[Photo]

Greg R. Notess
Reference Librarian
Montana State University

Web Wanderings

PubList and jake: Free Periodical Reference Sources

EContent, October 2000
Copyright © Online Inc.





Serials librarians are well acquainted with the abundance of potholes that make management and bibliographic control of periodicals so difficult–title changes, variations in frequency, split publications, supplements, special issues, publisher misinformation, and multiple formats. In this increasingly electronic age of periodicals, the dilemmas are increasing in complexity as new situations arise. Full-text periodicals online, databases with ever-growing full-text components, and a constantly changing list of online titles make the job of managing and researching information about periodicals an increasingly challenging task.

Fortunately, along with the potholes come some new tools, freely available on the Web, that can assist the information professional in finding information about periodicals as well as offer some interesting approaches to managing some of the difficulties. In particular, PubList and jake have much to offer. While neither comes anywhere close to solving all periodical problems, they are both tools that can be used to look up different kinds of information about periodicals and give pointers on the availability of full-text versions and URLs.

PUBLIST

One long-time standard source for information on currently published periodicals is the Ulrich's International Periodicals Directory. Available in print, CD-ROM, and via several commercial online sources, Ulrich's coverage of the majority of published periodicals provides basic, useful information about each publication. This crucial reference source is not yet fully available for free on the Web, but as an alternative, PubList (http://www.publist.com), from Bowes & Company, which was recently acquired by Infotrieve, comes very close and proves an excellent resource for quick directory-style information about periodicals.

PubList offers a searchable directory of over 150,000 periodicals, with much of the data originating from Ulrich's (and properly credited). The site reports that some additional information comes from Editor & Publisher as well as from participating publishers such as MIT Press, Jossey-Bass, the University of Chicago Press, and the New England Journal of Medicine. Secondary publishing services that are associated with PubList include the Academic Book Center. PubList has partnered with the British Library Document Supply Centre, CISTI, Infotrieve, and UnCover to supply document delivery services. It also has arrangements with Enews. com as its subscription partner, the Copyright Clearance Center as a rights and permissions partner, Alfred Jaeger, Inc. for back volumes, and Amazon for books. These partnerships mean that once PubList users find a record for a specific publication, they then have opportunities for subscribing, ordering individual articles, buying back issues, and managing copyright clearance.

PubList is a free resource and offers much of the basic information that appears in other forms of Ulrich's: Title, Subtitle, Publication Type, Status, Frequency, ISSN, Editor, Copyright Clearance Center Registration Status, Subjects, Circulation, Special Features, and Publisher. Some of these fields are not available in the initial record display but are available in the full record. The publisher information includes address, phone numbers, fax, and URL.

MISSING AND ENHANCEMENT ISSUES

Unlike the other versions of Ulrich's, some fields are not available via PubList. The commercial versions of Ulrich's add price information, first year published, Dewey and LC call numbers, British Library Shelf Mark, parallel title, and several other fields. They also offer the ceased title information that is not available in PubList.

These missing fields provide reasons to subscribe to a full version of Ulrich's or to use the version in online commercial systems. However, for many questions, the information available for free at PubList may be more than sufficient. PubList, for example, makes it easy to find basic publisher information, a URL, or circulation numbers.

PubList also adds some information to the basic Ulrich's starting point. For example, it has a Pre-Launch Publication Listing that has information on new, soon-to-be-published U.S. periodicals. This includes publication and publisher information as well as links to electronic versions where available. This added information comes from direct submissions to PubList and LC's National Serials Data Program.

PubList offers a premium service to publishers, where, for a fee, they can add their own logo, cover shot, title, promotional graphic, promotional message, publication description, publisher Web link, "anytime" updating, Web traffic reports, and subject selection. This premium service also includes preferred positioning within search results.

In general, this is not a big distraction for users, since a relatively small percentage of titles have paid profiles. On a search for marketing, the first hit is Marketing with Honors, a periodical that the publisher paid to add a profile, but the second through fifth hits are all periodicals that have the exact title Marketing.

SEARCH AND ACCURACY ISSUES

Searchers should note that the regular search does an automatic phrase search starting with the beginning words in the titles. Thus, a basic search on science teacher will not find the Journal of Science Teacher Education but only Science Teacher Education. PubList does offer several other search options under the Search button at the top of the screen. For keyword title search options and truncation capabilities, use the specialized title search form. However, the title search form takes much longer to respond than the regular search box, especially with a multiple keyword search.

With any directory of well over 100,000 records, there are many errors and discrepancies. The editor of this magazine, for example, is Marydee Ojala, not the Marydee Ofala that PubList claims. The URLs are often suspect. Sometimes they point to an accurate location, but many point to electronic periodical services, old URLs, related URLs, or just plain mistakes. This is a problem that is not unique to PubList, since some of these URLs are in the other versions of Ulrich's as well.

Try a popular title such as Newsweek. Both PubList and Ulrich's via Dialog give two URLs: http://www.newsweek-int.com and http://www.mediabuzz.com. Both of these have some Newsweek connections and content, but http://www.newsweek.com is a far more accurate URL that, while redirected, actually brings up the main Newsweek Web site. Even worse are records with completely incorrect information, such as the one for Searcher where the listed URL is http://www.well.com/–certainly not the Web site for Searcher. So be wary of the given URLs and do not take them as the definitive Web address.

INDEXING COVERAGE

PubList also offers its own information on which indexes or abstracting services cover certain periodicals. While Ulrich's lists abstracting and indexing services that index the articles within a particular periodical, PubList does not have that data from Ulrich's. Instead, it has created its own listing of coverage for about 40 indexes. This is not as complete as Ulrich's, but it can serve as a useful supplement.

Looking at Newsweek again, PubList offers seven indexes while the full Ulrich's has 44. It is a functional start, but it has a long way to go. PubList acknowledges the limitations and even states that "Our list of indexes is not intended to be comprehensive." Even the full Ulrich's is not comprehensive, although it offers much more information on possible indexing coverage. For even broader coverage, especially of online database sources, you need to move beyond PubList to jake.

FULL-TEXT ANOMALIES

Within libraries, information brokerages, and information centers, determining which service indexes which periodicals and to what extent is a frequent information need. And with the recent rise of multiple full-text sources for articles, it is helpful to have the ability to discover both the indexing coverage of databases as well as the full-text availability of titles.

PubList, Ulrich's, Dialog's Journal Name Finder, Full-Text Sources Online, Hermograph's Net.Journal, and many other traditional resources have proved helpful for finding this information, although none has been comprehensive or entirely accurate, given the rapid changes that occur in full-text coverage. Yet as more libraries move towards managing large collections of full-text resources, many are building their own databases to try to maintain this information, either in their catalog or in a separate database. As with so many serials issues, this database-building and maintenance engender a host of problems and difficulties.

The traditional indexing coverage problem could vary slightly from vendor to vendor, but in general, librarians expect that INSPEC via Ovid will have indexed the same journals as INSPEC via SilverPlatter. However, with full-text added to traditional bibliographic databases, the full-text titles can change significantly from one vendor to another.

VERSIONING ISSUES

Wilson's Applied Science and Technology database demonstrates one version of this variance. While the basic bibliographic database remains the same, several vendors offer it with full-text titles. The ProQuest version adds full-text titles from ProQuest's collection. Wilson has its own full-text version. Via SilverPlatter's SilverLinker, the same database could be linked to local subscriptions, with each individual library offering its own set of full-text titles. Most of the services have title lists of what is included for indexing as well as in full text. The better lists also include accurate dates of citation coverage, ISSNs, and full-text coverage dates.

The difficulties with maintenance and building of such databases and the complexities of including such titles in an online catalog are immense, and well beyond the scope of this column. Yet cooperative development has aided in the solution of similar problems in the past; jake is one approach that can help those looking at building their own database as well as the independent researcher who would like to determine indexing or full-text availability of specific periodicals.

IT'S JAKE WITH ME

The Jointly Administered Knowledge Environment, or jake as it is more commonly known (http:/jake.med.yale.edu), is a public domain resource, available under the GNU General Public License. Hosted at and supported in part by the Cushing/ Whitney Medical Library at the Yale University School of Medicine, jake is designed as a reference source for finding, managing, and linking online periodicals and articles. Its aim is to include the coverage of online bibliographic and full-text databases. In a sense, jake is a union list of database holdings, combined with an attempt at authority control and tools to link local holdings. It actually can have many functions, and can be used to look up the full-text availability of a single title. The entire database can be downloaded and incorporated into some local project. It has links to enable the creation of very simple MARC records for full-text titles. It can also be customized for the display of locally held databases only.

Do not be fooled by jake's bare-bones look and programming-oriented documentation. These disguise the value of the project, and no programming knowledge is necessary to use it. Try a search for a periodical title such as Newsweek in the basic search box. jake finds 13 matches, only two of which have "complete details"–one for Newsweek and the other for Newsweek International. The complete details for Newsweek list 68 databases that index Newsweek, and 44 that have it in full-text.

While this is a higher number than the indexing coverage listed in Ulrich's, the numbers can be misleading. In jake, databases can be listed multiple times, once for each service from which it is available. For example, four different versions of ABI/INFORM are listed–one from Ovid, another via OCLC FirstSearch, and two from ProQuest (the Global and Research editions). Given the variations in full-text availability among these services, this multiple listing can be especially helpful, as long as the user understands what the differences are.

In addition, there are many indexes that are not included in jake that do show up in Ulrich's, since the latter includes print indexes–jake only covers online indexes. On the other hand, Ulrich's does not cover full-text sources, nor does it attempt to differentiate between the various versions of a database as offered by different vendors, while jake includes both indexing coverage and full-text coverage. It also includes the full date range for each kind of coverage, another feature not available from Ulrich's.

WORK IN PROGRESS

jake is still very much a work in progress. While it is certainly useable and useful now, expect to see future developments and occasional instability. One current problem with jake is that it does not clearly display the dates when each title list was last updated. Many of these vendor lists change monthly, if not more frequently. With constant changes from publishers deciding to pull from or add certain titles to the full-text availability, these lists need to change frequently. And knowing what date jake updated its listing would help.

Neither jake nor PubList is going to completely fill the periodicals potholes scattered throughout the online landscape of electronic serials. Yet their free availability and continued development provide excellent resources for quick information about publications within the ever-changing realm of periodicals.


Communications to the author should be addressed to Greg R. Notess, Montana State University Libraries, Bozeman, MT 59717-0332; 406/994-6563; greg@notess.com ; http://www.notess.com.