Meeting Minutes for June 27, 2005, Chicago
CONSER Task Force on Long Term Storage of Publication Pattern Data, Monday, June 27, 2005, OCLC Blue Suite, Hyatt Regency McCormick Place
Attending: John Espley (VTLS), Linda Miller (LC), Diane Hillmann (Cornell), Les Hawkins (LC), Cathy Kellum (OCLC), Robert Boissy (Springer), Mark Needleman (SIRSI), Ruth Haas (Harvard), Amira Aaron (Harvard), Mike Markwith (TDNet), Peter McCracken (Serials Solutions), Elaine Ross (Thompson Gale), Bob Alan (Penn State), Jenny Walker (ExLibris), Charlotte Dewhurst (Elsevier), Ted Fons (III)
1.0 Serials Release Notice (SRN)
Linda began the meeting by giving a bit of history of the group’s recent work, in particular, discussions that began at Boston Midwinter. She discussed the SRN work, and its compatibility with MARC21 Holdings, which holds out much potential for beginning to map out a way forward. The work on the ONIX for Serials “Serial Release Notice “ (SRN) now includes both enumeration and chronology data elements, including captions. Open URL and ERMS applications require this more robust format that includes enumeration. There has also been an intersection with the ONIX coverage group and its been recognized that there is a need to account for embargo periods in both SRN and MARC 21. Comments on the SRN draft should be available soon.
Linda and Diane discussed two possible scenarios shown in handout diagrams. In diagram 1, SRN traffic flows from publishers to aggregators, subscription agents and directly to subscribing libraries. Libraries could receive SRN data about different titles from any one or all of these sources. Data flowing from publishers would need to be encoded in a sharable format that includes both enumeration and chronology.
Diagram 2 shows a traffic flow with PAMS acting as an intermediary between middle agents (aggregators and agents) and libraries for particular titles. It also shows libraries receiving data about a particular title directly from publishers. Many libraries like LC probably will want to work with only one or at least very few data streams. The idea that the PAMS can be the one source of SRN or SOH data for most titles is therefore attractive. Publishers could work with the subscription agents they currently are associated with in transmitting data and/or they can work through the PAMS who would transmit the data to existing library customers.
2.0 Publisher Participation
Bob Boissy spoke about the potential use of this data and how it’s being discussed within his organization. He feels it would be of use in marketing services and making them more visible. Also, it could be good for defensive measures like preventing claims: he calls it “data with a real purpose” or with the potential to decrease overhead costs. He would like to include people who do publisher fulfillment, who would have a direct interest. He volunteered to help spread the word within the industry. There is a need to sell this internally to the publishers. It needs to be seen as a way to highlight those elements of ONIX for Serials that create value. Being able to help resellers sell e-serials is important. Enhancing the claiming process and reducing the amount of time spent on claims a selling point. Bob proposed seminars for publishers to sell involvement with ONIX for Serials, possibly something could be made available at the Frankfurt Book Fair to spread the word among other publishers.
Diane suggested that another way to sell it is through the idea that publishers could receive data from libraries that can be reused. This would be holdings data not bibliographic data and sharing would be performed in a standardized model.
Bob: Obtaining data from libraries is a different/new process. This could be a one time project used as a check/verification of publisher owned data about a title. Or does it involve getting a feed of data that tells you what the runs are?
Discussion ensued about potential for sharing prospective and retrospective information on library holdings through the medium of an openly available Publication History database. Both publishers and libraries need retrospective data, and the potential exists for collaboration in building machine-readable representations for reuse by all. Receiving embargo information is of value for PAMS, sounds do-able. There is interest among the “middle” players, TDNET, Thompson Gale, III in testing the format.
An SRN pilot could allow many different players to determine the type of value that can be derived from sharing the data and how it can be incorporated. It could also allow them to experiment with adding value. This may be looking like a move toward electronic check-in, machine check-in is possible with the data encoded in XML.
3.0 Exploring Pilot Potential
Linda continued with the handout explanation, pointing out an example with the XML version of the SRN, and suggested that interested parties in the room could be involved in a pilot effort. Great interest was expressed by a number of organizations present in participating in a pilot. Linda will forward the names of interested parties to Christie Burke, who heads up the SRN effort. Linda mentioned that LC had suppressed some e-holdings information to users because the data, limited to spans of chronology, did not match their print holdings information and was confusing to users.
Boissy wondered whether some publishers and vendors who invested heavily in EDI might be resistant to this effort, expecting that end services might just as well transform the EDI data. Linda suggested that this is one possibility for the current intermediaries, whose role will certainly change and who may be open to other roles to be able to benefit from this effort as providers of new services to each end of the model. Some publishers using EDI X12 standards will say: we are already sending this data in EDI, we don’t have the time to encode it in a different format. When our ONIX for Books format is fully implemented, maybe then we can consider the move to ONIX for serials. Given this response, a possible scenario is that publishers send the PAMS the data as an EDI transaction, the PAMS tweak and convert and pass it along as SRN. Publisher could easily provide PAMS with EDI transactions to test this so there is no need to wait for publishers to fully adopt ONIX for serials to begin testing the transmission of the data.
Jenny Walker mentioned that for her organization SOH was a higher priority but they were very interested in this effort. Linda suggested that as the pilot begins, the listserv would be a good place for questions and concerns to be aired, and new participants integrated. There are a number of stakeholders, and good ideas to move information more efficiently and add value.
4.0 Related OCLC Projects
Cathy Kellum described her role at OCLC, and their migration work moving to a new system. The LDR data, a brief holdings record, is being moved, and MARC21 Holdings data created from the old LDRs. She described how OCLC was planning to use the LDRs to manage interlibrary lending transactions more efficiently.
She also described a pilot that OCLC was beginning, so that publisher data could be harvested and used to maintain library holdings data more efficiently. At present this is Level 1 holdings with limited coverage (basic summary). The plan is to move from this basic pilot to more complex and usable data exchange. OCLC is working with PAMS to set e-serial holdings for 20 participating libraries. In conjunction with registering the library’s OpenURL resolver with OCLC, the pilot will test new services in Firstsearch and other products. Right now it is strictly at the title level, perhaps later there is a possibility of including deeper holdings data like enumeration and chronology.
Diane suggested that we can look beyond the utilities and ILS systems to consider another way of providing holdings. Look at them as harvestable records through OAI repositories. Create a sufficiently detailed SRN so that publishers would not have to do it on their own. OCLC has the capability of outputting holdings data in harvestable OAI records. This could allow publishers to harvest the data from OCLC or receive it as a feed.
Diane suggested that OCLC might enable the publishers to harvest XML versions of CONSER holdings records, as part of the pilot, obviating the necessity of creating SRN detail from scratch. OCLC is well placed to experiment with this and might even be able to assist publishers in initiating harvests conveying this type of information. A question was asked of Les Hawkins, what the role of CONSER might be? Les suggested that the holdings information might be considered as part of the extended bibliographic information CONSER already provides.
Jenny Walker mentioned that information provided in the pilot mentioned above doesn’t mention the vendor, so isn’t currently actionable. The purpose is to see what kind of services might be provided, so the lack of firm deliverables isn’t too concerning. Linda suggested that by the time of Midwinter some of this work would be far enough along to report.
Peter reported that TDNET hasn’t really looked at ONIX for Serials because OCLC was so “forgiving”—willing to take data in the chronology only formats currently available. They’re trying to figure out how to participate in the OCLC pilot, and would be interested in participating in an SRN pilot as well. To provide the data in a more structured format for an SRN pilot requires setting up a different system from what exists, but there is value in the project and Serials Solutions would view it as a worthwhile effort. Mike Markwith of TDNET also expressed interest in pilot participation.
Diane mentioned that there are various issues being brought up as part of this, including additions to MARC21 to support:
- Release date
- Article level, and sub article level (including versions) (DOI would be persistent)
- Moving wall and embargos
- Other current awareness data (for other services)
5.0 Pre-Publication Articles
Bob Boissy asked about prepublication articles, and how these might be part of the pilot. Discussion ensued about how such information might be used for services and how rational data creation and exchange is critical for these new services. Opportunities for bringing discussions of data transmission possibilities to publishers and intermediaries were mentioned, including participation by entities like CrossRef. Bob raised the question about pre-publication articles issued in un-paginated HTML, and uniquely cited by DOI before the final with enumeration and chronology is known. Can the SRN handle this situation? Can the MFHD? This situation points out the need to be able to work at the article level and below. Prepublication article availability is a service for authors that publishers provide and, thus, perhaps a selling point within a publisher’s organization if SRN can handle it. Since the DOI is stable, this probably is the type of thing that could be handled by SRN. Helen Gbala pointed to SRN structures that could deal with data about pre-publication articles. Others in the group wondered about the need for transmitting this information; it would be a separate transaction from an issue type of announcement. It does not seem to fit in with the kind of publication history question in the way we are talking about describing the issues that have been published for a serial. How would this article level release transaction be used?
From the publisher and author [and probably researcher] point of view, there is very intense demand for pre-publication articles but there is no systematic way to find them, they are stumbled upon. Perhaps there is an opportunity for a new type of service or a use case can be developed for this type of transaction.
The group brainstormed on other ideas of new types of services or related projects:
For example, create records on the fly from dispatch data. Or teach publishers how to harvest data from digital library repositories. Techniques in digital libraries for sharing and harvesting data are widespread and well understood, they could be used in a business environment. This is something that could be the topic of future conferences. The Frankfurt Book Fair is a forum that attracts publishers, the London Online attracts both publishers and ILS vendors. NASIG is a forum that attracts publishers, many of the “middle” agents, and PAMS, and ILS vendors.
The listserv HLDNGS-L hosted by [email protected] will be used to hash out further steps that might need consideration while a pilot is planned.
Linda said that the SRN pilot would be headed by Christie Burke of Ebsco ([email protected]) The next JWP call is July 12 and it is likely that some of the planning would be done over those calls.
Our thanks to Diane Hillmann and Les Hawkins for contributing the minutes to this meeting.