TrAC - Internet Column

To cite this article please refer to the printed edition of TrAC: Trends Anal. Chem., 16 (1997) 68

The North American Chapter of the
International Chemometrics Society:
A Society's Virtual Office

C.R. Mittermayr

Vienna, Austria
cmitter@fbch.tuwien.ac.at

A community which is not only small but also sparsely distributed needs people with strong commitment who keep it alive or it will perish. The North American Chapter of the International Chemometrics Society (NAmICS) is both small with <465 members (see Table 1) and sparsely distributed. Despite its name, it also has about 20% non North American members, mainly from Europe (with increasing tendency). Although the NAmICS does not lack strong proponents who keep the society going, the internet provides new means for the community to communicate, which eases the burden of these few by distributing it over many (although 'several' is more accurate).

Table 1:

Member statistics of the NAmICS

Nationality	%	Affiliation	%
USA	78	Industrial	46
Canada	5	Academic	38
Scandinavia	5	Gov. & National Labs	14
Netherlands	3	Consultants	2
Europe (rest)	8	Non-profit organizations	1
rest of the world	2

The sparse distribution of chemometricians has led to the ICS-list which, besides conferences, is the main "meeting place" for chemometricians, who are not necessarily members of the NamICS (only about 50% are from the USA and Canada). In contrast to the discussion list which is a dynamic, rapidly changing and evanescent forum for the exchange of ideas and information, there is also a need for a more pertinent medium which preserves the information.

The NAmICS homepage can be considered as a virtual secretary's office. The main feature of this office is that it is a self-service office that is also open to non-members. This virtual office has an archive of its history and statutes (hopefully once), distributes tutorials, has a database of references, a blackboard with the programs of conferences, newsletters and a file with the actual addresses of its members and experts (maybe in the future). It has also a toolbox (i.e. programs in the jargon of chemometricians) available for everyone. But it should be emphasized that most of the information is not provided by the NAmICS itself, which only provides the link to the information. To share information is probably one of the main ideals of the non commercial part of the internet.

In building the NAmICS website, it would have been easy to put this metaphor into a picture and use a clickable imagemap to access the desired information, but due to the often very slow connections the coloured hype has been sacrificed in favour of speed and accessibility for everyone. This is of importance since the NAmICS homepage is located in Austria.

A lot of chemometric-related links point to each other [1], but this can also be found in other areas [2]. Therefore I will subsequently point out some links and archives which exceed the average information content of a page and are maintained with individual efforts.

Despite some enthusiasm during the first International Chemometrics Internet Conference, InCINC'94 and the second, InCINC '96, only very few tutorials on chemometrics have been made available. The most elaborate tutorial covers nearly all areas of chemometrics and is located at the University of Akron (J.K. Hardy). The only drawback is that it uses GIF images instead of pure HTML text, resulting in a poor transfer rate (at least across the Atlantic). The second one gives a broad view on factor analysis (C. Lochmueller, Duke Univ.).

R. Bro has put about 1200 references to chemometric articles (incl. abstracts) on the net, which can be searched free of charge. But there are still many more users than contributors.

This individually-provided reference source is not only of use to those who cannot afford an on-line search in CAS (as e.g. many students) but also for researchers. Brown mentioned in his review on chemometrics [1] that a considerable amount of literature relevant to chemometrics is not found by on-line searching in CAS. So this database might be considered the result of a rated hand search. Another project by Bro is a chemometrics glossary, which tries to explain chemometrics terms.

Two other sites which also provide information to the chemometric literature are the site of Brown, who has put three chemometric review articles appearing in the journal Analytical Chemistry on the internet. But unfortunately he had to remove these again, due to copyright enforcement by the ACS. Elsevier Science provides a search option to its table of content and Wiley is considering a similar service. In my opinion, making papers available via the internet should be regarded as distributing reprints (as long as it is free of charge). A copyright transfer should not deprive an author of all rights to his own work.

Another valuable source of chemometrics information are the presentations at conferences, but generally the results are not publicly available, except to the attendants who get a book of abstracts, although proceedings are sometimes to be found in special issues of some journals (which can take about a year - optimistically - between submission and publication). This is a pity since at conferences the results are generally very recent. The NAmICS homepage tries to build an archive for conference programs. An excellent example of a conference homepage is that of the 6th International Conference 'Chemometrics in Analytical Chemistry' in Taragona, which has made all its abstracts accessible via the WWW.

The main tools for chemometricians are computer programs. Some statistical software packages allow the programming of macros, which can then be given away to other users. Since the author and webmaster of the NAmICS uses MATLAB, the first links made public were links to MATLAB toolboxes. Links to other packages from other vendors are supplied depending on the contribution from users and readers.

As an example, R. Shaffer's ChemoWeb provides not only MATLAB code for neural networks and genetic algorithms, but gives a nice overview of the different fields of chemometrics and a list of homepages of prominent chemometricians. Another example is the Three-way Chemometrics Toolbox by R. Bro.

My intention here is not to give a comprehensive list of chemometric-related links but rather my subjective choice of some interesting and informative links. A more extensive list may be found on the NAmICS homepage.

Finally, I want to note that making a homepage popular still heavily relies on the basic media of communication and the internet, i.e. visiting conferences and email, respectively. Figure 1 shows the server access statistics of the last two months. The first sharp peak originates from the announcement of the NAmICS homepage at a conference and the postings to several related newsgroups. The following constant access numbers increased twice by the announcements of a newsletter and a new page of MATLAB toolboxes. The last two postings were made only to the ICS list. Even after two months the homepage could not be found by the most popular search engines. The author is looking forward to discovering whether the addition of the NAmICS homepage to the internet search engines will increase the number of accesses.

Fig 1. Access Statistics from 26/08/96 to 04/11/96 for all NAmICS related pages. Overall Hits: 1,378 accesses.

References