Chemical Information System (The CIS)

THE CIS NEWS	OCTOBER 2002

This issue of The CIS News is coming to you from the CIS E-mail address -- cissupport@nisc.com. Comments, questions, concerns, etc., should be directed to that address.

FULL-TEXT LINKING OUTSIDE CIS

The recent update to CIS introduced out-of- system full-text linking to CIS for the first time. This is different from the in-system full-text linking we've had before (for example, to TSCATS documents in the TSCATS database); that linking was -- and is -- to documents actually resident inside The CIS. The new linking is to documents -- journal articles, reports, etc. -- that are OUTSIDE CIS. You can simply click on a link in a CIS record, and -- presto! -- there's the full text of the article cited in the CIS record. There's terrific power in this -- but there's also a bit of complication because of the multiple parties that are involved. So there are some things you need to be aware of, and some things you need to do to make this work for you.

First, there are two ways to take advantage of CIS's new full-text linking capabilities:

If your library has signed up for an open URL service, please let us know which one. We can currently interface with: SFX (Ex-Libris), 1Cate (Openly Informatics), or LinkFinderPlus (Endeavor). If you have one of these services, then you will already have done most of the work needed to identify the electronic full texts to which you should have access. It merely remains for us to activate the link to your open URL service. Once this is done, live links to full-text articles that were never there in CIS records before will begin appearing.

If you do not have an open URL service, then you need to provide us with a list of your subscriptions that give you online access to electronic full texts. For now, these may be through the following aggregators/publishers: EBSCO HighWire Project MUSE IDEAL Swets-Blackwell JSTOR BioOne We will load information about your subscriptions into NISCLink, our own utility for managing this kind of information. Once this is done, you will begin seeing live links to full-text articles that were never present in CIS records before.

Here are a few warnings and caveats about this process.

If you subscribe to a journal that gives you some electronic access to the journal, but you do not go through one of the aggregators/publishers listed above, let us know. We are already involved in discussions with additional aggregators/publishers. However, it may be some time before we can set up a functional link directly to a small publisher's electronic site.

We have done our best to prevent "dead links" from showing up in CIS -- links which, when you click on them, lead you nowhere useful. For example, if you subscribe to Journal X, we do NOT insert a link into each CIS record that cites Journal X -- but ONLY into those records that cite articles that are actually available online. However, dead links WILL turn up occasionally. Let us know when this happens so that we can refine our processes and eliminate these wastes of time.

When you see a live link in a CIS record, make a note of the bibliographic citation before you click on the link. We link directly to individual articles whenever possible, but sometimes you can only get to your article by clicking to the year of publication, then a particular issue, then to your article -- and it's a nuisance to get into this process before realizing you have to go back to get the necessary information.

We would very much appreciate hearing from any CIS subscriber about your experience with this. Send an E-mail to cissupport@nisc.com with any comments or observations you may have.

FULL-TEXT LINKING INSIDE CIS -- TSCATS, AQUIRE

As noted above, the full-text linking that has been present in TSCATS for some time is completely different from full-text linking OUTSIDE the system. For TSCATS, all you need is a CIS subscription. You click on the link that says "Click Here to View the Original Document," and the image appears in a PDF box. (You need Adobe Acrobat Reader 5.0 on your PC to make this work, of course.) There are now approximately 1,400 images installed in the system. This is still only about 6% of the documents cited in TSCATS, but the percentage is growing, and the documents currently available tend to be the more recently indexed documents (i.e., the ones in which most people are interested). The new year will see the addition of some more images from the backfile -- shorter documents (which are better suited than long documents to online presentation) that are cited in both TSCATS and RTECS. (There will be a link from each database to the full-text document.)

The technique we use for presenting full-text documents in TSCATS has now been extended to AQUIRE as well. Some 10,000 AQUIRE records (out of about 200,000) now have full-text document images attached to them. This actually represents only thirty or so studies, since the same studies are cited again and again in AQUIRE, but more full-text images will be attached to AQUIRE in the near future, too. If you want to see how this looks in a real record, enter AQ-0005460 in the search box marked Record ID/Reference No. on the Advanced search screen. (The box is on the right-hand side, about halfway down.) This will call up one record that includes the "Click Here..." remark. The document itself is a little bit pencilled up (since it was used by the AQUIRE editors in creating the database), but it's perfectly legible.

NEW DATABASES, UPDATES

Four new databases were introduced in the course of the recent update: RISKLINE, EMIC, CRISP, and NEW JERSEY HAZARDOUS SUBSTANCE FACT SHEETS. Most CIS users will know the first three from the National Library of Medicine site, where they have been available for some time. The fourth is a collection of MSDS-like information sheets for some 1,300 individual substances. Detailed information on each of these new databases is available through the CIS home page at http://www.nisc.com/cis/.

Updates were recently made to the following databases: AQUIRE, CERCLIS, HSDB, RCRIS, TRI, TSCAINV, and TSCATS.

The "update" to the TRI database -- Toxics Release Inventory -- was particularly noteworthy. The database previously covered reporting years 1997-1999 on our system. With the latest update, however, we now have the complete array of years of TRI coverage, 1987-2000. This resulted from user comments that the full array of TRI data was not conveniently available anywhere. The database on CIS can now be searched for all years simultaneously or for individual years, as you please.

For those who like to restrict searches in AQUIRE to only those records added/updated, note that the last add/update date in the last version of the database on CIS was 14 February 2000. Consequently, if you want to restrict AQUIRE searches to only those records that have been added or changed since that date, enter ">20000214" in the search box marked "Add/Alter Date"; it's on the right-hand side toward the bottom on the Advanced search screen.

RTECS STATUS

By now users of CIS will know that it was MDL to whom we were coyly referring when we talked about RTECS in the March issue of The CIS News: We were waiting for MDL themselves to make the announcement that they had been selected by CDC/NIOSH to continue editorial development of RTECS in the future. They now have an update ready, we are in discussions with a view toward acquiring the rights to it, and we hope to be able to present it to our users shortly. Stay tuned.

AQUIRE DATA ANOMALY

An AQUIRE user recently pointed out to us a data anomaly that sent us to the AQUIRE editors for an explanation. It was several LC50 records that were identical -- except for the concentration values. That is, the species were the same, test conditions and duration the same, water conditions the same, even identical citations to the same source -- but DIFFERENT LC50 values. Which value was correct?

Our inquiry to the AQUIRE editors at EPA brought this explanation: In the past, it was for a while the practice to create two records when the nominal substance could exist in ionic form in solution. The reported concentration would then be apportioned between the two records based on atomic weight. However, both records would show the same substance as the subject of the test -- Sodium sulfate, in the case that brought this up. There would be nothing obvious in the records to indicate that this ionic apportionment had been done.

This practice was abandoned some time ago, and the AQUIRE editors now include information on multiple ions in a single record. There are only a small number of extremely old records that still reflect the misleading former practice. Consequently, while the editors would like to seek them out and change them to match current practice, it is not a high priority.

This means that you may run into some of these records from time to time. You can recognize them quite easily, fortunately: They present identical data from the same source -- EXCEPT that there are two apparently contradictory concentration values reported. In this rare situation, there's really nothing to be done but consult the original source document cited in the AQUIRE record.

- END -
View March 2002 Newsletter View October 2001 Newsletter

National Information Services Corporation NISC USA Wyman Towers, 3100 St. Paul Street, Baltimore, Maryland 21218 USA
Tel: +1 410 2430797	Fax: +1 410 2430982
cissupport@nisc.com
www.nisc.com/CIS
A Company in the Public Interest
Copyright � 2002 National Information Services Corporation. All Rights Reserved.