Péter's Digital Reference Shelf
October 2004

Title: Kluwer Journals Online
Publisher: Kluwer Academic Publishers
URL: http://journals.kluweronline.com/
Price: bibliographic records with abstracts are free
Tested: August 18-20, 2004

The Context

Kluwer Academic Publishers has been known as the second largest publisher of scientific-technical and medical journals (behind Elsevier), but this does not mean that the two publishers are in the same league — they are not, in more ways than one. But first, a little background.

In 1997 Kluwer was on the brink of merging — poised to walk down the aisle with more than $30 billion in their bank account. By the spring of 1998 the prenuptial talks concluded that it would not be a match made in heaven. When the wedding was canceled you could hear the collective sigh of relief from other large publishers (like Wiley, Thomson and Springer) and, most of all, from librarians who were concerned about the implied price increase. Kluwer kept looking under the guidance of her parent company Cinven, the prominent buy-out company which finally arranged a marriage late last year between Kluwer and Springer.

While the press refers to the couple as Kluwer & Springer, Kluwer did not keep her maiden name. So why do I refer to its online archive as Kluwer Journals Online? Because the two have not yet moved in under the same roof, as the logo and Web address clearly indicate.

The Content

Kluwer has more than 650 journals (a little fewer than Springer) and even the joint assets represent just a bit more than two-third of Elsevier's journal collection. More importantly, the Kluwer Journals Online (KJO) archive has less than 300,000 articles, while Elsevier's ScienceDirect has six million.

True, ScienceDirect is searchable only for subscribers, but Scirus, an open access archive of Elsevier, has free bibliographic citations and abstracts for almost all of Elsevier's journals nicely integrated with more than 10 million indexing and abstracting records from PubMed; close to two million free full-text articles, reports, conference papers and patents from BioMed Central; preprint and reprint archive; and the U.S. Patent and Trademark Office databases.

If it is true, as claimed, that Kluwer publishes 50,000 articles annually, then its retrospective coverage must be rather short and/or the availability of the articles in digital format must be highly selective. Test results have shown that Kluwer's coverage goes back to 1984, but for the first eight years the digital coverage is negligible — not even a blip on the histogram I prepared. Although, since 1997 the digital coverage started to rise sharply. But even from the early 2000s, the records are only in the low 40,000s per year. For the earlier years of some journals, the lack of coverage may be explained by the fact that Kluwer acquired the journal only in the mid and late 1990s.

In spite of the relatively small sized archive of the world's second largest scientific-technical and medical publisher, KJO is a useful resource by virtue of their many respected journals. Nearly 300 of its journals are covered (with some overlap) in the Science edition and the Social Science edition of the Journal Citation Reports from Thomson ISI. Some of the journals are among the leading scholarly periodicals in the specific fields, such as Sex Roles, Archives of Sexual Behavior, Educational Psychology Review, Law and Human Behavior, Journal of Abnormal Child Psychology and Scientometrics.

Not surprisingly, Kluwer's journal collection is truly multidisciplinary, even at first glance. Clicking on a category often reveals subcategories and sub-subcategories under which journals are listed in Kluwer's catalog. With that said, some of the sub-subcategories are excessive, created for a single journal that may be listed under several other subcategories as well.

The diversity is apparent even with a cursory look at some of the results that show articles on highly specific topics, such as citation analysis, including journals one may not have thought of for this topic. The result's layout is visually pleasant and showing the journal cover page makes it more attractive. Still, it would be better to offer the option of displaying a tighter set page with the bibliographic information and relevance ranking of more records on the screen.

The abstracts are predominantly substantial and informative, as you would expect from scholarly journals. About 80% of the records have abstracts. Book reviews, letters to the editors, short obituaries and other editorial materials usually don't have abstracts, which explains the 20% gap to some extent.

It is, however, disappointing when substantial research articles do not have abstract, such as several important research articles in Scientometrics. It is an even more frustrating experience when the abstract button is not greyed out when there is no abstract for the record.

Although the full text of the articles are available only for the print subscribers and for pay-per-view users, I found quite a number of journals with many (temporarily) free articles. The above mentioned Scientometrics is one of those. This is good news because this excellent journal is inaccessible in many libraries that cannot afford to pay the more than $1,500 subscription fee, which was raised from the already very expensive $1,100 price before Kluwer acquired the journal from Elsevier. (The appallingly high subscription fee is not due to some exorbitant honorarium paid to editorial board members who would probably appreciate it if their existence and contribution were not ignored. It is quite an insult that the home page of the journal claims that there is no editorial board information and does not identify its editors who have been editing it since its inception.)

The really bad news is that the substandard implementation of the software makes much of the archive's content poorly accessible and some of it inaccessible.

The Software

Kluwer uses the expensive and capable Verity software, but perhaps it is more appropriate to say that it abuses the software. It is difficult to understand why the full text of a document is not searchable. These days the processing and storage costs associated with full-text processing is negligible for a company like Kluwer. Even publishers that are not exactly the most progressive ones in digitization, like the Haworth Press, offer full-text searching of articles.

It is just inexplicable that a search cannot be limited to a publication year range, document type, journal title or ISSN. The options available on the advance search menu are author, title, keyword and abstract. But this does not mean that you can find documents by using these options. For example, searching the title field for the article about the relational charting approach shown earlier finds no matching records. The same is true for a search for the word "Nobelists" (short for Nobel Prize winners) in the title field, whether you search in lowercase or uppercase. Dozens of known-items test searches yielded zero results, even though the records are there with the search terms in the titles.

Author searching is worse, as the help file does not provide any information about the syntax of the author index entries. I searched for articles by Tibor Braun, the editor in chief of and frequent contributor to the journal. Trying "Braun, T", "T Braun", "Braun Tibor" with and without truncation yielded no results. The query "Tibor Braun", which is an unusual format in abstracting/indexing databases that use the last name with the first name or initial sequence typically separated by a comma, finally yielded two hits. But there is no cause for celebration because his other eight contribution to this journal in the past five years were not found. You can find them if you browse issue by issue. Eugene Garfield had four contributions in the past four years for the Kluwer journal, but his last name brings up only one of them. This is unacceptable.

It might be nice for some users that they can highlight the hit terms in maroon or green or orange or mustard, bolded or plain, italicized or straight. Color highlighting indexing/abstracting records is not a big bonus, but more like the vibrating, pulsating neon signs of small Nevada casinos way off the Strip. Nor can I envision that users would like to choose the options of sorting by authors in Z to A sequence (instead of A to Z) or in reverse article title order, but good for them if they want to as Kluwer offers such options. I have the feeling that the programming was given to some gamers who grew up on MTV and Nintendo and are not much interested in scholarly journals.

The merger with Springer is likely to result in further price increases, which makes the pathetic implementation of the software even more appalling. One positive aspect of the merger may be that the digital Kluwer content management tasks will be outsourced to MetaPress, which has been offering much more powerful and intelligent search and browse options for Springer. That's my wedding wish for the couple (really for us, the users).