The Dilemma of Access Describing Open Access Journals

The Dilemma of Access: Describing Open Access Journals with MARC and [Other] Metadata Schemes Monica Berger, NYC College of Technology Gloria Rohmann, New York University

Open Access Journals • are indicative of broader trends that will affect librarianship • provide an opportunity to consider broader issues relating to the role of libraries and librarians as controllers and mediators of content.

Open Access: Definitions Walt Crawford: “The fundamental principal of open access is that scholarly research should be freely available to anyone who can use it, at no direct cost to the reader. In practice, that currently means two different initiatives: ” http: //cites. boisestate. edu/civ 4 i 2. pdf • Full access/always free journals • Preprint and hybrid journals

Full access/always free OA journals • no charge for electronic access to the published journals (or collections of articles) • work is done on a volunteer basis • a university or other institution provides the small support costs, so that neither author nor reader pays a direct charge. • Example (from Pub. Med): ELECTRONIC JOURNAL OF DIFFERENTIAL EQUATIONS (EJDE) –University of Texas

Preprint and hybrid OA journals • Author-initiated article archiving in institutional or topical archives • Author adheres to metadata standards =OAI-PMH • Deposits to standards-compliant archive – British Project SHERPA ('Securing a Hybrid Environment for Research Preservation and Access') or – Physics, subject-based ar. Xiv at Los Alamos • These OA archives often also include postprints of articles, dissertation/theses and other scholarly webborne content • Type 2 OA journals hybrid free/licensed. • Hybrid example (from Pub. Med): Arthritis Research & Therapy all research articles Open Access; subscription required for other content; free trial available

Highwire A familiar hybrid collection is Highwire. Ejournals can be: • Entirely free, all issues; • Recent issues embargoed (most common); • Free text via trials; • Pay-per-view • [another common scenario, doesn’t apply to Highwire: back content embargoed]

Recap: Preprints and other hybrids • In the traditional journal world, we have microfilm and ejournal analogs. Sometimes, publisher web versions of print articles are abbreviated or different. • Preprints are particularly problematic since they are typically similar but not identical to the published version. Postprints typically have updates as well. • The issue of versions is a great example of why it is important to consider issues of identification and description of OA journals and all ejournals. We’ll repeatedly return to this issue …

OA Opportunities Much of the discussion on open access (OA) journals is scholar-focused and relates to: • New, alternative modes of scholarly communication and production • Economic impact of OA on traditional publishing industry OA is creates many opportunities as well as challenges: Some of the most difficult challenges for librarians relate to: • description and collocation • access • preservation

When considering issues of description and collocation/access/preservation … Role of: • librarians as mediators • libraries as repositories/archives of content • libraries as the focal point or hub of bibliographic research Will OA content find its way into: • OPAC and other bibliographic tools? • bibliographic utilities as union catalogs?

The Chicken and the Egg Description and access are really two sides of the same phenomena and one can’t consider one without the other. I will focus on description and Gloria will focus on access and hopefully we’ll get some discussion going about whether or not we librarians are critical stakeholders in the OA movement.

OA Journal and “our collections” • We and our users understand our collections: – What are “our collections”? – Do scholars think of collections as “ours”? • Our users search and use our increasingly interdisciplinary collections • Scholars communicate with librarians about our collections • Scholars communicate with each other and create robust digital communities

The Discovery Stage • We presume scholars know the key journals in their field, traditional or open access but what about … – Interdisciplinary research – Students and other consumers of scholarly research – Librarians doing collections development

The Discovery Stage • Where do librarians identify OA journals? Many are in the Directory of Open Access Journals (DOAJ)—we don’t know how many are not… • Randomly searched DOAJ titles in OCLC and not all were found: WHY? • Some libraries do their own thing locally cataloging OA journals/buying bib records

OA Journals: Roots in Print Universe OA journals have their roots in traditional library media: the print and proprietary ejournals that we’ve been describing and making accessible to our users the last 20 years.

OA Journals and Traditional Library Tools What role will the OPAC and MARC play in the future? Will they become obsolete? Will the OPAC lose its primacy as the center of the library? Is MARC a one-dimensional relic? Or will extensible web services integrate traditional stewardship into description and access?

MARC 21 “Extended” • Extensibility = the ability to add new features to an existing program without disturbing any existing code • MARC 21 was designed to be extensible • XML as general metadata format for “open” web • LC metadata schemes for easy cross-walking to MARC: – MARC XML, Library of Congress, version of XML that maps to MARC – MODS=Metadata Object Description Schema (MARC XML “lite”), Library of Congress metadata scheme, for easy crosswalking to MARC, not equivalent to MARC – MADS (Metadata Authority Description Schema) under development Further reading: Yee, Martha M. (2004). New Perspectives on the Shared Cataloging Environment and a MARC 21 Shopping List. Library Resources and Technical Services, 48(3), 165 -178

But Do Scholars Use the OPAC for Finding E-journals? • • • Scientists (especially Bio. Med) use Pub. Med Linking from A&I databases Direct to scholarly associations and publishers Listservs, scholarly blogs Alerting services; RSS direct from publishers, vendors Google; SCIRUS

As an example, let’s consider the impact of the Googlization of our users and our bibliographic universe

A Google Universe More and more users search Google rather than browse or use structured sources because of ease of searching Google Scholar: citations from ejournal vendors, including Ingenta OCLC loading bibliographic records into Google: an portent of things to come

A Google Universe Prediction: in the future, links to far more “library” content will be dumped into Google, bringing users back to our OPACs, link resolvers, databases and A to Z lists

Description, Versioning and Appropriate Copy: Open. URL • • Google Scholar as example Subject search: “social security reform” A citation may have multiple versions Are any of them actually available (for free, or course!) • How to link to “Library” copy? • Link resolver allows linking to OA titles too

Google Scholar extended with Open. URL

Open URL and Open Access Journals Link resolvers: solve “appropriate copy problem” Most work on ISSN as identifier SFX, Serials Solution, etc. harvest Open Access linking information Examples: DOAJ, PLo. S

Versions and Appropriate Copy: Registers • ISSN: – How many OA journals have ISSNs? --most – What about the ones that don’t? – Separate ISSN for “electronic” versions • DOI – How many OA journals are registered? – Plo. S, Biomed Central-- not as many as ISSN

What is DOI? Digital Object Identifier establishes a persistent link to a digital object. provides an identifier "container" which can accommodate any existing identifier: e. g. BOOK: • 10. 5678/ISBN-0 -7645 -4889 -4 Open Access e-journal article (PLo. S Biology) 10. 1371/journal. pbio. 0030127

DOI Resolution Paskin, Norman (2003) On Making and Identifying a “Copy”. D-Lib Magazine http: //dx. doi. org/10. 1045/january 2003 -paskin http: //dx. doi. org--resolver /10. 1045 --D-Lib Magazine /january 2003 -paskin—issue, author (page) URL: http: //www. dlib. org/dlib/january 03/paskin/01 paskin. html

The Versioning Problem We touched on this earlier in our presentation: • Journals available in multiple databases as well as open web, microfilm, paper: not all content equal • Related versions of journals: supplements • Articles available in multiple versions=preprints, postprints • Granularity: connecting the pieces • Dreaded journal title changes • New dreaded article title changes

FRBR to the rescue? Most users may not care about bibliographic relationships between versions and parts of ejournals but librarians and some scholars will. FRBR = Functional Requirements for Bibliographic Records, approved by IFLA in 1997 to address relationships between versions and show hierarchy between bibliographic entities at four levels (from most specific to most abstract): 1. 2. 3. 4. Item Manifestation Expression Work • • Based on ideas by Smiraglia & Tillett Easy example: versions of “Romeo and Juliet” Will improve how catalogs work U. S. library community plans to implement it in the next few years. • • • NYTSL presentation IFLA FRBR Review Group Tillett What Is FRBR? publication/. pdf

FRBR Group Entities and Primary Relationships

Display in VTLS FRBRized catalog

FRBR could be very helpful for OA journals re. collocating • Different bib records for e-journals – “one–record model” CONSER model isn’t always implemented, particularly as we load vendor-supplied bib records • Different physical and intellectual versions of the same ejournal article – eprint versions – paper versions • Connecting different levels of related entities such as indexes, supplements • Easy display of dreaded title change history

FRBR and serials These slides come from the VTLS website and represent a journal and journal issues in a FRBRized catalog Source: http: //www. vtls. com/documents/FRBR 9. PPT

Journal Indexing Analytics • Component Parts – Work = Journal title – Expression = Issue information (enumeration and chronology) – Manifestation = Article record

Journal Indexing Analytics The Work record for the Journal, Brigham Young University Studies

Journal Indexing Analytics The expanded tree showing three Expressions: One for the title itself and the other for individual issues

Journal Indexing Analytics The tree expanded to show the Manifestations of the first two Expressions The Expression record for the issue v. 36 no. 1 1995 as shown in the 866 tag

Journal Indexing Analytics Manifestation (analytic) record for the issue v. 36 no. 1 1995

Journal Indexing Analytics Manifestation (analytic) record for the issue v. 36 no. 2 1995

Let a million standards blossom Standards for describing, identifying, providing rights information and preserving web-borne content are still very much under development so it will take a couple of years before we reach more consensus

Let a million standards blossom Interoperability is the key concept here: since we have so many metadata standards, they need to be interoperable.

Let a million standards blossom • OAI-PMH as developed by Herbert Van de Sompel at Los Alamos has emerged as the leading standard for web-borne/open access content • OAI XML Schema to represent MARC records: translate MARC to XML but not back • OA publishers/authors use it. Repositories register with the Open Archives Initiative. • Is description of content is truly robust? • How does it compare to bibliographic information in the OPAC? Very different but difference is important.

Let a million standards blossom Consider mapping ejournal metadata back to MARC: • CUFTS at Simon Frazier University offers MARC records • Cornell University has a done similar project for their ejournals, translating xml-based metadata. (See http: //nylink. suny. edu/docu/sl/184 j_a 04. pdf) • Anna Hood talk @ NASIG, two year project importing DOAJ records into Kent State OPAC • The University of Illinois at Urbana-Champaign Open Archives Initiative Metadata Harvesting Project provides software to convert from MARC to other metadata and back

Let a million standards blossom • Vendors like Serials Solutions and Ex. Libris offer MARC records for ejournals including Open Access journals but they may not be high quality or may be records for paper version of journal with 856 field added.

Let a million standards blossom • Dublin Core continues to be important. It provides a unique identifier element that links metadata to content • Users tend to not use enough elements: Jewel Ward’s 2002 study of 82 Open Archives Initiative providers found use of an average of only eight Dublin Core elements (Bruce & Hillman) • DC lacks the 362 field for designation (volume/number and dates of coverage) and a date formatting issue (21 -22) [article says this is under discussion @ MARBI] (Copeland)

Stewardship is key • Who is describing the open access journal? Someone in the academic department creating the journal? The author? A librarian? A granting agency? • Who will take care to preserve the open access journal and follow the standards and protocols that the open access community has embraced? • How is it being described (MARC vs. other metadata) and at what level? Are the levels interconnected?

Conclusions • OPAC will evolve in different ways – FRBR – Greater harmony w/XML – will be integrated with ERM and link resolver systems – Will OA journal records find their way into bibliographic utilities? • Standards including identifiers will hopefully become more cohesive • User behaviors on the Web will change methods of description and access