Journal Title Display and Citation Practices

Transcription

1 The Serials Librarian From the Printed Page to the Digital Age ISSN: X (Print) (Online) Journal homepage: Journal Title Display and Citation Practices Les Hawkins, Regina Reynolds, Steven C. Shadle & Deberah England To cite this article: Les Hawkins, Regina Reynolds, Steven C. Shadle & Deberah England (2009) Journal Title Display and Citation Practices, The Serials Librarian, 56:1-4, , DOI: / To link to this article: Published online: 13 Mar Submit your article to this journal Article views: 90 View related articles Citing articles: 1 View citing articles Full Terms & Conditions of access and use can be found at

2 The Serials Librarian, 56: , 2009 Copyright The North American Serials Interest Group, Inc. ISSN: X print/ online DOI: / Journal Title Display and Citation Practices WSER X The Serials Librarian, Vol. 56, No. 1-2, Januray 2009: pp. 0 0 Journal Tactics Sessions Title Display and Citation Practices LES HAWKINS, REGINA REYNOLDS, and STEVEN C. SHADLE Presenters DEBERAH ENGLAND Recorder In this tactics session two Library of Congress catalogers and one former Library of Congress cataloger discussed how metadata issues with journal title presentation on provider and publisher websites can be problematic. Through a series of examples, the speakers demonstrated how lack of metadata and citation practices result in users not finding content and libraries not getting access to content for which they paid. Reactions from publishers and attendees were welcomed with the goal of enhancing communication among stakeholders and generating interest to form a collaborative effort in establishing a NISO working group to develop best practices with journal title display and citation practice. KEYWORDS metadata, journal title display, citation practices, NISO, publisher websites, best practices In this tactics session presented on the last morning of the conference, Les Hawkins and Regina Reynolds, both of the Library of Congress, and University of Washington s Steven C. Shadle, addressed metadata problems with journal title presentation and their impact on user access and electronic resource delivery systems. Les Hawkins, CONSER (Cooperative ONline SERials) Coordinator at the Library of Congress since 2003, introduced the topic by stating that this topic of metadata on content provider websites was a result of discussions that came up on the SERIALST listserv last October. Hawkins cited this was an easy topic for librarians to talk about, but he and his colleagues wanted to draw perspectives from other players in the serials chain. The speakers anticipated the discussion would be an opportunity to build understanding with what s happening on the provider and publisher end of the journal supply chain. 271

3 272 Tactics Sessions To that end, Bob Boissy of Springer graciously offered to act as a reactor to present a publisher s perspective on the topic at hand after the speakers concluded their presentations. Hawkins noted the panel welcomed comments from other service and content providers with the aim of enhancing communication among stakeholders and to form a collaborative effort to resolve issues surrounding title presentation on provider and publisher websites. A desired outcome of the presentation, according to Hawkins, was to gauge and develop interest in the topic with the goal of forming a group of stakeholders drawn from all sectors in the journal supply chain with the objective of forming a NISO working group to establish best practices in regards to metadata for journal title display and citation practices. ISSUES WITH CATALOGING AND ISSN RELATED WEBSITE DISPLAY FROM A USER PERSPECTIVE In illustrating the individual user s perspective of connecting journal citations with content, Regina Reynolds, Head of the U.S. ISSN Center at the National Serials Data Program, introduced Serials Sam Jr., son of Serials Sam and Mona Monograph and as such brought a welcomed lighthearted story approach to the topic. Before launching into her examples Reynolds expressed concern that we re stranding our users, and leading our researchers right off a cliff with nowhere to go. In her first example Reynolds described a situation whereby Sam Jr., who is researching worm disease, finds a citation in Military Medicine for the article The Results of Hookworm Disease Prophylaxis, which was published in volume 2 (1922) of the American Journal of Hygiene. With citation in hand, Sam Jr. searches a bit with no success until he happens upon the title in WorldCat where in the exploded view he sees three microfilm and one Internet resource icon. Sam Jr. clicks on the Internet resource icon, which instead of linking him to the American Journal of Hygiene, links him to the American Journal of Epidemiology. According to Reynolds, like his father, Sam Jr. is not the brightest light bulb in the chandelier and does not know hygiene from epidemiology. All he knows is he s been given this assignment, which he does not understand very well and thus he s understandably puzzled. After a bit, Sam Jr. decides to check the index for American Journal of Hygiene and even though he has a bit of trouble with the alphabet, he concludes it is not there. He cannot find his journal and per Reynolds we ve led him off the cliff again. After obtaining some help from his public service librarian, Sam Jr. tries WorldCat again. This time he sees a Web resource link. He clicks the link and finds he s back in the American Journal of Epidemiology! He then decides to browse the archive after the librarian explained what an archive

4 Journal Title Display and Citation Practices 273 is. When he clicks on the all online issues he finds January 1921 to June Because he s looking for 1922, he thinks he will find it now. So he scrolls and scrolls and sees that the library has January 1921 to He then notices the fine print note at the bottom of the page that indicates the journal was published under the title of American Journal of Hygiene before Armed with this knowledge, Sam is successful and finds the 1922 issue he needs. Even though Sam Jr. finally found the issue for which he was searching, Reynolds remarked that most students would have fallen off the cliff five steps ago. If Sam Jr. had come in through HighWire citations he would have been confronted with a different metadata issue whereby the citation itself was updated to display American Journal of Epidemiology January 1922, even though there was no such title in January According to Reynolds, somebody has kindly, so they think, rewritten some history with this citation and has probably caused confusion. Even if all citations could be updated to the current title, Reynolds emphasized we can t go around and erase all these citations that are in printed publications or are now being digitized. It s impossible to rewrite the whole citation infrastructure to display the current title, even if some of the citations online can be rewritten as with the HighWire example. Reynolds declared you can t blame Sam Jr. and our users for being frustrated! If Sam Jr. had looked up American Journal of Epidemiology in the ISSN portal record or in the catalog he would have seen that the title is continued as the American Journal of Hygiene. If he had pulled up the corresponding record for the American Journal of Hygiene he would have seen the continues and continued by, even if our systems don t always make good use of the metadata, per Reynolds. In a subsequent example Reynolds demonstrated issues with inconsistent use of metadata in website displays. The journal SHINe, which is published by a library, is listed as Volume 44, Spring 2004 on the publisher s index page. However, when a user clicks the link, the title page displays Issue 44, April 2004, and in yet another location on the same page the issue is referred to as no. 44. In addition to this inconsistent terminology the title is actually Interim. It s no wonder students get frustrated and confused, Reynolds concluded. Sometimes breadcrumbs are left along the way for users. In Reynolds third example, she offered a GeoRef citation for an article that appeared in the Journal of the Geological Society of Australia. The citation contained both the current ISSN and an earlier ISSN of When clicked, the ISSN from the citation takes the user to the current title, Australian Journal of Earth Sciences, which includes the note previously published as Journal of the Geological Society of Australia. By presenting both the current and earlier ISSN and including the previously published note, users could find their way to the desired article.

5 274 Tactics Sessions Reynolds stressed, If I found following these leads somewhat tedious, and I know what I m doing, theoretically, think about a student who has no idea how serials work, has no idea about latest and successive entry, and title changes, and all that. The breadcrumbs are good, but they re not good enough. Reynolds self-described plea to providers and publishers was that if you do nothing else, please index all the earlier titles because you do have coverage for that title. Reynolds suggested this practice would allow a publisher to state they covered more journals when they typically don t include all the earlier titles in a title list. It would also, posited Reynolds, add to a publisher s statistics. In her fourth example, Reynolds presented a publisher s index page for the Journal of Interdisciplinary and Multidisciplinary Research, which listed volume 1, issue 1 (2007) as the earliest issue. When a user follows the link from the index page to the volume 1, issue 1 page, the title displayed is the earlier title, Journal of Multidisciplinary Research. Once again, per Reynolds, an earlier title has been buried. Reynolds noted some of the examples she spotlighted were a little more obvious than the first example of using American Journal of Hygiene and American Journal of Epidemiology. She noted, It will never be perfect; serials are a very imperfect discipline. Regardless of the imperfections, Reynolds challenged that We can t rely on the fact that some title changes are obvious and some are not. There s an awful lot [of title changes] that look like a totally different publication. A MODEL FOR JOURNAL TITLE DISPLAY PRACTICE In wrapping up her portion of the program, Reynolds used JSTOR as an example of doing it right. According to Reynolds, JSTOR is not alone in this; there are many others who are doing it right and can provide a model of how it could be done in order to align by sense, align with cataloging records, and align with citations that were based at the time the journal first appeared. To make her case, Reynolds offered for an example JSTOR s title presentation for Journal of Hygiene. JSTOR s title presentation is clean and provides a note and link to Epidemiology and Infection, the later title. The journal title presentation contains the continues and continued by and lists the coverage under each title. The JSTOR model aligns with the catalog records, the ISSNs, and citations. The ideal, at least from an ISSN and cataloging perspective, and from the perspective of lining up the citations that were created at the time the article first appeared, and keeping everything synchronized would be to identify each major title change with a separate ISSN. In Web displays each title would be listed separately. Each title would be indexed to ensure that a

6 Journal Title Display and Citation Practices 275 user who arrives at the index would know the publisher has coverage for the other title changes as well. If a publisher does not know which changes were major and which were minor, their friendly ISSN Center can help, advised Reynolds. Publishers need to check the ISSN Center when they make a title change and at that point they can receive information on each major change in title. If cataloging, ISSN, and citations were all aligned this would lead us to the promised land where we can all rest easy, according to Reynolds. IMPACT OF TITLE DISPLAY DECISIONS FROM AN E-RESOURCE DELIVERY SYSTEM PERSPECTIVE Steven C. Shadle, Serials Librarian at the University of Washington, began his portion of the program by recognizing Ann Ercelawn s Best Practices for Ejournals: Publication and Website Design Guidelines, which are available on Vanderbilt s website. Per Shadle, Ercelawn s practices provide really good, clear, straightforward guidelines for the kind of e-journal content that should be displayed on websites. Shadle stated his approach to the topic at hand was from a behind the scenes, library system perspective for a typical cataloger or serials data manager. Using one title, Zoonoses and Public Health, which had a title change, Shadle illustrated some of the problems that surface when libraries and providers or vendors do not receive the correct metadata from the publisher. Shadle used Zoonoses and Public Health, which began in 2007 and continued Journal of Veterinary Medicine, Series B ( ), to show the data going through the data streams in the various environments and systems that are used to provide access. On Blackwell Synergy s home page for the title, the entire run or what has been digitized to date is listed. The latter title has a reference that the title was formerly known as Journal of Veterinary Medicine, Series B, but in this particular case, per Shadle, you don t know which set of issues were published under which set of titles. Using Sam Jr. as his example of a user, Shadle observed there might be a breadcrumb that will help Sam Jr. realize this really is the same thing. He might, if he s motivated, see the years and think to click the link for 2005 because the article he s looking for was published in University of Washington uses their PAMS (Publication Access Management Services) vendor, Serials Solutions, as their data provider and Serials Solutions, in turn, gets their data from content providers and publishers. Using a screen shot of Serials Solutions Client Center, Shadle pointed out that nearly all the sources listed, IngentaConnect, OhioLINK s Electronic Journal Center, and various Blackwell packages list Zoonoses and Public Health as starting in However, four entries begin with the earlier dates and those are

7 276 Tactics Sessions actually the earlier title coverage. The data provider, which is not necessarily the publisher, provided the entire run of the serial represented with this one title. Shadle commented that he has advised his Serials Acquisition staff that if they are working with a particular title and they see several start dates that are the same and then they see one or two sources that have an earlier start date, investigate that. Check the MARC record to make sure the dates are okay. If staff do not investigate the dates and select an entry that has dates of because they think we are getting access through the back files what kind of problems are we getting into at OCLC? What kinds of problems does it cause if those dates do not match? In his next example, Shadle offered an example of Zoonoses and Public Health displayed in the WorldCat Local version of University of Washington s local catalog. In a search of the title a record is retrieved that shows the holdings for this title are 1999 present. Per Shadle, that s sort of what we expect. The data is supplied by Serials Solutions who provide the link and the link is propagated through to the library catalog and displayed to the public in this form. While looking at the example, Shadle also checked title displays for Zoonoses and Public Health in WorldCat Local for consortial partners, namely Wazzu, Washington State University. Washington State s display also shows 1999 to date. They have the same exact issue. So it s not just my issue, concluded Shadle. If a data provider does not provide the data for the earlier title, the earlier ISSN, then there s no earlier catalog record and no reference to the online content when a title is searched in the catalog. The user would retrieve a record indicating print through 1995 was available when in reality electronic content is available beginning with 1999 up to Metadata issues with journal title display can also have an impact on link resolvers like WebBridge and SFX resolving correctly. To illustrate this, Shadle pointed out how the Journal of Veterinary Medicine, Series B, and the earlier title for Zoonoses and Public Health is not displayed in the list of titles from Blackwell Synergy. Series A and Series C are available, but not Series B. Again, per Shadle, this is because the publisher or the content provider did not provide the metadata to Serials Solutions who in turn could not make the data available to us. The title, the ISSN, and the coverage are not surfacing in the link resolver and knowledgebase. However, farther down in the Blackwell Synergy STM package, Zoonoses and Public Health with the ISSN and the start date of 1999 is available. In this case the data is getting propagated to our coverage to our knowledgebase. What affect does this have on our systems? Shadle demonstrated the impact using the terms dog rabies in a search of ISI Web of Science. The search retrieved a citation for Zoonoses and Public Health, a Journal of Veterinary Medicine, Series B citation from 2004 and 2008, respectively. If users click the link resolver link for the 2008

8 Journal Title Display and Citation Practices 277 citation, they are served with a link to the full text along with a link to the title; from there they can access the article content. Everything is as expected. However, if they click the link resolver link for the 2004 article, they are not served with a link to the full text because the title is not being profiled in the knowledgebase. Instead the user is given the option to search the catalog because that s University of Washington s fall back. Shadle noted the University of Washington s practice is if no full text is available, serve them a title and title link. If no title link is available, serve them a link to the catalog. How many people do you think will go the extra mile? Reynolds asked. Would they go the extra mile only if they re really desperate? Shadle responded with they might look at the catalog link and respond why do I want to go to the catalog anyways? There s definitely that possibility, Shadle surmised. If they go to the catalog, the ISSN and a title search is going to take them to that catalog record, which again, doesn t have the online link because the data hasn t been propagated. If it s in print forget it both Shadle and Reynolds concluded. Per Shadle, It s not the years associated with the title because the citation is for 2004 and the print holdings only go to 1995, so the user surmises that he has to go to Interlibrary Loan, or Google. In short, the library is paying for content that the user isn t getting, noted Reynolds. The content the library paid for is basically hidden because the metadata that is associated with it has not been propagated through the system. I m not telling you anything that you don t know, noted Shadle. You all know this; we re sort of preaching to the choir here. Using the same earlier title example, Shadle checked Washington State s display to see if they had the same problem. University of Washington s earlier title was hidden; however, Washington State s content provider provided the earlier title with the same coverage and the earlier ISSN. Somehow they got that electronic content to that earlier title. Shadle then pointed out the coverage shows 1999 to current, but the catalog record shows Shadle wondered if perhaps we are heading towards a situation in which the publisher may be unable to identify the individual ISSN assignments through the chain, which are needed to correctly match up those assignments with our data. Maybe it s enough to at least get the ISSN in there? We re getting back to that Super ISSN thing so that the ISSN can serve us the date as the linking mechanism for it to connect with our systems and who really cares about the coverage? Shadle stated. Is it enough that the coverage can be set for the entire range of years? Is it enough to get the links in our system so the user can get to the full text and our link resolver will continue to work? In concluding the speakers presentation Reynolds observed that one of the issues is the different ways a user might access content and if the citation no longer matches up and the ISSN is not leading them to the content,

9 278 Tactics Sessions the user is getting confused along the way. Reynolds remarked we need to try and come up with use cases and walk through the various use cases to determine where a user can and can t get access, and where, if they re really patient and sharp, they still cannot get access. Reynolds also expressed concern about OCLC s xissn, which goes a step further than a linking-issn in connecting an entire cluster of earlier, later, and subsequent data due to the problems with ISSN in WorldCat. Since the session was held, OCLC released their xissn Web service, which assists libraries and participants in the journal supply chain and related services with facilitating the management of serials metadata with the aim of retrieving serial relationship information for Web-based applications, such as library catalogs, knowledgebases, and link resolvers. At its core, the WorldCat-based xissn service provides information about earlier, later, subsequent, and alternate titles, as well as electronic ISSNs for print titles by retrieving lists of related ISSNs and titles associated with a submitted ISSN. The results retrieved are based on relationships in WorldCat. To enhance look up, the OCLC service offers a free ISSN History Visualization Tool, which displays in chart form, the history of a journal with a given ISSN. For Web-based applications, the xissn server supports OpenURL and unapi protocols and retrieves data in XML, Python, JSON, or Ruby formats. A PUBLISHER S PERSPECTIVE Springer s Bob Boissy contributed to the discussion in his role as volunteer reactor. He noted the topic was very interesting and raised awareness for an issue of which most publishers are not aware. To illustrate, Boissy recalled some conversations he had with Springer s Web design team who were proud of all the new things they are putting on their new site and how they have connected old and new titles. Boissy said, I m surprised at how many times editors have come up to me and asked questions about how they could put a journal on the website. Per Boissy, editors have completely diverse ideas about it. Some of them want to pull all content under the current version of the title because they have a marketing bias. Basically, they want to put all content under the current form of the title. Boissy acknowledged that he tries to make very clear to them that they should not do that; they should put the current content under the current title, and the older content under the earlier title. In other words, the content should be placed under the form of title at the time it was published. Springer has, noted Boissy, hundreds of editors and only a handful actually talk to the resident librarians. The issue is a real one, remarked Boissy. The Internet has not been around since He noted that we are talking about spending a lot of time on historical reconstruction to include content about a title that maybe

10 Journal Title Display and Citation Practices 279 was acquired in the 1970s, even though it s been published since the 1850s. As a publisher, considering what is going on with the content and the need for usage statistics, for that reason alone, I think it s worth spending time to work on it, Boissy noted. For Springer, there s pride of publisher to link content and to work on it. Another factor contributing to the status quo, at least at Springer, is the layer of product managers who concern themselves with making the product better. They make the case for usage and promoting older titles to add title value. However, they do not get detailed with metadata. They are concerned with new products and product placement for the new products in the marketplace. Springer has library advisory boards, but again, the problem with the boards are they do not ask the kind of questions we are raising today. They do not ask the detailed questions; they typically comment on the design of the website, the user interface, product performance, and do not get into details. Moreover, the librarians on the boards are usually associate university librarians (AULs) or above and obviously are not at the level you d like us to be. They are not as specific as needed to address issues like this. Another problem is with small publishers because they typically do not have library advisory boards. Boissy acknowledged that NASIG might be willing to step in and provide volunteers for library advisory boards, particularly for smaller publishers. He said, That s a very appropriate role for NASIG and it will provide input on the problems, the very crux of what we re talking about. Boissy added, I hope more librarians will think of a career in publishing. Reynolds quipped, that may happen regardless of your plug with library budgets decreasing. PARTICIPANT PERSPECTIVE Discussion after Boissy s reaction first focused on price lists, including what to include in the coverage and how to present the information without the price list being ominous, before turning back to the topic at hand. Reynolds commented that she was surprised when she found the 1922 article cited in every current publication. Reynolds concluded what has come out of the invasion of back files is they still have use. Publishers should offer libraries the ability to get as much of their collection in the online form as possible. Lettie Conrad of Sage Publications noted that she s thought about all the ways to do this and understands what the speakers are saying about users not finding the content. She s noted two key points regarding metadata and the need to get the history credited in the Web display. Conrad also mentioned Sage is working on and welcomes library advisory boards. Reynolds reiterated for the publishers in attendance that the ISSN Center

11 280 Tactics Sessions can help with aligning the metadata. The ISSN Center cannot solve all problems, but they can help track the ISSN for the publisher. Christie Degener of the University of North Carolina at Chapel Hill Health Sciences commented on a couple of observations from the discussion. First, this is going to turn up more and more of the older citations. Secondly, from a collection development standpoint, we are not getting full value for what we have paid. Finally, when Sam Jr. goes to Reference for help, we re tripping up our Reference colleagues. East Carolina University s Eleanor Cook commented that she finds this ironic. We ve been through retrospective conversions. Now that publishers are the ones providing the platform, what they need are catalogers. They need consultant catalogers, catalogers, period. The records are in WorldCat. It s all there; we ve done the work! In response, Reynolds pointed out that WorldCat s focus is as a portal. The problem, according to Reynolds, is that the United States and Germany disagree on the ISSN. We need to become shadow data wranglers and publishers need to realize that it is important to have data wranglers. NEXT STEPS The panel s goal is to obtain NISO s agreement to put together a working group with representation from a variety of stakeholders to hammer out what is the best approach, what is the best solution, because there may be some solution behind the scenes to help this work. Reynolds noted she hopes this will provide the opportunity to work with ISI providers and knowledgebase providers to clean up some of the bad metadata. Cleaning up that data regardless of what else we do is a good goal, according to Reynolds. Hawkins announced the next step is to begin the process in the CONSER community, and then get stakeholders from all players involved to submit a proposal for the working group to NISO with the aim of establishing best practices journal title display and citations. Mitch Turitz of San Francisco State University pointed out that you can pass a standard but that doesn t mean publishers will follow it. Turitz noted earlier standards, like SICI (Serial Item and Contribution Identifier), that were not implemented across the board by publishers. Reynolds cited that the panel s objective was to establish a best practice instead of a standard. Best practices, according to Reynolds, are implemented quicker, tend to be less rigid, and NISO likes best practices. We need to have the involvement of all the players to make this work, emphasized Reynolds. This is never going to be perfect, but if we aim for as much consistency as possible, we ll still have enough inconsistency to deal with. The session wrapped up with a call for volunteers.

12 Journal Title Display and Citation Practices 281 RELATED ARTICLES AND RESOURCES 1. Ann Ercelawn, Best Practices for Ejournals: Publication and Website Design Guideline. August Vanderbilt University Library. vanderbilt.edu/ercelawn/bestpractices.htm (accessed June 8, 2008). 2. National Serials Data Program, Library of Congress, What s in a Name? Presentation Guidelines for Serial Publication Library of Congress. (accessed June 8, 2008). 3. Robert Boissy and Linda Beebe, Drinking From the Firehose Two Responses to Top 10 Suggestions to Publishers of E-Journals, Against the Grain 18, no. 4 (2006). (accessed June 8, 2008). 4. UKSG, Link Resolvers and the Serials Supply Chain. sites/uksg.org/files/uksg_link_resolvers_final_report.pdf (accessed June 8, 2008). 5. UKSG, KBART: Knowledge Bases and Related Tools Working Group. (accessed June 8, 2008). 6. UKSG, Transfer. (accessed June 8, 2008). CONTRIBUTOR NOTES Les Hawkins is CONSER Coordinator for the Library of Congress. Regina Reynolds is the Head of the National Serials Data Program, the U.S. ISSN Center of the Library of Congress. Steven C. Shadle is the Serials Access Librarian at the University of Washington Libraries. Deberah England is the Electronic Resources Librarian at Wright State University.