ERC ST 061006Minutes of the 10/6/06 CMDC-ERC Joint Meeting
Portico Presentation (Ken Difore) Recorded by Ken Watson, Eastern Oregon University
Portico:
Currently there are 19 publishers, with 3900 journal titles, signed on with Portico to provide content. There are currently 130 libraries who have signed on to the project.
Portico’s approach is to provide just the text image of the publication, not the “look and feel” of the publisher added features. Working from the SGML source files the focus is on long term preservation not immediate content access. Portico is committed to migration of content as technology changes, for example jpeg -> jpeg2000.
Access to supporting libraries is dependent on certain trigger events. Trigger events include:
Until a trigger event occurs select librarians at participating institutions are granted pass worded access for audit and verification purposes to check on the files and on the status of the database.
Support for Portico comes from publishers who pay from $250 to $75,000 in contributions. Libraries pay from $1500 - $24,000 based on their library materials expenditure LME
The role of the publishers is to articulate their archival strategies. Participate in at least one archival arrangement. Monitor the digital preservation efforts of others. Stay a breast of the emerging and changing legal requirements.
The benefits to publishers are Relieves publishers of the obligation to convert source files to archival formats and conduct future content or format migrations as technology changes. Creates an opportunity to reduce or potentially eliminate current internal archiving activities and related costs. Portico provides a means to meet library demands for a trusted, reliable third-party archive. It also provides a means to meet library demands for perpetual access, where appropriate, without incurring an ongoing operational obligation that lacks an associated revenue source.
The following are the benefits to libraries of joining Portico:
· It is a practical means to act upon their traditional preservation mandate in support of the parent institution's educational mission. · Portico is an opportunity to accelerate the transition to preferred electronic resources without sacrificing future accessibility. · Libraries have reduced system-wide preservation costs. Because of the collaborative effort, this eliminates the need for each library to locally receive and archive electronic journal literature. There is an opportunity to rely more heavily upon electronic resources and realize savings through reduced costs associated with print subscriptions, handling, and storage. · Finally Portico is a means to secure perpetual access if a participating publisher chooses to designate Portico as a provider of post-cancellation access.
LOCKSS – Claudia Weston (PSU)
Claudia recommends that people wanting more background read “Social Model for Archiving Digital Serials” by Michael Seadle
See Claudia Weston’s Powerpoint presentation for details on LOCKSS: http://www.orbiscascade.org/staffhome/
Question and Answer Session
Q – Publishers agree to post cancellation viewing rights if they go out of business? A – Scenario I - Publisher gone the archive opens Scenario II - Publisher allows access to a subset of their archive
Q- Does Portico work with consortia? A- LOCLSS Alliance works with consortia on pricing. JISC in the UK has joined. A- Portico asks for individual participation first and will give an additional 5% discount to consortial members.
Q- LOCKSS notifies of new titles regardless of subscription? A-Institutions only get the content you have paid for.
Q-How are migrations to other formats handled jpeg to jpeg2000 for example? A-Portico has a review team that continually reviews new formats. Portico is concerned about backward compatibility at ingestion. A-LOCKSS does migration on output based on “Typed Object Model” TOM idea of conversion.
Q- How does Portico view lots of copies? A- Portico replicates its archive online around the country, thus keeping its content stable.
Q- How to offer complete content when some authors may want to withdraw content? A-LOCKSS publishers guarantee access to all of their published content.
Recorded by Carrie Fry
Panel on e-journals
Participants:
Donna Packer Head of Technical Services Coordinator of Collection Development Western Washington University Donna described a medium-sized academic library’s experiences with e-journal packages. She spoke of the difficulties working with this complex and fragmented environment including the need for subscription agents and their expertise in helping to facilitate the completion of these e-journal package deals. Electronic invoicing was noted as being especially important for subscription agents to handle. A criticism Donna had of subscription agents is the difficulty in contacting them which could be very time-consuming.
Additionally, Donna described the difficulty in navigating publisher web sites that are overly complex and filled with too much information. All librarians really want to know is the price of the subscription but at times obtaining this information can be very difficult with so many pricing structures. Donna felt the most effective way to get a price for her institution is to speak directly with someone.
Allocation of funds adds an additional layer of complexity. Currently, at Donna’s institution funds are not bundled together even though the package is. Instead, funds are allocated by discipline.
In ending, Donna commented that she doesn’t see the complexity of the e-journal environment changing any time soon and hopes that more efficient process are developed in the future.
Andrea Cernichiari Library Business Development Managerm Cambridge University Press As representing a publisher, Andrea spoke about what she terms “the missing partner” in the world of e-journals, namely the societies. Many journals that libraries want to subscribe to electronically are published by society presses. These societies are not very understanding of libraries needs in the electronic realm. Societies are also quite leery of the big e-journal packages and many times either don’t want their publications to be a part of such a package or want to impose unrealistic conditions on those subscribing institutions. One of the main reasons that Andrea cited as a reason for this disconnect between societies and libraries is that societies work under a predominantly editorial driven relationship with publishers and are removed libraries.
Andrea sees the role of the subscription agent as complex and necessary. Agents must manage across publishers, dealing with multiple choices in access and pricing. Agents notify libraries of subscription changes, follow up on issues, help with e-claiming, and much more. From Andrea’s perspective publishers do not want to take this part of the process on. Moving journals to an online environment is very costly, but the benefits outweigh the work. The subscription agent ensures access to paid for content, closing the circle between the publisher and library.
Tina Feick Vice-President, Customer Relations Swets Information Systems Tina spoke of the concept of subscriptions to serials being the same regardless of format. Fundamentally it is the same process but the format adds a layer of complex that is challenging for all parties. Tina stated that 51% of what libraries pay for serials is in electronic format. Some of the challenges agents face as a result is increased staff to track e-information, added fields to record e-data elements, increased costs, negotiations, establishment of an e-contact person, management reports, license management, access assistance, claiming, EDI line by line invoicing for e-packages and more. In addition to these added services there is also the beginning of what Tina termed E-Services such as A-Z journal title programs, open URL resolvers, mini ERM systems, license banks, federated searching, consortial services, usage statistics, and more.
Because of this long list of services added to their plates, subscription agents are searching for ways to be more efficient. One of these ways is standards. By being at the table in the development of standards, there is an increase in interoperability. Initiatives such as SUSHI, the License Expression Work Group, ICEDIS, and JSCEI are such examples.
Chrysanne Lowe VP Global Customer Marketing Elsevier In describing the evolution and history of electronic licensing, Chrysanne talked about the dearth of such things as licensing standards, strong consortia input, and knowledge of even how to sell e-journals on the part of subscription agents who were still heavily entrenched in the print model. As a result, Elsevier was forced to work more and more directly with customers. Echoing what Tina talked of earlier, the absence of standards when dealing with e-journals and e-journal packages has created confusion and complexities. As a consequence, this lack of standardization brings more options but no more instant estimates across publishers. The more complexities in licensing, the more customers wanted to deal directly with Elsevier, cutting out the third party of subscription agents. And on the flip side, the more Elsevier works directly with customers, the better they feel they understand libraries’ needs. This in turn shapes the company’s business models and licensing terms. Because Elsevier is taking on more and more of the role traditionally filled by a subscription agent, Chrysanne showed some statistics of improved customer service on the part of Elsevier as a result of making innovations in the ways they work directly with customers. She did stress that Elsevier is still working with agents, just in different ways.
Faye Chadwell Head, Collection Development & Acquisitions University of Oregon Faye talked of some significant changes in collection development as a result of moving to the electronic realm. Libraries are moving more towards aggregation. In managing these e-collections, we need other skills like math, accounting and statistical besides the traditional subject discipline skills. She posed the question is collection development a matter of art or economics? This question is rooted, in part, in the concern that by looking more closely at cost per use and the bottom line may effect our collections. Faye used the humanities as an example of an area whose collection could suffer as a result of looking at collection development from a more economic viewpoint. Additionally, how has emphasizing the bottom line hampered the relationships between the librarians and discipline faculty?
Another aspect Faye brought to the audience’s attention was digital archives. These archives are coming closer to center stage so how are we going to deal with them? Do we outsource or build the archive in-house? And how are we to pay for this?
Acquisitions issues have become more difficult with the move to electronic. Documenting for ourselves and our users and staying on top of it is a real challenge. It’s all about maintenance. As related to the question echoed above as to whether libraries use serials agents or work directly with publishers, she believes it is a case by case basis, depending on many factors. It goes back to showing accountability in our support of certain disciplines. Importantly, we can sit and talk about all these issues and new innovative processes, but ultimately managers have to go back to their staff and implement these changes. And lots of change is disruptive. A manager’s job is to help staff deal with this change. The need to move outside of our silos and deal with our budget offices and other departments adds another level of complexity with implementing innovation and change.
Moderator: Diane Carroll Collection Development Librarian Oregon Health & Science University
III Statistics
Linda DiBiase, Collection Development Librarian at UW, summarized the recent Zoomerang survey sent by the CDMC Steering Committee to CDMC members. The goal of the survey was to gather information about member library local use of Innovative statistics. The results will be added to the Orbis Cascade staffhome page.
There is wide variation in local use of Innovative statistics. The most frequently used statistics were Circulation activity by title and Circulation activity by call # (SCAT) table. 5 libraries look at Circulation activity by title annually, and 7 never. 10 libraries look at Circulation activity by SCAT table annually. Many libraries utilize other types of circulation reports.
16 members used the Collection Development Report through Web Management Reports, although many (10) never do, and 2 reported not knowing what it is. Diane herself uses this, but isn’t enamored of it. Libraries also used “Age of Collection” reports, “Collection by Material Type,” and other reports.
10 respondents use INN-Reach circulation transaction reports in collection development and management decision-making. 18 libraries do not utilize INN-Reach borrowing statistics. Several use the paper pull slips.
The comments showed that people would like way to integrate separate reports (patron type, call number, etc.) into a list of titles with call numbers, ISBN, OCLC#, borrowed from other libraries in a spreadsheet format. Several other commented that would like the ability to integrate multiple fields and user-defined time periods.
[Diane mentioned that the Steering Committee is also grappling with the question of what is the optimum number of copies of a title in the system.]
Diane introduced Nancy Nathanson, Orbis Cascade Alliance Resource Sharing Program Manager, who was instrumental in the creation of the UO III catalog, the Summit union catalog, the Orbis-Cascade courier system, and refining user-initiated searches and requests.
Nancy said that SCAT (Statistical Classification Analysis Table) data is normally based on LC classification. Most libraries also have some other types of classification (Dewey, SuDoc, etc.). SCAT data can be used to record volume of activity, and can be helpful in budget requests for supplies and staff, in reports to administrators, and to compute lending and borrowing ratios. Catalog statistics help understand the relative strengths/shortcomings of members, and when making purchasing and weeding decisions.
Collecting data from INN-Reach is difficult. It is important to ask, “How will the data be used?” Is it necessary to collect all data for a time period, or is sample data adequate? Can the data be aggregated or is necessary to have individual data (item level, library level, member institution level, server level, etc.)? Do you need regular, periodical, or occasional data? You may need to consistently collect data for the same criteria over a period of time. Is the data scalable as consortial members are added? It is very time consuming to use Summit for data; it takes three days to collect data for a report now, due to the size of the Summit database.
It is tricky to provide individual member data, which is based on servers. UW has 2 servers and the OSU server is also the server for COCC. St. Martin’s data is included in the Evergreen server data.
All lot of data available on the Alliance Staffhome page (www.orbiscascade.org/staffhome/), in the Union Catalog & Borrowing section. The statistics are derived by server. You can look at the number of:
In the initial stages of the union catalog, there were 82% unique records. This percentage then went down to 65%, and now has increased slightly to 67%, probably due to the addition of records for electronic resources, and the addition of large sets of records.
Summit also provides information about number of searches and the use of the limiting feature. Summit search information is limited to searches that originate in Summit, not pass-through searches.
It is interesting to note that LC searching is decreasing, accounting for only 9% of patron searches. Author searching is slightly down to 18%. Keyword searches have maintained an even level, and are now 23% of patron searches. Title searches are the most frequent search, and have steadily increased since 2001 to 38% of all searches today. (Title indicators do matter, a note to catalogers.)
There are now new capabilities for gathering statistics. Millennium Admin is designed for use on the local catalog level, not INN-Reach.
Orbis Cascade Alliance is looking at a “Data Harvesting Initiative.” Yesterday, Council authorized a charge for the catalog group to, “Explore Summit’s potential as a source of data (using bibliographic, item, and order, circulation, etc.) for the purpose of developing improved services and informing decisions made by members libraries and the consortium.”
From Summit, we can also receive circulation information, site-to-site and system wide, on requests and fulfillments, transfers and circulations, who’s borrowing, loan rules, and characteristics of user request. However, it is very laborious, and involves a huge amount of transactions and data.
It is also possible to look at monthly and annual aggregated information on circulation and materials sharing. Nancy encourages member libraries to compare locally derived data with data provided by Summit. If there is more than a 1% deviation, there should be an explanation.
Summit borrowing statistics for 2005-2006 showed that 335,789 of 358,175 requests were filled. 47% of requests from undergrads; 42% of requests were for a title with 5 or more owners.
On the web (www.orbiscascade.org/staffhome/) you can see aggregated statistics of who requested what from who, and patron type (grad, undergrad, etc.).
Nancy asked why collect local statistics if aggregated statistics are available? One reason is ease of use. Local data is easily exportable, tab-delimited, for use in spreadsheets. A caveat is to be aware of the possibility of bogus data due to classification type.
Libraries can access local information about Summit borrowing from their local Web Management Reports. Select INN-Reach at the bottom of the list on the left. You can choose data range, and then look at checkouts from your school to and from other members. This is easily downloaded to Excel and can be used in purchasing decisions, etc. Be aware that the addition of new member libraries can change your rows and columns, and can affect your spreadsheet formulas.
When you look at requests by title and Ptype, you will see your institutions patron types, not Summit’s. You can also look at requests by author and Itype (based on length of loan). Some members look at Itype (length of loan) for information on A-V borrowing. Don’t assume that a short loan is always for a film or audiotape.
Nancy showed an example of downloaded data from SCAT categories for requests by Call # and IType. She then shaded numbers over a certain threshold using Excel’s conditional formatting, which makes it easy to see higher usage. Nancy mentioned that the reports now hold 256 characters in the title field, so title information is more useful. Nancy issued a warning about using INN-Reach cross-tab locally. There is something fishy about the way that cross-tab works with INN-Reach.
Nancy looks forward to discussions about statistics and feels that we can make more use of the statistics that we have. However, we should continue to ask, “What data do we need and why? Is it interesting or really useful?” If the data is needed for collection development decisions or for gaining more resources, then it may warrant the use of time and resources to collect and analyze. Libraries need to determine who will collect data, how it will be collected, and when it will be collected.
A question and comment period followed. When asked about the decline in use of the journal title index, Nancy responded that the journal title index was discontinued. We now have scoping by journal, which searches all serials regardless of frequency of publication.
There was a question about load balancing, which Nancy explained. Orbis Cascade looks at load balancing over several months. There are often short sporadic spikes in requests at member libraries due to assignments, research projects, etc. This is why a longer time period is necessary,
Someone asked about finding data on PUA (pick-up anywhere). PUA borrowing shows up as an INN-Reach statistic of Visiting Patron borrowing. Do these impact statistics? PUA borrowing will be included in local INN-Reach statistics soon. At this time PUA statistics are difficult to pull out and are included in overall Summit borrowing statistics.
Why is Nancy surprised by increase in title searches? Nancy was surprised because she has heard anecdotally that keyword is the way things are going, with Google, etc. Someone suggested that the increase in title searches may be due to the high cost of higher education and that more students are using Summit to borrow course textbooks instead of buying them. Perhaps more people use Summit to look for a specific book, and keywords may be used more on local catalogs. Or, the growing percentage of title searches could be a reflection of librarians using Summit for collection development, or for checking before placing ILL requests.
Does the Alliance participate in Innovative enhancement voting? Nancy replied that the Alliance does participate, although their track record in term of response by Innovative hasn’t been great. Recent Innovative enhancements seem to used as a way to market to new large consortia.
|