_ADS Abstract Service: A Remarkable Eclectic Index_ In response to my recent posting for candidates for my new column in Library Hi Tech News tentively titled e-Profiles, Michael Kurtz, an astronomer and computer scientist with the Smithsonian Astrophysical Observatory [ http://cfa-www.harvard.edu/~kurtz/vita.html ] reminded me about the ADS Abstract Service of the Astrophysics Data System [ http://adsabs.harvard.edu/abstract_service.html [See below for edited reponse which is re-posted with his permission] Upon reading the 'What's New' [ http://adsabs.harvard.edu/abs_doc/whatsnew.html ] section (as well as Michael instructions below) I've concluded that this is perhaps the finest example of an Eclectic Index/Abstracting service that I know. I defined Eclectic in the context of e-journals in a recent posting: "By the "Eclectic Journal", I mean a Web-based resource that at its core provides access to e-journals that offer not only access to the conventional content of a digital form of a journal but also provides or permits interaction with novel and innovative _features and functionalities_ (e.g., reference linking, cross-publisher searching, page customization, open peer review, etc.) _AND_ novel and innovative _content_ (e.g., e-Books, pre-publication history, electronic discussions, translation services, e-prints, bibliographic databases, etc.)" [ http://sunsite.berkeley.edu/Web4Lib/archive/0101/0131.html ] Here's are some the features, functionalities, and content I find noted for in the ADS Abstract Service from its 'What's Next' page: * Citation indexing * Collaborative Filtering * Font, Format, and Display Control * Database Linking * Sorting * Data Extraction * Access to Dissertations and Theses * Personalization * FAX Delivery * Access to Conference proceedings [For examples of some of these features in e-journals, see my EJI(sm) registry [ http://www.public.iastate.edu/~CYBERSTACKS/EJI.htm ] I invite MyWebColeagues to explore this Most Remarkable Eclectic Index at their earliest convenience, if only to test its Collaborative Filtering functions and Citation Indexing feature described below. Regards, /Gerry McKiernan Remarkable Eclectic Librarian Iowa State University Ames IA 50011 gerrymck@iastate.edu **************************** I saw a note on the **** listserv from you requesting web resources for your column. Essentially every astronomer uses the NASA ADS Abstract Service http://adsabs.harvard.edu/abstract_service.html The average use by the average astronomer is more often than once per day. For a set of articles describing the system go to: http://adsabs.harvard.edu/pubs/A+AS/ For your column I suggest you look at our use of second order information retrieval functions. I believe they are unique in their IR retrieval power among literature search facilities. I will give you an example of how to use these functions. First go to our main query page: http://adsabs.harvard.edu/abstract_service.html. Next type "redshift survey" (use the quotes) into the Enter Text Words/Keywords window, then click the Send Query button. You now should have a list of the most recent 100 papers which contain the phrase "redshift survey" in the abstract, title, or keywords, ordered by publication date. Next page to the bottom of this list and click on the button which says: Get also-read lists for all articles in the above list. You now have a list of the 500 articles most read (in the last three months) by people who read articles from the first list, of recent articles on redshift surveys. The list is ordered by how frequently these articles were read, thus the list is, in order, those papers currently most read by people interested in recent work on redshift surveys. Now page to the bottom of this list and click on the button: Get reference lists for all articles in the above list. You now have a list of papers referenced by the 500 papers most read by people interested in redshift surveys, ranked by the number of times they were cited. A large number of these papers are not about redshift surveys, but they are the most generally useful for people interested in redshift surveys. A more specific list to usefulness to redshift surveys, as opposed to usefulness to people who are interested in redshift surveys, could have been gotten by clicking the return references button after getting the results from the first "redshift survey" query. Now (finally) go again to the bottom this list and click on the button which says: Get citation lists for all articles in the above list. You now have a list of those articles which cite the largest number of the articles in the previous list, sorted by the number of articles from that list which they reference. This is a list of articles with the most extensive discussions of topics of interest to people interested in redshift surveys. The high number of references found, both as a fraction of the papers in an article (49% of the papers referenced in Cole, et al were in the previous list, 57% from Cohen, et al), and in absolute terms (105 papers from the list were referenced in Ellis' review article, out of 237 in his reference list) demonstrates clearly that these second order query methods work very well. So finally, in response to a simple text query for "redshift survey" we have four lists of papers: 1. The most recent papers on redshift surveys (What's New?) 2. The most currently popular papers among people interested in redshift surveys (What's Hot?) 3. The most currently referenced papers by people interested in redshift surveys (What's Useful?) 4. The papers with the most citations to papers useful to people interested in redshift surveys (What's Instructional?) [snip] Cheers Michael Kurtz kurtz@cfa.harvard.edu 617-495-7434