source

They had the Espresso machine printing books.  It took about 5 min for a 169 p. book.  Prints from PDFs.  Prints anything in the Lightning Source db.  I talked to one of the salesmen for My iLibrary.  For ebooks, publisers were charging 150% of hard copy price.  This year it is 120%, he expects it to be even next year.  They have a deal with Springer to provide POD Print on Demand, and they are getting that deal with other publishers who want to publish e-only but allow an option to print.  Espresso costs $80-100K.  It’s out there in about 20 sites they had listed.  There has to be annual maintenance on the machine as well.

LITA Program Planning Committee I left with the charge to manage revision of the LITA manual pages for program planning for the annual conference.

LITA Top Technology Trends.
On mobile technology someone envisioned patrons being able to download to their mobile device from a workstation in the library. Clifford Lynch made the point that while mobile devices are ubiquitous actual computation is moving off devices and on to the “cloud”. but Preservation role of Research Libraries is not transferable to commercial vendors. So what info will we put in the ‘cloud’? We have yet to see a big scale failure there but it will happen and it will change how the ‘cloud’ is used. On mobile devices it is easier to capture video and images than text. This could impact communication.
Eric Lease Morgan commented if it’s in the “cloud” you are not preserving it. He predicted “open everything” in 5 years library software will be turned on its head, libraries will own and develop more of theri software. Libs will band together to get Elsevier to lower their prices. He did add that “open” needs to be tempered with reality and a mixture of open source and proprietary systems will be in use. He complained that libs don’t have the hutspah to force vendors to lower prices.
Joan Williams thought the vendors would target endusers and skip libraries. She thought Ebsco’s ads on NPR were an indication of that.
John Blyberg said enduser tools are where the money is right now but we are using frontends built on backends that have not been updated. In 10 years licensing software to run the backend will be replaced by Software as a Service offerings. This is separate from Open Source but they go together nicely.
Geert thought libraries would move from very rigid, completely closed systems to basic, simple software you can add on to and build from.
Cliff refused to speculate about future of library systems. More significatnt is future of authoring and communication in the academy which is becoming more open, collaborative, research data more widely available, scholarly commons will grow.Researc libraries will need very different kinds of systems than the past but there is an enormous historical mass of literature that embodies the past and it will be very problematic to deal with the link between scholarly lit of the past and future scholarly production.
Other issues brought up as of importance, digital humanities, “tools for narrative”, and then how to port across tools, and keepers of the narratives. An incredible amount of original documentary material available on the web but only a few scholars competent to deal with some, via paleography skills, etc. Bandwidth could be a problem. Amazon fedex’s drives for some installs for “cloud computing”. Economic pressures will cause sudden radical changes. Corporate records and histories could just disappear. Difficulties for the future of journalism might be an opportunity for libraries to help newspapers going online. Used to say “customer service” changed to “user experience.”

LLAMA SAAS Open Source meeting
Joe Lucia of Villanova University spoke on Open Source. He sees it as congruent with the social mission of libraries. Referenced Lawrencd Lessig, Yochai Benkler “Wealth of networks” Stephen Weber “The success of Open Source”. Libraries support eh cultural “commons”. Move ideology to the heart of the systems used. We must see libraries as “social”. Open source ia paradigm of a library. He sees the challenges against adoption of open source in libraries:
timid leadership
self destructive addiction to legacy data standards and practices
complex application in very small application space Acq systems especially
lack of practical alternatives to sunk costs of existing systems
allegiance to current work processes
unrealistic expectations that all will work perfectly immediately
Fixed resources, funding, budget
Culture of vendor dependence
lack of technical confidence
lack of broad and deep technical talent pool
dependence on individuals not growing a group
reluctance to hire and compensate technical staff

His ideal end state would be 3 years out 30% of all libs use Open Source and if you aren’t using it you feel you are falling behind.He suggested a need for community owned data infrastructure based on RDA to be used on teh Semantic web with networked data. This would need a rich commercial and nonprofit ecosystem for Open Source. He thought the Kuali type structure Ole was looking at was needed. I talked to him later and he’d like to move his ILS to open source. He mentioned the need to capitalize migration cost. If a model could be developed that did not front load the costs but allowed a library to migrate, end up with reduced support cost and payback on the migration with savings from ongoing costs. Hosted systems need to be an option and folks need to let go of legacy infrastructure and redirect money saved from support costs.
Other speakers talked about the Evergreen implementations of Michigan Library Consortium. They are starting with small libraries where III could not meet their price point. It still is not cheap enough for all small libs. Larger libs may be interested when Acq is available. How to share costs for it is an issue. Grand Rapids Pub Lib data center hosts their servers as they want to promote open source. Software development issue is how to contract to share costs, or wait for someone else to add a feature.
There was a report from the Waldo Consortium moving to Koha with support from LibLime. There are 15 libs. They have invested $600K in development. They had 21 dev projects after Liblime did a “scoping” report for them last year. That has become 50+ projects but some are when they split one project into 2, others are additional. Not all will be done. 7 libs are up, 7 will go up end of July, 1 Jan 2010l
Meadville PA reported on their experience as an early small library Koha user. they paid $35K to have Indexdata’s Zoom incorporated in the front end of Koha. After that he believes the php and Perl skills needed to add to the system are skills taught in any community college.
There was a question about problems with VuFind development and Joe explained the main developer, Andrew Nagy, had left and gone to Serials Solutions and that had caused problems. But they have hired in talent now.

I also attended the Evergreen get together and talked to Bob Molyneau. He said King County public library had been working with them and now would allow them to publicize the fact. They want to fund necessary development to go live in 2011, I believe.

Talking about the settlement, Dan Clancy mentioned that with it the libraries will get copies of anything they own, not just things scanned from their collections. Without the settlement they get only files of things scanned from their collections.
Pat Steele mentioned that CIC sites want to concentrate efforts on what they can do to complement Google work. Hathi is part of that. She said the earlier, go it alone digitization projects were a disaster from her point of view. There was no way the libraries could get the stuff digitized.
Lee talked about negative issues concerning the proposed settlement. For the 4-5 million in copyright out of print titles without a subscription libraries could view 25%, view 100% for a fee. No copy/paste. With a subscription you can copy/paste 4 p. at a time, print 20 p. at a time You can annotate and link to files from Blackboard and other Course Management systems. But no libraries were party to the settlement only Google authors and publishers. Libraries could be sued for up to 5 million. She had concerns
Kevin is a librarian and a lawyer. He was concerned about Google developing targeted advertising based on what you have already read. He has big privacy concerns on the part of the library. He was concerned there would be a disincentive to get things into public domain on the part of Google or the publishers. As a basis legal position he pointed out that in this class action suit, Google and the plaintives all share an economic interest in definind the class. Normally a class action suit is about defending a basic right. Here it is about a basic right being licensed.
Crosby pointed out that the source of copyright goes back to a 1710 UK court decision that was done to reverse monopoly on printing/publishing. So it was to get rid of monopoly. And now we may be going back to give Google a monopoly. We are going from a royalty model to a fee model. Public libraries have never been fee based, so it doesn’t work for that. Public libraries need to be at the table. Privacy needs to be covered. Limit of 1 terminal for a big public like NYPL is ridiculous. Fair use needs to be defined, as he fears fair use will be narrowed. the 20 pg limit for printing is too arbitrary.
Dan Clancy commented that Google thinks there should be orphan title legislation. They want that spelled out. He also mentioned there are 2 research databases specified by the settlement that would be administered by libraries. This would be for full text processing research. Now only Google can do text processing with the settlement others can do it.
Dan said for privacy issues were not an appropriate topic for the settlement, this will be covered by Google and customers including the libraries and he claimed libraries are tough negociators. For photos, Google scans but they won’t make available. Libraries do get.
Pat mentioned that HathiTrust is looking to develop services like the Research Corpus, the text processing. She mentioned space as a service in libraries saying public libs have offered this but academics have not. She sees the books taking up too much space and wants to move them out and offer the space for other campus activities.
Dan Clancy said they estimated more books will be deaccessioned from libraries. He thought libs would concentrate on special collections, smaller subsets that Google would not be interested in. He said Google wants this to be a good consumer product and the only way for that to work is if it is inexpensive. He said they thought most users of Google are not in the university and aren’t going to the public library, and they are the ones whose usage of these materials can increase with availability via Google. He anticipated costs of something like $5.99 and thought you could offer links to libraries who would drop ship a copy to borrow, or you could buy. Libraries could buy access to a copy rather than doing ILL.
There was a concern that by forcing public libraries to a fee based scenario this would deepen the digital divide.
Dan Clancy said thinking about what will happen to new books published in the future is a real challenge. He said in public libraries users want the most recent imprints. What will happen there> This settlement won’t impact what happens there. He called it the Big Question
He also responded to critcism saying Google hoped the settlement would open up public domain and make it easier to clear up questions about who had rights to things and what could be declared as being in public domain.

It was an interesting presentation in a full room with quite a few questions.

http://librarygang.talis.com/2009/02/10/google-books-and-libraries/

The really interesting idea is for libraries to put out their catalogs in an agreed upon format that would make it easy for search engines, including Google to crawl. She says Google would be very interested in using that information to make sure if you found a book in Google book Search you could find out that a library held it.

OCLC provides a layer that does this now, but it is not necessary for OCLC to try to be a monopoly and corner the market by trying to say libraries can’t put out their catalogs in this format. The really valuable information in this case is item level data, what the library holds, not really the bib data. The bib data could just be a number like ISSN to connect cites from one place to another. I emailed Marshall Breeding who was on the call to ask if he had followed up and he said, indeed there is a DLF group that wants to agree on and promote a standard for this. This is a necessary move ahead for libraries, I really don’t think OCLC will be successful it they try to put a finger in the dike and hold back this kind or progress. WE need to keep forging ahead.

I also dispute Karen Calhoun’s explanation that only 2 % of a library’s records are original cataloging so that’s all they ever paid for. Excuse me. Libraries paid OCLC first time use fees that used to be $1.50 apiece. I believe we paid more than that for a 1.3 million recon project we did with OCLC. So we have literally paid OCLC millions of dollars. I would suggest that the purchasing and legal departments of universities would not allow them to sign up for OCLC under the terms of the proposed policy. So, OCLC is saying, pay us millions of dollars but you can only use those records if we say you can. I just don’t think institutions would agree. Not for these MARC records where the bulk of them could be had from LC for free or for a minimal subscription price.

OCLC is not advancing the interests of its members by trying to exert control over the use of MARC records. As I commented at the meeting, it is solidifying into a lump of rock in the stream and I think there should be a concern that the world will flow past and that will be the end of its usefulness. OCLC needs to have a different business plan instead of just depending on cataloging as the cash cow and then using such a lame policy to try to prop it up. You should put some kind of free use license on the records and charge for cataloging, ILL and other services like timely inclusion in Worldcat.org. Don’t try to rent out the records, or worse prevent their use.

I remember having to argue with developers at OCLC when I first started working there after being in a network. They were afraid to let libraries download MARC records into their local systems and tried to resist allowing that to happen (remember that’s how Innovative Interfaces got started? By screen scraping to do that). Imagine what would have happened if OCLC had maintained that stance and refused to let people download to their ILS’s . They would have gone elsewhere. The genie is out of the bottle. OCLC has not magic word to get it back. They need to move on to innovations, not protect records they never produced in the first place. They should find a way to help LC to continue to produce the many, many records we depend on getting from them. They are the ones who do the bulk of the work.

*

Creating and Sustaining Communities Around Shared Library Data. * Karen Calhoun (VP, OCLC WorldCat and Metadata Services) will speak on the environment for library data sharing and the process of revising OCLC’s 21-year-old /Guidelines for the Use and Transfer of OCLC-Derived Records/. Brian Schottlaendar (University Librarian, University of California San Diego) will consider the topic of library data sharing from his perspective as a library leader with a background in collections and technical services. Peter Murray (Assistant Director, New Service Development, OhioLINK) will comment from the perspective of a library membership organization that provides consortial access to a large union catalog, licensed content, dissertations, and digital media. John Mark Ockerbloom (Digital Library Planner & Architect, University of Pennsylvania Libraries) will speak from the perspective of a library practitioner with a keen interest in freely accessible data and content.

Karen Calhoun has been 18 months at oclc. She led a review of their record use policy. She compared it to a community swimming pool and had a picture of her local swimming pool up on the screen. She said OCLC is experiencing a change of business model from the value in creating and controlling data to the value in exchanging and linking data.

They found the norms for membership were voluntary. They looked at other ways of sharing. Open data commons with a public domain license is one. Voluntary sharing requiring attribution, giving URL of source. She complained that bloggers think data should be free but in reality it isn‟t. (She was comparing only to corporate, vendor sites, not library or government sites where it is free, actually, my comment). She complained that they just think “other people‟s data should be free.” She cited her own IFLA presentation on the web.

http://www.slideshare.net/amarintha/calhoun-data-sharing-panel-ifla-aug-2008-presentation

She complained that OCLC was not a villain and there needed to be agreements not acrimony. She said the revised policy is a legal document. She apologized for not getting more review before putting it out. There will be a review board and “OCLC is not choosing the review board.”

(I understand from friends who are former OCLC employees that the director of Notre Dame who heads the review board is an old buddy of Karen Calhoun‟s however)

She showed a chart of average library catalog where only 2% were created by the library itself so seemed to say you couldn‟t say you had bought all your records from OCLC, you only “owned” 2 %. (However that 2 % are the ones you did NOT pay for. You paid per use charges for all the rest, and you paid plenty. I did not make that point in this forum but will later).

John Ockerbloom from UPenn talked about the cost of information restriction. He used the example of Chemistry as a domain where information has been restricted and as a result advances have been slower than in physics and genome research where shared info has made advancement quicker.

He gave the example of subject maps web page that he put up as part of his research. He had used 30K records from Penn‟s db. He asked what if he used all 3 million instead of 30 K? What if he scaled up to use all of worldcat? We don‟t know what would happen because he does not have access to the data. He pointed out that with the Google Book project the more info about the status of copyright we have the more information will be freely available. The Registry is under the publishers control. And independent db of copyright info is maintained by OCLC BUT it is at least as restrictive as Worldcat. How does that help libraries?

He pointed out that when the original record guidelines were written, the technical environment was totally different. He showed the tech specs for a flash drive that could hold all of worldcat. The new proposed policy is much more restrictive of creative use of the data. Furthermore he quoted the clause that says OCLC had sole discretion to judge the case for use and that OCLC could unilaterally make changes in the policy without consulting anyone.

He pointed out a different way to approach it would be to use an open source type of license. He said there is a cost to not making the data available. He suggested OCLC needed to look for other support strategies to fund themselves and look to other services like ILL, Worldcat, etc. for funding. OCLC is not the only source for data. He mentioned Library Thing, biblios.net, RLG which is now gone. He pointed out that the descriptive data is factual and, as such, is not copyrightable. He also made the point that in a cooperative members decide how to proceed, they don‟t get dictated to by something like OCLC, so the membership ought to decide the issue.

Peter Murray from OHIOLINK talked about records as surrogates for the desired objects. Sources include OCLC, LC, vendors, biblios.net (a new free source based on the Open Library Internet Archive set of records, where they want one web page per book and were donated records). He called these Human generated surrogate records. He wanted a different term for cataloger and suggested “descriptologist” (which was later shot down). Metadata is what we know, data is what we seek. The distinction is becoming blurred on the web though.

It is also possible to have Machine generated surrogates, e.g. from search analyses of a big datastore like Google Books you might generate such surrogates.

As far as ownership goes, facts are not copyrightable and the description is factual. Creative is probably subject headings and classification. There may also be computationally created attributes in the record. Then there can be user generated content like tags and reviews. This amounts to one very messy surrogate. Library Thing gives the rights to reproduce and share their data. Users can put in info about books, facts. Might assign a Creative Commons Sharealike license. Biblionet uses that. Internet Archive just says anything there is in public domain. So, it becomes very messy to split out rights of surrogates, very messy for reuse.

He recommends we put data out in open data licenses with no roadblocks to reuse. Where we see these problems are in new discovery laters like Extensible Catalog and Serials Solutions Summons. Subject specific portals. New workflows. He mentioned Ole might push for a

collaborative type of design, where libs collaborate in purchase and tech services. This policy would be a problem in that case. OHIOLINK has an RFP out for a way to improve workflows by collaboration.

He said the elephant in the room is this record use policy. Setting aside the actual text of the proposed policy, the philosophy of restricting or creating uncertainty about the use of records creates problems in doing new things and that cooperatives need to be less risk averse to do new things.

He pointed out that OCLC is trying to mandate that OCLC be in the middle all the time and that is not effective. He said he can see the value of Worldcat driving usage to libraries by providing a middle layer to move from say a Google search down to a local library but we can‟t have OCLC always in the middle. He questioned whether we did want to always have to go to OCLC in the middle for permission. He quoted an OHIOLINK motto things commonly done are best done in common but said, the higher cooperative cannot trump the interest of the members. He said he will continue to blog this issue under Disruptive Library Technology Jester. And encouraged discussion.

http://dltj.org/

Brian Schottlander attempted to summarize.

Karen described worldcat as a valuable asset intellectually and financially and said the falue was changing from creation to linking. She reviewed community norms using the swimming pool analogy. She found a correlation between openness and whether an organization was dependent on the financial stream from data usage. She mentioned many existing data agreements over the years that OCLC has granted and that most had no money involved. The intent was not to change what members do but to limit use outside the membership. He was glad she acknowledged how badly OCLC had handled this issue.

John described missed opportunities if you disallow pursuit of many things. He also distinguished between telling people vs. asking people and pointed out having to request permission is a block to progress. He also pointed out that it was a troubling point in the policy that it could be changed by OCLC at any time. He pointed out the public domain nature of parts of the records and the need for participatory decisionmaking with the members collectively deciding to agree, not being told what they could do.

Peter made the point that the distance between data and metadata has been reduced and messy record make rights issues messy, also the fear that our records will become fractured based on these rights issues. It is bad to jeopardize reuse of the records. Mandating that OCLC always be in the middle may not be where we want OCLC to be.

Brian thought the distinction between community and cooperative might be important saying Karen had talked of OCLC as a cooperative when recounting history but a community for more immediate things. He thinks Cooperative is more formalized. In a cooperative there is a trust norm. In the area of digital preservation if you gave OCLC files to preserve you might want a more formal agreement about responsibility he thought maybe record use was like that.

There were comments from the floor including someone who disputed the idea that they only owned 2 % of their records. And someone from Newbery Library who said they needed to work with a vendor to digitize part of their collections and would need to use the MARC records but those records were greatly enhanced by local special collections data from the records in OCLC. By the way, Karen Calhoun thought if someone wanted to do a web site of an MLK collection they would have to get permission from OCLC for the book cataloging part that had come from OCLC.

I said I was glad John and Peter had been so articulate about the problems, especially the messiness of the records. I said I didn‟t think having to go to OCLC every time you wanted to use records would scale and that people would just dump the subject headings and “non-factual” part and use the records anyhow. I think OCLC is like a rock in a stream and everything is going to just stream past them, and that won‟t be good for OCLC in the long run.

Peter Mmurray pointed out that biblios.net has a db of 30 Million records freely available and that OCLC is at risk unless they change their business model.l

Karen said they are trying to shift their model. They have to sustain the cooperative.

John said google is a good example of openness in a financially successful company.

Karen questions biblionet where did records come from. (She‟ll probably go home and try to sue them but the genie is out of the bottle on that one).

Tim from Texas commented that we can assume the world is sliding towards interoperability but OCLC, contrary to that, is becoming more proprietary. OCLC is trying to compete with ILS vendors and going in the opposite direction from the rest of the world where there is a split between data stores and the services on top of them. He suggested they should make the data free and charge for services like Worldcat Local, xISBN, etc.

Another library questioned record ownership. They are and ENHANCE library and have found 37% of their records are upgraded, so want to redefine original cataloging. Karen said they are doing a review (?) And they are launching an “expert community experiment” by opening ENHANCE to all cataloging members. (I realized afterwards they are probably doing that because biblios.net, see below, is set up Wikipedia style to allow updates to stick).

280867 Top Technology Trends Committee 1/25/2009 Sunday 08:00 am – 10:00 am CROWN Office

Marshall Breeding, Karen Coombs, Roy Tennant, Clifford Lynch, Karen Schneider, Karen Coyel

They also ran a LITA blog, twitter and another channel

Trends they see:

issues ard mangement of technology open vs closed systems and data

coombs new you don‟t have to have programmers

tool creation libraries are back to doing it instead of looking to vendors

using wordpress and druple outside library vendors, those are open source non-library software

flickr commons Lynch noted the LC project using flicker to find out about a photo goes beyond description to include narrative added by people.

He thought the LC report was good and was a new threshold moving description to a larger set of players

Schneider can you go beyond initial lib and community

Tennant need to manage process vufind is forking Australia has a different version from the original one

Coombs likes that, she sees a core with plugin modules

Lynch next 10 yrs will change a lot about name authorities. What to do about names There will be markup of biographic and historical data and it will be in dbs

Coyle info more geographic based

usg wireless signal strength to see where you are will tell you which way to walk to get to the book

ubiquitous use of cell phones in future

oclc has cell app for worldcat in testing

Coombs too many look at zip or ip need to get to id you‟re a student at x institution so have access to xyz. google scholar lets you set yr affiliatns

oclc has somw of data to do this

There was mention of a need for an open street map for libs they walk a go click on gps to get open info google is closed txas st lib has or can gather

geo coding breeding has for US libraries, based on addresses

Tennant semantic web linked data out on web with stable URIs instead of apis into systems. linked data he likes as a term better than semantic web. Thinks term semantic web is

indigestible but linked data is incremental. Ed sommers lcsh as linkd data will be up in 4-6 wks. Scientific communities putting up research data in this way.

ore standard is for data on the web. It allows you to move or reference objects or aggregations of objects as opposed to metadata about objects. It is semantic web friendly but not required.

Coyle linked data question how do you get links to be meaningful or maybe the problem is we dont use html right use it easiest way.

Economic pressures pub libs are experiencing high traffic at the same time as their budgets are cut. Self check and rfid might be economic. Coyle had done a study and found RFID was worth it for public libraries but not good ROI for academics. Some other things not in library systems that could produce economies were mentioned, Adv shipping notice, tracking orders as Amazon does, ecommerce fulfillment more streamlined than library ILL. Ought to be more fulfillment oriented, provide home delivery to faculty.

Universal telecomm a problem banwidth not there for rural libs a policy issue needs help

Gates has big nw prog

lightng round

Coyle open vs closed depnds on bus models closed models of data more expensive

good enough winning over perfect

move frorn dbs to web as data platform

using tech to answer rights questions

Schneider print publication is on life support death of newspapers

awareness of carbon footprint

Lynch

print demise newspapers melting down striking how quickly its happening. historic collections of papers used in research, so if only web how will they be archived etc? libs in medium sized towns and cities may need to take on this.

The way people interact with dispalys is changing. People use multiple displays multiple monitors. Think about the ramifications. Libs are single screen oriented. He cited a post by google larry spars on optical computers. Powerful for scientific image analysis

Greater demands for accountability for cost benefit of technology. luddites arise. recent report says all cost rise in university admin overhead is for technology. will ask how do we know what we get for tech evidence based hos tech improves teaching will be required.

Tennant does not think print books are going away like newspapers.

self publishing increasing

book industry more complicated than newspaper

libs to collect books cant just look at traditional pubs distribution is changing

Tennant tough economic times require good econ decisions how do you decide to go in and cut off technologies when you have less money to spend

Coombs dig preservation nightmare scary complex technical issues. We have to rely on others like journal publishers, google books, content harvestingg

Lynch scholarly journals in best shape not small mags though

Breeding dig pres journal best off more worried about local dig content building a real repository difficult. ex libris has rosettta, canada is buildg. lots of local digitized files in jeopardy

discoverry solutns Serials Solutions Summons system added to the fray. There are a lot of products but not enough libs have deployed. Do‟nt wait till product is perfect

Integration of social computing

in magic has a social knowledge network. librarythg for libs now distributed by bowkerr and amazon has an interest.

Bibliocommons another product.

lib automat industry perceptn report put out by Breeding says smaller libs happier than bigger libs with their ILSs and open source products are in the mid level or satisfaction not top level which he though probably reflected growing pains for them as they are taking off.

LITA

Program Planning Committee Sunday, January 25, 2009 10:30 am – 12:00 pmWorldCat Web and Data Services: The Developer’s Network Luncheon

Roy Tennant of OCLC ran. There were posters for the following offshoots of the program:

wordpress widget

Notre Dame term finder

worldcat search calif state u

zotero uses oclc open url as dflt gateway

compare ev gs android dev challenge

primateli wisc madison open url registry

Notre Dame xisbn send it to me

oclc Don talked about grid services

ezproxy for id management

web services

grid services

search api

id services

data services

institution registry

ezproxy

developers networkk

they put out web services and are surprised what people do with them

search api uses open search and sru. It will bring back holdings too

restful uris

combine w id services

facebk a word press widgets use these

id services xid frbr

xissn can pull up serial history and grouping. There is a title evaluator app on their website.

Sruw is used to sync Australian Natl catalog with Worldcat, also a Dutch union catalog. They will eventually want to replace batch services dataloads with this machine to machine type of load.

worldcat identities view by authors sru search service interface also uses this.

worldcat registry of institutions

need to improve quality

worldcat.org uses to find libs so it is important to have correct links

generates deep links

oclc mobile pilot will use for discovery delivery

first search admin duplicates this info. Next summer new admin platform will combine

wrldcat local a other new stuff uses registry.

there is a service for this for the public schema

ezproxy

oclc startg to integ for their own products

new version for shibboleth suppt (we use for Refworks, testing with Ebscohost)

OCLC had hackathon in NY 40 people will do in europe and west coast

They are also doing a pilot to host ezproxy on oclc servers for 3-5 institutions

updats to search xid were released

bundle data eservices validations will be coming out

id managmt reduce amt of auhenticatn is being worked on

dev netwk has listserv blog and wiki

Holly Egglestn Calif Dig Lib talked about shibboleth development and the InCommon Shib project. We are part of that. We are ending phase one and beginning phase 2.