This is a summary of the discussion of the LITA Library Consortia / Automated Systems Interest Group meeting on Monday morning of the ALA Annual Convention in New Orleans. The meeting consisted of a managed discussion of the use of Electronic Resource Management (ERM) systems in consortial environments. In some cases, comments from the two primary speakers and discussion among the commingled and unattributed. Inaccuracies and comments taken out of context are the responsibility of the author of this posting, and corrections or embellishments are welcome in the form of comments to this post or as private e-mail messages.

The first speaker was from Yale (unfortunately, I arrived late and didn’t get her name — a helpful hint in a comment to this posting would be appreciated). Yale purchased and “implemented” (actually getting any ERM system up and running is not a trivial task, so it is hard to say when “implemented” is) Verde about 14 months ago when it was in alpha development. Ex Libris delivered a production version to Yale in September 2005 and from them until two weeks ago Yale has been working through configuration and getting an the electronic journal coverage load into the ERM system to know what they have(“took quite a while to get there”). Next they will be migrating information about which databases and e-books they have, followed by data entry for subscription, license, and administrative information for all records in the system. They will probably set up these data entry projects with three different groups, one for each metadata type, operating concurrently.

Yale’s ILS is Endeavor Voyager. They looked at Endeavor’s Meridian and Innovative’s ERMS in addition to Verde. ERMS, although it was in production, was less attractive because it would add third vendor in their automation mix. A major part of their selection process came down to which of their existing systems needed to be more closely integrated with the ERM, and they decided it was more important to be more closely aligned with their existing SFX and Metalib implementations that with the ILS. Yale’s systems department thinks they will be able to pull read-only acquisitions information from the Voyager system via a custom ‘report’ function using the purchase order number as a key. Yale is not putting print serial information into Verde. (Verde has the capability to create a “print record” for a “work” (e.g. bibliographic record).) They have an automated lookup from SFX to their ILS that the user can use to find information about print subscriptions.

The second speaker was Diane Carol from the Oregon Health Sciences University (OHSU). They have an Innovative Interfaces ILS and because of existing consortial arrangements, had no choice but to use ERMS from Innovative. Prior to implementing an ERM system, they used things like the 856 fields in the ILS to point to electronic journals plus an external SQL/ColdFusion database for public users and spreadsheets for administrative information, license information, and statistics. At one point they were trying to maintain holdings in five or six systems (above plus Ovid, PubMed, etc.) In 2003 they decided to integrate the information about electronic products into one place. This single source of data would be used to upload to other systems and OpenURL resolves, allow wider access to the license data, and centralize the collection development decisions.

Information from their old coverage database (ISSN, URLs, and Holdings) was loaded and attached to a resource records. ISSN connected it to the existing bibliographic record in the ILS. Also has a license record that is also attached to the resource record. Doing things this way created “skeletal records” for resource records — the minimal of what is needed to start loading the system. Because OHSU is part of a consortial union catalog, doing this added a lot of brief records in the union catalog; they are now looking at the process needed to clean this up. OHSU’s OpenURL resolver is Innovative’s WebBridge, and coverage data is automatically updated.

Helpful Hints

Yale started by walking through the DLF ERMI data dictionary to determine what fields they wanted to use what values to put in those fields (e.g. establishing standard terminology), but found it very different when they started the data load and thinking about how they would use it. For instance, does one need to record in the ERM the state of where the license will be adjudicated? If it isn’t part of reporting needs or a public display, then don’t bother. (One can always look it up later out of the paper files.) In other words, don’t create a “high bar” for the initial implementation by thinking that you will fill every field of every record.

If you already have a link resolver and/or a metasearch engine from a vendor other than that of your ILS, decide whether you need tighter integration with the link resolver/metasearch engine or the ILS and have that weight your selection criteria. For instance, Yale reports that Verde can eliminate some duplicative data entry — information entered into Verde can be pushed into SFX. SFX can also be configured to look up permissions, license rights, and technical information in Verde (an “i” information button in SFX). However, from the same vendor, Metalib is not talking to Verde in any way; it is a concern of Yale for when they start to record database information into Verde. Yale is seeking from Ex Libris a best practice for data entry now to be prepared for the day when the integration is done. The general observation of the group was that institutions tend to go with their link resolver product vendor versus their ILS vendor.

To pay attention

Implementation is done for Verde 2.0, documentation done on July 1, early adopters are using the software, public demonstration sites are ready, general release date announcement coming soon. Verde 2.0 has some new features for consortia. Although it not available yet, it is still useful to have their 1.1 implementation to start addressing how the consortium will deal with the definition of fields and how they will be used across the consortium.

What follows is a summary and commentary on the LITA Top Technology Trends meeting at ALA annual conference in New Orleans on 25-Jun-2006. What I’ve tried to do is collate comments from the panel members and add my own commentary (marked off as such from the rest of the summary) where I thought I had something useful to add. It is my hope that this summary is a faithful representation of the statements made by the participants in the panel. If not, please let me know privately or in the comment area here and I will make the appropriate corrections on the body of the blog post.

Please note that this is not intended to be a complete summary of the comments of the panelists; in some cases I forgot to write things done, in other cases what was said didn’t fit neatly into this collated set of topics. For a more complete accounting of the topics, please see Karen Coombs’ and Michelle Boule LITA Top Technology Trends postings.

Evolution and Interim Solutions

As a profession, we need establish a collective mindset that “everything we do is an interim solution.” When our perspective is that of managing interim solutions, we begin to describe our activities with the language and context of interim solutions — that this is work not completed. For instance, faceted browsing is not the solution. Its adoption is part of an iterative process. (Tom Wilson) There is lots of experimentation in the arena of “next generation” OPACs and interfaces — navigation, scope of OPAC content, consolidation of purchased and subscribed content — and none of them are “good enough” to be long-term solutions. (Marshall Breeding) And if our gaze rests solely on the OPAC, we are in trouble. The catalog is but one source of information about library content yet it receives the lion’s share of attention and effort. (Roy Tennant)

Approaching change with the mindset of managing interim solutions will encourage flexibility and more experimentation. I agree with these statements, and I think we need to be prepared to move a little more nimbly in the coming years. Perhaps not at “internet speed” — we have firm roots in sound practices — but certainly no longer at, say, “committee speed.”

Research in information retrieval is now being explored by those that are not in our profession. We are no longer the landlords of the information space that we were before (but perhaps we can reclaim some of it through a lease-back arrangement). (Andrew Pace)

My recent trip to JCDL brought this home. The “joint” of JCDL is, by the way, the Association of Computing Machinery (ACM) and IEEE’s Computer Section — nary a “library organization” in sight for this “digital library” conference. There is top-notch information retrieval experiments and practices being explored here…stuff that we could apply or consider applying to our own systems.

There has been lots of consolidation in the business side of the library automation industry, but it is still fragmented and more consolidation will be coming. The future will have fewer companies and probably a fewer number of hopefully better products. The large automation companies are outsourcing development and integration of some modules, particularly for ERM functions (to companies such as Serials Solution and TDnet). (Marshal Breeding)

The has also been the rise of “managed” open source. Some open source has an audience that is wide enough to be community-maintained (the Apache web server, for instance). For applications of a more limited interest, companies are making it their buisiness to provide support for open source software (IndexData and others). (Karen Schneider) One example in particular, the field of Institutional Repositories was initially open source, but this capability is now being marketed as a complimentary part of an ILS. This puts institutional repository capabilities into the hands of more institutions. (Clifford Lynch)

Focus on the Service Aspects

The results of the mass-digitization efforts will change library operations. What is the role of library if everyone has content on their gizmo? Our role must be to provide services on that content. (Eric Morgan)

Faceted Browsing

NCSU’s Endeca-enabled catalog is part of a long-term strategy for improving access to the items in the catalog; is is not the end. (Andrew Pace) Any decent search engine in 2006 will have this capability. Faceted navigation does a good job at marrying the search and browse modalities. (Karen Schneider)

Findability as a Service

The recent focus on “findability” is very healthy and the dissatisfaction of the library catalog is part of a reorientation to better serve the user. (Karen Schneider) Software for faceted browsing and personalization has reached a commodity status. Automatic classification and subject assignment and natural language processing are the first part of the last mile. (Andrew Pace)

Along with this has to be the realization that the OPAC is not the center of the library universe; other services are of equal importance to the users. Users with full-text expectations are coming to our metadata universe. And for them, tutorials, screen captures, and desktop movies are not going to cut it. (Karen Schneider)

Karen has a new twist on the “librarians like to search, users like to find” axiom. Our users new expect content — not pointers to the content — to be at the end of their finding process. And increasingly they are familiar with finding modalities coming out of the web-as-a-whole and will not sit through a bibliographic instruction session or watch a “screen cast” (movie of desktop capture) to learn how to use a new service. Think Jacob Nielson here — if everyone is doing it a certain way, you probably should, too, regardless of the fact that you mike know a better way to do it.

At the same time we see the breakdown of barriers to publication — with blogging and wiki software, anyone can be a publisher. Along with the benefit of the capability for everyone to publish, we have the detriment of everyone being a publisher. Should the library offer a filtering and selection service for this content? (Roy Tennant)

I can’t remember the exact context of Roy’s mention of this, but it seems that we should be offering this kind of sorting and filtering service of the “unwashed” blog and wiki content so it is inter-filed with selected and vetted content from our commercially-produced collections. That would be a service to our users, I think.

Actionability as a Service

Eric was talking fast, and so not all of the actions he mentioned are listed above. I can’t remember if one of the actions he listed as “excerpt”. Technology has made it easy to excerpt and recombine pieces of content into a new work and the users have taken advantage of this capability — they call it a “mash-up.” As we put content online we need to bring along the enabling technologies that allow it to be excerpted — when appropriate — and track back the provenance of that excerpted content.

Publishing Platform as a Service

Web pages created in the form of blogs and wikis are becoming the norm rather than the exception. How do we think about of these sorts of things? Content is married to the software and underlying database, and it will be difficult to migrate these things forward. (Eric Morgan) The rise of community sistes (collectively, systems like Flickr, MySpace, GMail, etc.) increases the confusion between services for sharing versus services for preservation. “Over the next few months this lesson will be driven home.” (Clifford Lynch)

This is a concern of mine as well, particularly with the conflicting value systems of the entities in question. As corporate bodies, accountable to venture capital firms or shareholders, will sustain a service as long as the business model is profitable. When it is no longer profitable, what happens to the content on those systems? These corporate bodies also seem to be relying — again — on revenue from advertising to sustain their activities. Despite the success of Google in reviving this method of moving money around cyberspace, do we really think that advertising-supported sites will continue indefinitely?

And everyone is publishing, with resulting decreases in this think we call “privacy.” Teenagers are now being councled that what they put in Facebook will follow them for the rest of their life. (Karen Schneider)

If libraries do service these communities, what is our responsibility to inform users of the risk to their privacy and/or take proactive steps to protect their privacy. This question goes beyond, of course, statutory requirements in the United States and other countries regarding the solicitation and position of information about minors.

Granting agencies and university administrators coming to understand the importance of long-term data management and curation. The National Science Foundation’s Office of Cyberinfrastructure will be putting out guidelines on this soon, and it will be something to watch for. (Clifford Lynch)

Network Services

This is a report of the presentations from the LITA Standards Interest Group at the ALA Annual Conference, 24-Jul-2006, in New Orleans. Pat Stevens, interim director of NISO, moderated the panel discussion.

ISSN Regina Reynolds, Library of Congress (U.S. ISSN Center)

Structure

The ISSN International Center is located in Paris. It assigns the prefixes to ISSN centers and holds a master copy of descriptive metadata — the “Key Title” plus other metadata elements in MARC format — for every assigned ISSN. It also provides documentation, a manual (about 80-100 pages in length) and support for new centers coming on board.

Activities for ISSN matters comes from a Governing board elected by the membership. The directors of ISSN centers also meeting annuallyto resolve operational issues. An “ISSN Users Group” has recently been formed as well. The ISSN standard itself is undergoing revision now.

Funding

For the U.S. this translates to about $120K/year dues to the International Centre, plus the cost of staff salaries and benefits, office space, and operational expenses. The Library of Congress pays this out of its budget.

Statistics

The U.S. assigns about 6,000 ISSNs per year, which is considered a low number because publishers should be assigning different ISSNs to different media types.

Challenges

The assignment center receives a lot of “Vanity” ISSN requests: personal newsletters, publications of only local interest (“the town gardening club”), and those seeking a “free” standard number (since there is a fee for receiving and ISBN assignment). The rules typically applied to requests for ISSNs are based on whether the serial is “in the chain of trade” — will it be cited elsewhere or included in indexing and abstracting services or in OpenURL resolvers. Recently the blogging community have been seeking ISSNs, but this has been ruled as not an appropriate use of the ISSN standard.

Future ISSN Network Directions

One question from the audience was about the possibility of running out of numbers. Almost 10,000,00 individual numbers are available for assignment, of which only about 2,000,000 have been used. The pressure of running out of numbers may cause some structural changes to be adopted. One such change that could be considered is the addition of a suffix for media type — a journal in print and the electronic manifestation of that journal would have the same ISSN (which is different than the standard practice now) and be distinguishable from a suffix added to the base ISSN.

Brian Green, International ISBN Agency

Brief History

It has universally adopted as the key identifier for books. The standard was last rewritten in 1992 and last revised in May 2005. The 13-digit ISBN comes into being on the 1st of January next year.

By definition in the standard, ISBNs are only for books. It is a “manifestation” or supply-chain identifier. Coverage includes digital monographic publications on physical carriers (CDs) or online. A separate identifier required for each electronic version separately traded. ISBNs can also be allocated to parts of books traded separately (e.g. chapters).

The migration to 13-digit ISBNs to be encompassed in the EAN-13 standard was an interesting problem. In the EAN-13 barcode system, prefixes are assigned based on country (e.g. “Germany”), not on products (e.g. “books”). So the question was how to incorporate internationally established yet product-specific ISBN into the country-specific EAN-13 system. The answer? Create a new country, “Bookland”, and give it a country prefix 978 (with 979 in reserve). GS1 (formerly EAN Int.) and ISBN now discussing the incorporation of ISBN into RFID tags as part of EPC (Electronic Product Code) standard.

Management and Governance

In 2005, the International ISBN Agency Ltd was created, and it will take over management and governance of ISBN in 2006. The International ISBN Agency is a not-for-profit, limited by guarantee organization with the guarantors/members are the national agency (160 in all). Each national agency has one vote in the governing body with a governing board elected by the members. ISO is keen on this change because it prefers to contract with an organization to be the maintenance agency of a standard; this is now possible.

Structure and Process

The move to 13-digit ISBN means that all 160 agencies have to understand the change and tweak their systems. (Each country has their own system and software, but standard management tools are now under development.)Most agencies have worked with libraries and the book trade in their countries to facilitate change.

Unlike the ISSN international agency, there is no central repository of metadata about assigned ISBNs. ISO 2108 states that ‘publishers “shall” supply local agency or its nominee with basic information about the publication to which ISBN is assigned’ but in reality this is unenforcable once the prefix has been given to the publisher. Publishers, of course, want their items to be bought, so they generally will participate in a local books-in-print effort. The International ISBN Agency publishes a list of identifier prefixes assigned to national agencies (the Publishers International ISBN Directory or PIID). National agencies compile or collaborate in producing a local books in print (for instance, Bowker in the US, Nielsen BookData in UK). In many cases, the compilation is aligned with a national bibliography effort.

Funding

Chuck Koscher, Technology Director, CrossRef

Service Infrastructure provided by CrossRef and DOI

• Corporation for National Research Initiatives (CNRI): Develops and maintains the Handle system. It is a technology partner to the International DOI Foundation (IDF) and an advocate for broad technical solutions.
• International DOI Foundation’s Digital Object Identifier (DOI): Develops and maintains the DOI standard and is an advocate for DOI-related technology.
• CrossRef: Operates a metadata look-up service for the registration of metadata about a DOI identifier, servicing queries to discover the appropriate DOI identifier, and redirection of resolver requests to provide a stable space for these persistent identifiers as the underlying objects move around. CrossRef also sustains a community of users (publishers, libraries, aggregators, secondaries) making use of DOIs as well as monitoring and maintaining the integrity of the resolver service (quality of metadata and links to objects).

Structure and Process

• A resolver address (http: //dx.doi.org/) which itself is not a formal part of the DOI.
• The DOI Prefix (10.1016) assigned to publishers by the DOI maintenance organization.
• The DOI Suffix (S0040-4039(01)80789-9) created by publishers.

DOIs are a special subset of the CNRI Handles; any CNRI Handle that beings with “10.” is a DOI prefix. A DOI and a CNRI Handle are technically the same thing sharing the same resolving infrastructure. One can resolve a CNRI handle against a DOI resolver (for now, may be blocked in the future) and one can resolve a DOI against a CNRI handle resolver.

To assign a DOI, a publisher sends the article metadata to CrossRef with the assigned DOI prefix and publisher-defined suffix. Systems can query the CrossRef database using citation metadata to determine if a DOI exists for that citation. If so, a user interface can present the referring article as an active link to a DOI resolver. When the user selects the link, the browser contacts a DOI resolver and receives in return a URL to the reference document.

Finances

The text was modified to remove a link to http://www.ala.org/ala/lita/litamembership/litaigs/igstandards/standards.htm on June 9th, 2011.

Attending: M. Anderson, U of Iowa; A. Laas, LexisNexis; Y. Han, U of Arizona; P. Howell, Western Michigan University; Y. Kaganovia, Princeton U; P. Murray, OhioLINK; K. Thompson, Smithsonian Libraries

Participants talked about their interest in JPEG2000 and their institution’s use of the standard: Western Michgan University is digitizing manuscripts and other special collections materials and using JPEG2000 for access; LexisNexis is using JPEG2000 in the maps portion of the U.S. Serials Set digitization program; the Smithsonian Libraries has started converting archival TIFFs to JPEG2000 and is considering use of the standard in the Biodiversity Heritage Library project (including a capability to cross-link taxonomic names in digitized text to oneline databases).

There was discussion of the state of JPEG2000 in library systems. Companies such as Luna (“Insight”), DiMeMa (“ContentDM”), Ex Libris (“Digitool”) now use JPEG2000 in their back-end systems and use some sort of server-side transformation to deliver an image to the user. Peter spoke of the Google Summer of Code project to build a JPIP applet in Java that could be embedded in web pages to do the panning and zooming on the client side rather than the server side.

Ideas for Programming