Tools for Information Gathering
Submitted by Francis Heylighen on Tue, 07/22/2008 - 17:24
Web of Science (including the Science Citation Index) is the most authoritative database in the scientific world, especially for its citation count for papers and authors. Compared to Scholar, however, its coverage is much more limited; it does not give direct access to the papers; and it only searches titles and keywords, while ignoring the content. You must access it via an institution that subscribes to the service, which is the case for the VUB.
Scirus was set up by Elsevier publishers as a competitor to the two previous databases. While it claims to be the most comprehensive scientific search engine, it seems to find much few papers than Scholar, though it is more comprehensive and userfriendly than Web of Science. Its subscription counterpart, Scopus, also includes citation data, but the VUB does not offer access to this database.
An important part of research is gathering existing information from papers and books. Thanks to the Internet, most of that information is now within easy reach. However, the number of information services on the web is overwhelming, and therefore it is worthwhile selecting only the best.
A good information collection tool is as exhaustive as possible in its covering of sources, but provides a good ordering so that you immediately see which results are the most important. Moreover a good tool is maximally simple and transparent in its use, avoiding any cluttering with irrelevant options, pictures, advertisements, etc. A strict adherence to this philosophy explains the phenomenal success of Google and Wikipedia—probably the two most important web services overall.
The following is a selection of tools based on the extensive (but of course still limited) experience of ECCO researchers. Using such tools may save you days if not months in the preparation of your projects, papers and dissertations.
Google is the best tool to quickly search the web as a whole, thanks to its very extensive coverage, clean, user-friendly interface, and smart algorithms to order the results so that the top entries indeed tend to be the most important ones.
Dogpile is a meta-search engine: it combines the results from several existing search engines, including Google, Yahoo and MSN. Because these search engines together cover a larger number of pages, this is useful when you are looking for something very specific that is hard to find. Otherwise, Google will give you a better ordering and presentation of results.
There are three large search engines for the scientific literature, with databases that only partially overlap. This means that if you are looking either for a comprehensive picture, or for something difficult to find, it may be necessary to use all three. Unfortunately, there is no clear method to find out what is included in each database, as these commercial services don't provide full information about their sources.
- Google Scholar is the first choice, as it covers a very wide range of material, and orders the result according to the number of citations and other relevant indicators of quality. Most usefully, it provides immediate access to the papers it finds, either freely for self-archived papers or preprints, with restricted access via your university library for the journals that the library subscribes to, or for payment to the publisher. Another useful feature is that each paper is linked to the papers that cite it, so that it is easy to find related material.
The VUB library allows you to search for papers in its holdings and has links to a number of more specialized databases. However, in general papers in journals held by the VUB can be found and accessed more quickly via Google Scholar.
Authormapper.com is a very useful database of articles published by Springer. It allows you to finds all papers that contain certain words. Moreover, it classifies those papers according to geographic location (on a world map), country, year, journal, author, institution, etc. Like that you can immediately see in which countries, journals or years terms like "cybernetics" or "memes" are used most frequently.
Google Books searches inside books, and allows you to read a few relevant pages of most books. It also provides links to libraries or booksellers (such as Amazon) where you can order the book.
Amazon.com, the largest bookseller in the world, has a very extensive catalog, and now also allows search of book content (not just titles and authors like other booksellers).
The VUB library lets you search title words or authors for books available at the VUB, or within the network of libraries in Belgium and the neighboring countries with which it has established agreements for "interlibrary lending". This means that if the book is not available at the VUB, you can order it via the database and collect it at the VUB after a short waiting period, provided you pay a small fee.
The following databases allow the rapid identification of scientific journals to publish your research in, or to browse relevant articles from.
Journal Info is an extensive database collected by Lund University, including open-access journals and journals listed in the Journal Citation Reports below. It lists (or provides links to) basic characteristics of the journal: home page, self-archiving policy, impact factor and other quality indicators, price, number of issues, etc. You can search for (combinations of) title words, including parts of words using "*"as wildcard operator (e.g. "complex* systems" will find both "Systems and Complexity" and "Journal of Complex Systems"), but not for more general keywords.
JournalSeek claims to be the largest database of freely available journal information, including description (aims and scope), home page, and subject category. The advantage compared to Journal Info is that it also searches for keywords within the "aims and scope" and thus finds more relevant journals. The disadvantage is that you get less characteristics at a glance.
Journal Citation Reports is the database used by the Web of Science, which means that the journals it includes are more authoritative (so-called "A1" publications), but much more restricted. It is most famous for its calculation of the "Impact Factor", which is traditionally seen as THE measure of the importance of the journal, in terms of the average number of citations it gets.
Zotero is an extremely useful "plug-in" for the free web browser Firefox. It can extract bibiographical data from practically all the important publication sites, such as Amazon, Web of Science and Google Scholar. The data are kept in its own database, which you can export in all the traditional bibliographical formats. Especially when combined with a Zotero plug-in for your word processor, this is extremely useful for creating bibliographies for papers and dissertations:
- using Firefox/Zotero search various websites for relevant publications
- mark the publications you want and have them imported automatically to your Zotero database
- when you need to make a list of references, select the right entries from the database, and paste them straight into your text in the required format!
CiteULike is a website where you can collaboratively create, store, search and publish bibiographies. It uses similar methods like Zotero to extract bibliographical data from common website formats.
OneLook searches hundreds of general and specialized dictionaries (such as our own Web Dictionary of Cybernetics and Systems) at once, and provides links to their results. This is particularly useful for rare, technical terms, such as "stigmergy", that may not appear in common dictionaries.
Dictionary.com similarly checks a (much smaller) number of dictionaries, but immediately shows you the most important results. This is useful for more common words where you want to compare the subtleties of the different meanings.
Wikipedia, the free encyclopedia to which everyone can contribute, is the best and most comprehensive way to get a good introduction to about any subject you can think of, including advanced scientific topics such as stigmergy, self-organization, or the Theory of Constraints.
Encyclopaedia Brittannica (freely searchable from the VUB) is the most comprehensive of the traditional, commercially published encyclopedias. It is smaller and less up-to-date than Wikipedia (e.g. it does not find "stigmergy"), but its results are more authoritative.
Google Translate is the best of several translation engines we tried out. It provides translation of webpages or inserted text between the main languages you are likely to encounter, such as English, Dutch, French, Spanish, etc. The main competitors are Babelfish and Systran. Anyway, you will still have quite some work to manually "clean up" the translations if you want to use them, but the work is least with Google Translate.