Reference Reviews

Péter's Digital Reference Shelf

April 2009

Title: ticTOCs
Publisher: University of Liverpool and the ticTOCs Consortium
URL: http://www.tictocs.ac.uk/
Cost: free
Tested: March 18-23, 2009

THE CONTEXT

In the era when many expect even traditional library and information services to talk, sing and dance, to show photographs of the authors and play video interviews with them, a Table of Contents (TOC) service may not be that exciting. It is for me, as I know from practice how much time one can save by using such a service. In my salad days, when I was in charge of an information and computer science library, an indexing/abstracting unit and a small software development unit (an unusual but extremely educative combination), the highlight of the week was when I thumbed through the most recent issues of the dozens of journals to see the latest developments in the hardware, software and “dataware” arenas.

As my managerial tasks and travels increased, the chance to do this decreased and I had to switch to using a print current awareness service based on the ISI Current Content database and a search query that reflected my specific topics of interest. Although it took away some of the pleasure and benefits of glancing at the pages of the journals, it still helped me to keep abreast of the journal literature.

With the coming of the Internet, getting e-mails from the publishers about the most current issues has become the main modus operandi for current awareness services. By now almost all the journal publishers provide free alert services through e-mail and so do most of the database aggregators. I did subscribe to several of them, but frankly, I often skipped opening those useful e-mails because I fell behind in my deadlines wading through the ever-increasing junk mail to find and read these important messages.

There are only a few functionally similar digital alternatives to what ticTOCs offers – and most of them are not free. The classic Current Contents database of the former Institute for Scientific Information, the mother of all TOCs, covers well more than 8,000 scholarly journals. It is available on some commercial hosts, such as Dialog, Ovid and of course, on Web of Science, in different subsets and in combination with Current Web Contents on Current Content Connect.

The Zetoc service —based on the table of contents records provided by the British Library — is maintained and operated by MIMAS, the National Data Center of the UK Higher and Further Education administration. It is free, but only for students, staff and teaching and research faculty at qualifying institutions within the U.K. It is about twice as large as ticTOCs in terms of the serials covered. I used the term serials because Zetoc includes TOC records for both journals and conference proceedings, but it does not have abstracts, while in ticTOCS about half of the records have abstracts.

There are some discipline-specific, free TOC services, such as the one in astronomy and astrophysics from the ADS (Astrophysics Data System). It covers 25 discipline-specific journals, plus Science and Nature. It is an additional free service of the excellent digital pre-print depository.

The closest free service to ticTOCs is the splendid Feed Navigator of the National Library of Health Sciences (Terkko) in Finland, which covers about 4,300 sources. However, it is limited to health sciences – and luckily, to library and information science! As the number of sources also includes some highly respected media sites (Reuters Health, BBC Health News, etc.) and good sites of the blogosphere (Virology Blog, EmergiBlogs), the extent of coverage is not really comparable.

THE CONTENT

This open access TOC service hosted by the University of Liverpool has highly relevant and current content. It is the result of a two-year project supported by the Joint Information Systems Committee(JISC), which is a good sign, as JISC has been sponsoring practical and innovative projects for a long time.

Size

The homepage says that there are TOCs from 12,135 scholarly journals from 433 publishers. Apparently, since the time the homepage was made, there has been a 20% increase in the sources covered. How do I know? I downloaded the file with the list of journals and saw the titles of 14,148 serial publications. There are gaps in the id-number of the journals, possibly for reasons of title changes, such as many of those with IEEE as a lead term in the title that were recently changed to IET and hence don’t have current articles under the former title. The number I referred to above (14,148) includes only the journals currently published.

This is quite a feat and obviously the University of Liverpool needed partners for this –and partners it got. It is an interesting and impressive mix of academic and commercial entities, including the Heriot-Watt University, CrossRef, ProQuest, Emerald, RefWorks, MIMAS, Cranfield University, Institute of Physics, SAGE Publishers, Inderscience Publishers, DOAJ (Directory of Open Access Journals), Open J-Gate and Intute.

True, Ulrich’s Periodical Directory provides information about 25,500 active, academic/scholarly journals, that is appropriate for bibliographic control perspectives, but not necessarily for benchmarking the size of the source for a TOC service. In addition, many of the periodicals are not journals and some of the journals don’t have digital content on the Web, let alone RSS service, so the size of the source base of ticTOCs, is very good although there are regretful omissions, apparently beyond the control of the management of the ticTOCs service

The number of records (with links, but not necessarily with access to the full text articles) is reported to be 344,713. This may not seem impressive at first, considering that about 1 million scholarly articles have been published recently per year – based on my statistics from searching Web of Science. Scopus has somewhat higher data for the most recent years, but its numbers include a higher ratio of conference papers, books and articles from trade journals (which I don’t mind, because some trade journals have better articles from better research than some scholarly journals – even if this is heresy from a university professor).

However, it must be realized that ticTOCs is a current awareness service, not a historical archive. It does not aggregate the records for the back issues of journals, (at least not for the public), therefore the size of the database reflects the number of records for the recently published papers and for the ones in the queue of papers that were accepted and are to be published.

When you consider that at the end of March Scopus had 267,417 records for papers published or to be published in 2009 (i.e. including the preliminary records for 53,617 articles in press, which is an excellent feature), then all of a sudden, the 344,713 records in ticTOCs gets very impressive, especially for the purpose of a current awareness service.

Currency

There is no way in ticTOCs to limit the search to publication year, but scanning many result lists, it appears that the majority of the records are for current issues, i.e. the ones published in or to be published in 2009, or for recent issues, published in the last quarter of 2008. If a publisher is belated it is not ticTOCS’s fault that the TOC record for the most current issue is several months old.

Fortunately, Haworth Press, whose belatedness in publishing could be measured in years rather than weeks or months for some journals, was acquired by Taylor & Francis a good year ago. It is not under the former management team, so the pages of a TOC list of former Haworth Press journals are not spoiled with entries of appallingly belated issues. These would have been technically the most current issues from that dysfunctional former publisher, but not from the perspective of the researchers who worked hard on their papers and the ones who would have loved to read those papers when they were much more fresh. (I must add that I was not affected, just irritated by the attitude of that publisher and its pathetic software. For more details, see my earlier column in ONLINE magazine.)

To be realistic, the majority of indexing/abstracting services have an average lag time of 4-6 months from the time the issues are published until records about their articles appear in the database – because indexing and abstracting is a time consuming process.

There is no such time lag in ticTOCs because it does not do abstracting and human indexing, it takes the bibliographic records and the abstract (when available) from the publishers directly, “as is” through RSS feeds. No wonder that there are many highly current TOCs in ticTOCs. For example, as I was finishing this column, there were already 20 records in it for the articles in the upcoming issue of Scientometrics.

It is far more efficient to find information about upcoming articles through ticTOCs than through jumping from the Early View section of Wiley, to the Online First section of Springer, then to the equivalent sections of other publishers’ sites, one after the other. A quick scan of the TOCs of your preferred journals stored in MyTOCs saves a lot of time and provides a lot of gratification.

Scope of coverage

The journals are classified by major subject categories and sub-categories as assigned in the Ulrich’s database. For example, there are 59 journals under Library and Information Sciences and 38 under Information Science and Information Theory (a subcategory of the Computers major category). This is a good number right at first sight and the list of journals confirmed the first impression by showing the most important subscription-based and open access journals from the perspective of library and information science and technology — with a few exceptions.

From the perspective of my personal interest, there were only a few scholarly serials that were absent, such as Library Trends, College & Research Libraries and Annual Review of Information Science & Technology (ARIS&T) which are essential publications for LIS practitioners, faculty and students. On a broader scale, the entire lack of coverage of the journals of ACM (Association for Computing Machinery) was the biggest disappointment. I missed some good trade publications but ticTOCs makes it clear that its focus is the world of scholarly/academic journals. It is a consolation for missing journals that I found TOCs for First Monday, Ariadne and D-Lib Magazine, excellent open access LIS journals.

I understand the reason – unhappily— that publishers of the missing journals may not have RSS feeds for TOC and/or don’t cooperate with ticTOCs. In either case, it is as much their loss as ours simply because ticTOCs records would be much more effective and much less expensive for PR than glossy ads.

The case of ARIS&T is somewhat different. The formal publisher, Wiley, is not aware of the fact that its information about ARIS&T is already two years outdated, not reflecting the 2008 and the 2009 editions. Being well aware and critical of Wiley’s mediocre software features (except for the good links to the items in cited reference lists), this does not surprise me. Fortunately, you can look up the TOCs of the 2008 and 2009 issues of ARIS&T at the Web site of the real publisher, Information Today, Inc.

I did a few tests by looking up publisher names and hit counts in tocTOCs. The number of hits in ticTOCs from the largest publishers suggests that their coverage is comprehensive. There are chances for exception as described for Wiley, but these seem to be really exceptional. I got hit numbers by publisher names that rhymed well with the number I had from a recent, unrelated research where I needed to know the collection size in terms of active serials by publishers.

Here are a few stats with the number of serials following the publisher name in parentheses: Elsevier (2,202), Springer (1,651), Taylor & Francis (1,335), Wiley-Blackwell (1,276), InderScience (255), Sage (250), Emerald (170).

Very importantly, the rich content of ticTOCs is free of charge to anyone. You don’t even have to register for the latter, unless you want to have its special features, like having a list of preferred journals selected and stored, which obviously requires some kind of authentication for retrieval, like a user-id and password combination.

THE SOFTWARE

Overall, I had good impressions about the interface, the navigation, browsing, personalization and output features, which are summarized below. These are followed by some suggestions regarding searching and linking, where I see a need for enhancements and corrections.

Interface and navigation

The interface is very simple, self-explanatory and adjustable. It is like a triptych, the three-panel altar paintings that can be made a diptych or even a single panel screen. The left panel offers a small box to search by title, subject or publisher. It is actually character string scanning, so there is no need to type in library or know if the title includes the singular or the plural format, or one of the variants: libri, librarian, librarians, librarianship – all these will be picked up without truncation symbol. Of course, there may be some cuckoo’s egg, like the journal Fluid Phase Equilibria, but it is a very small price for the convenience.

The intelligent design accommodates exact match and allows to limit the look-up process to lead terms or character strings. Either comes handy when you need to locate single word journal titles like "Science", which appears as the only hit when the exact option is chosen and as the first entry when the “start with” option is chosen — ahead of the other 28 that start with the string "science". This is smart.

The middle panel shows the table of contents of the journal selected on the left panel, in short hit list or detailed format . On the top of the panel the essential bibliographic data about the journal issue are provided. More about this later.

The third panel shows the list of journals that you stored in MyTOCs. This is excellent as you can quickly move from one journal to the other and there is no need to select them again in each session, as long as you have registered and sign in. It is very much worth the registration and sign-in. Strangely, my browser (Firefox 3.1) here did not offer to remember my user-id and password.

The journals are added to MyTOCs as you select them when browsing in the left panel and adding them. They remain tagged in the list on the left panel when you sign in later and can easily scan the list and decide to add a few more journals, or when you look up another category, see the journals that you have already selected – in those cases when a journal is assigned to more than one category/subcategory. Removal is equally easy.

In a generous and intelligent move, ticTOCs offers the option to export your feeds to other feed readers in a fairly standard OPML exchange format. I appreciate this because I still vividly remember from my early days in library automation projects, that the vendors held us captive using a proprietary record format without offering the option to convert them into the MARC communications format, needed to switch from one system to another.

By the way of MARC format, I wish there were an international standard for TOC records and publishers would comply with it, to facilitate the display of TOC records in a consistent, consolidated way, without the need to manually trim and tag them. I could see problematic records in ticTOCs and it was obviously due to unusual format used in the feed and that strongly tempers my criticism.

Record content and format

The result list displayed in TicTOCs is slim – but not in the skinny way of Twiggy, but in an appealing style. It includes only the titles of the papers. I find it good because that is the essence in this phase of the TOC lookup and it allows a quick scan. However, for some journals, like those of Emerald, the character string “Table of Contents” is also included with each article. This could be easily removed from the feed programmatically and it should be done because it is just fat and it kills the essence of the smart design of the quick-to-scan result page.

More details are available when the records are expanded. However, this area is overcrowded with unnecessary and redundant details, which should be pruned from the feed, not just taken as is. There is no reason to include the ISSN here, or add the authors’ detailed address as if we were to pay a personal visit, or send them a snail mail (instead of e-mailing them for a reprint as a last resort).

For fairness, the detailed records look much better if you close the left or the right panel of the triptych and make the screen a diptych with more space for the remaining two panels. You may go wild and make the middle panel the only one for most comfortable layout, akin to having all the three seats to yourself and without the equivalent of the fixed armrest, because the middle panel extends to the maximum width of the entire screen automatically. This is when you would appreciate the option to expand through a single click all the appealingly slim, pure muscle records on the hit list to see more details and as of now, unfortunately also some fat.

The title, authors and the abstract would suffice (although abstract is available only for about half of the papers). Smartly but unorthodoxly , there are no chronological and numerical designation data in the detailed record . It is obvious, that these are titles, authors and abstracts of the most current articles from the journal selected. It does not make the article more attractive if the records inform us that it appears, say, in volume 39, issue 5 and starts at page number 47. We are not in Kansas and in the print world anymore with ticTOCs.

I do the same in the digital reading list of my courses when there is a direct link to the article hidden behind the title, thus not wasting space and attention by adding volume, issue and page numbers which are irrelevant in this context. Still, I have to explain this in the first class to make sure that students don’t get the impression that this outsider does not even know how to create a bibliography correctly, applying all the very sophisticated rules for authors’ first and last name and middle initial formatting, data element sequencing, uppercasing and interpunctuation for which there are hundreds of format standards beyond Turabian, APA, etc. to the embarrassment of the publishing community and to the alienation of students, potential readers and writers. For them these insane variations in bibliography formats is as comprehensible as mass was in Latin wrapped in heavy accent. Requiescat in pacem, if I may say.

Searching

There are two things that I am critical about. One is the lack of searchability of the TOC records. I very much wish to see an option for searching the title and abstract fields for all the 300,000+ papers, to find information about the potentially most relevant new papers on my search topic.

Quite often I find information about highly relevant and pertinent papers in journals that I have never heard of, let alone added them to MYTOCs. I still don’t add them to MYTOCs in a neophyte’s state of enthusiasm because they are not likely to publish another paper on the topic anytime soon. Doing so would lose the first letter from SDI, the Selective Dissemination of Information, that used to be the technical term for the personalized current awareness service by topic.

A search option to find terms in the title and the abstract fields of 300,00+ records published in any of the journals ticTOCs would multiply the great value of TicTOCs big way. Such a search option is a real bliss in the excellent Feed Navigator of Terkko, the Finnish National Library of Health Sceinces mentioned earlier.

Author searching may be also useful, but it is not nearly as critical as title and abstract search, because I know that only a tiny proportion of the users would want to know if there has been anything published in the past few weeks by professor Jacso and by extension by researcher/professor John/Jane Doe.

Linking

My other recommendation is to find a solution to a link problem that I have run into and that may affect many other journals and users. The records for the upcoming issue Scientometrics that I mentioned earlier appear both in the Springerlink database (in the Online First section) of the publications of Springer and in the digital collection of its imprints, in this case that of the Akademiai Kiado (Academic Publishing House) in Budapest where the journal has been actually edited and published since its very first issue. This has an important consequence for those users whose library has full text access rights.

Why is this important? Because ticTOCs happens to link to the www.akademiai.com site for Scientometrics (which is technically and ethically correct and the full text of papers are indeed there). However, that site may not recognize your university library as a subscriber. It certainly does not recognize my university, so I am in a dead end street either through proxy searching or through IP address recognition from within the physical library.

The average users would not know about this scenario and following the link they just would assume that the library does not subscribe to the digital edition and they would need to pay for it. They can’t be expected to know that if they log in to the Springerlink database through proxy or from the library, the full text article would be there. Yes, they could inquire at the reference desk, but we know that few do and at 2 a.m. when they must have that article to meet the morning deadline for submitting a related paper, they can’t inquire.

This link problem may not be an exceptional situation because the largest publishers (Elsevier, Springer, Wiley, Taylor and Francis) kept acquiring the smaller publishers, laudably integrating their digital collections into Sciencedirect, SpringerLink, Wiley Interscience and InformaWorld.

However, if the links in TicTOCs are not to these sites but to the small publishers’ archives, then users may not be considered authenticated by those. This needs some solutions because this may apply to many other journals published by the smaller, formerly independent imprints of the largest publishers and may be available through two different Web sites, but authenticated only through one. It may be easier for the developers of ticTOCs to explore this issue, as they are likely to have a broader perspective as a team.

In spite of some deficiencies, ticTOCs is an excellent free service and the corrections and enhancements would really justify to apply for another grant to make it even more excellent in helping users keep abreast of the current journal literature more efficiently and in making better use of the journals subscribed to by the library.