European Libraries and Electronic Resources
in Mathematical Sciences

Annex

"Zudem sei an dieser Stelle nochmals auf die Tatsache hingewiesen, daß Bibliothekare häufig an den Einstellungen und Bedürfnissen sowohl der aktiven als auch vor allem der Nicht-(Mehr)-Nutzer einer Bibliothek interessiert sind. Letztere können jedoch nicht in ausreichendem Maße von der Benutzer- oder Besucherbefragung erfaßt werden. die ebenfalls keine Gründe für die Nichtnutzung einer Bibliothek liefert. Diese gibt vielmehr Aufschluß über die soziodemographischen Merkmale, die Gewohnheiten und Einstellungen der aktiven Nutzer oder Besucher einer Bibliothek. `Daher ist zu empfehlen, die Repräsentativbefragung einer Nutzerinnenbefragung anzuschließen. ´(Kemper, 1992a, 48)."
Jaeger, Marcus: Planung einer empirischen Untersuchung in Bibliotheken am Beispiel des Projekts „Kundenbefragung in der Bibliotheksregion Ostwestfalen – Lippe“ Fachhochschule Köln – Fachbereich Bibliotheks- und Informationswesen (Hrsg.),
Kölner Arbeitspapiere zur Bibliotheks- und Informationswissenschaft Bd. 2,
Köln: Fachhochschule Köln, November 1997, p.13.
(cited Kemper from Jaeger: Christine Kemper: Befragungen: Marketing-Instumentarium zur Erhebung von Daten. In: Die effektive Bibliothek. Endbericht des Projektes "Anwendung und Erprobung einer Marketingkonzeption für Öffentliche Bibliotheken". Dbi-materialien Bd. 119, Berlin: Deutsches Bibliotheksinstitut, 1992.)

PART I OF THE QUESTIONNAIRE

1. Which of the Following Listings Describes Your Profession?

Five persons completed the field "Other" with the following answers:
"System Administration", "student", "Research Organization", "researcher" and "webmaster". Further one person completed "University" and not the field "Other" wrote in the text field to the questions point "other" "research institution". So we got six entries.

2. Which of the Following Describes Your Employer?

One person not completed the field "Other form of employer" but "Educational institution" and "Research institution" completed the text filed with: "Private University (Catholic Church)".

3. Are You Mathematician by Education?

7 persons completed the text field belongs to the statement "Yes" without clicked on it. They labelled their work field with
"Harmonic Analysis", "Statistics", "COMPLEX VARIABLES", "Geometry of Banach spaces", "analisi numerica", "mathematical analysis; differential operators", and "discrete geometry". One person who confirm to be a mathematician by education completed both text fields with "Stochastic Differential Equations / Large-scale systems" and "Applied mathematics". Another two persons completed the text field belongs to the statement "No" without clicked on it. They labelled their work field with "bibliothécaire" and "Theoretical physicist". Because of this 'wrong' completed forms we add 7 persons to the mathematicians and we can add 2 persons to the other group. So we get 90 mathematicians and 39 other persons in the group of 129 persons completed boxes belongs this question (the values of percentage are for mathematicians 70% and for the other group 30%). From the group answered with "Yes" six gave no field of work and from the other group answered with "No" two gave no statement of their educational background.
84 mathematicians specified their field of work.
The greatest group nine persons work on the field of numeric ("Differential Geometry, Numerical Analysis", "numerical analysis", "Numerical analysis", "Numerical Mathematics, Control Theory", "Numerical Analysis", "Numerical Analysis", "analisi numerica", "NUmerical Analysis", "Numerical Analysis").
Further answers were: "Algebraic Geometry", "algebraic geometry", "Algebraic Geometry", "algebraic geometry", "Operator Theory; Algebraic Geometry", "number theory", "Number Theory", "Number Theory", "Discrete Math, Combinatorics, Number Theory", "PDEs, functional analysis", "Harmonic Analysis", "Abstract Algebra", "computer algebra", "COMPLEX VARIABLES", "Algebra", " Ph.D , Algebra", "classical analysis", "spectral theory", "systems theory", "Dynamical systems", "Dynamical Systems + Teaching", "teaching", "mathematical analysis; differential operators", "topology", "General Topology", "set-theoretic topology", "theoretical computer science", "theoretical computer science", "applied mathematics", "Applied Mathematics (Differential Equations)", "computational/applied mathematics", "Differential equations", "Differential Equations", "Differential equations, dynamical systems, analysis, geometry etc," "Stochastic Differential Equations / Large-scale systems", "seismic", "Mathematical physics" "Mathematical Physics", "mathematical physics", "geometric mechanics", "applied harmonic analysis", "Harmonic Analysis", "algebra, logics", "PDEs logic etc", "logic", "logic, theory of computing", "logic & applications to language", "IT, linguistics", "applied probability", "Mathematical Optimization", "pure maths (algebra and combinatorics), optimization", "combinatorics", "Calculus of Variations", "operations research", "probability", "Optimization", "Optimization", "stochastic analysis", "University; Stochastics and OR", "Statistics", "geometry", "geometry", "geometry", "Geometry of Banach spaces", "Non Commutative Geometry", "computational geometry", "discrete geometry", "Differential Geometry", "Differential Geometry", "Ph.D. in differetial geomtry", "complex analysis, differential geometry", "field theories and differential geometry", "Technical Informatics, AI", "Information Retrieval" and "nor already specialised".

38 non-mathematicians declared their educational background: "Applied mathematics", "Computer science", "informatics", "computer science", "computer science", "computer science", "computer science", "Computer scientist", "physics", "physics degree", "Ph.D. in Physics", "theoretical physics", "Theoretical physicist", "Physiscs", "physics", "aeronautics", "technical university", "electronics", "engineering", "engineer", "Economist", "Dr. phil.", "historian of philosophy and related fields; librarian", "Philosophy", "HUMANITIES", "Humanities, Linguistics (Information Science)", "BTS SECRETARIAT TRILINGUE", "history", "Information Specialist", "Information system", "Electrical engineer and librarian", "physician and librarian", "Humanities and Library science", "M.S. in Library Science", "bibliothécaire", "HUMANITIES, BIBLIOTHECONOMICS", "librarianship", "Librarian (formerly radio operator!)". From the group answered with "Yes" six gave no field of work and from the other group answered with "No" two gave no statement of their educational background.

4. What Sort of Electronic Information System Do You Use Today? How Often Do You use These Different Forms?

General other online access
16 participants wrote to other online access:
ftp, telnet, email and news ("FTP, eMail, news, telnet", "email", 3 "ftp, telnet" (in different spellings), 2 "ftp, e-mail" (in different spellings), "telnet", "ftp", "news groups"), 1 "online databases (Dialog, etc.)", 2 university library ("Local network (University library system)", "university library"), "STN International", "LANL electronic preprint server (with subscription)", and "Zentralblatt & Mathematical Reviews".

CD-ROM access to bibliographic and reference databases
11 persons wrote to use the following CD-ROM :
"OVID", "university network", "ULTRA*NET", "Math. reviews", 2 Compendex ("Compendex, Science Citation index inside and many others", "Compendex plus, Science citation index, Iconda and many others"), 4 MathSci ("Mathsci", "MathSciNet", "math-sci", "MathSciDisc") and "ERL".

Electronic books
From the groups of librarians (26 persons answered in question 1 and/or 2 to work in a Library or to be an librarian) 7 answered with "3", 8 with "4", 7 with "5" and 4 gave no answer for the frequency of electronic books (Frequency 1/day, 2/week: 0 0%, 3/month: 7 (32%), 4/less: 8 (36%) and 5/never: 37 (33% of the group of librarians)). From the group of librarians 1 answered with "1", 3 with "2", 5 with "4", 9 with "5" and 8 gave no answer for the relevance of electronic books (Relevance 1/important: 1 (6%), 2/important: 3 (17%), 3/no opinion: 0 (0%), 4/unimportant: 5 (28%) and 5/not at all important: 9 (50% of the group of librarians)). This shows that the group of librarians labelled the relevance of electronic books no so high as the total group of participants.

5. Which WWW Browser Do You Use? Which Version?

The answers to the used browsers are:
1 "explorer", 4 "Internet Explorer", 2 "lynx & netscape4", "lynx, netscape", 3 "Microsoft Internet Explorer", 1 "Netscape (MS-Explorer at home)", 97 "Netscape" (in different spellings), 9 "netscape comminucator", 1 "netscape navigator or communicator", 2 "Netscape navigator" and 1 "yahoo",
37 of 132 persons give no version, two state that they do not know an done person said to use the last version.
Version: 3 persons answered with "lynx, netscape", "lynx & netscape4" "yahoo" gave no version. 25 persons use "Netscape" and did no statement to the used version. One person use Netscape version "2.0". 22 persons use Netscape version 3 (6 "3", 3 "3.0", 1 "3.0.4", 1 "3.01 Gold", 3 "3.01", 2 "3.04", 4 "3.1", 2 "3.x", 1 "F-3.01"). One person use Netscape version 3 and 4 and another person 2.01. and 3.00. 45 persons use Netscape version 4 (13 "4", 5 "4.0", 1 "4.02", 3 "4.03", 2 "4.04", 11 "4.05", 4 "4.06", 3 "4.1", 2 "4.5", 1 "4.x"), one person version 5 and one version "gold". 9 persons "Netscape Communicator" with once version 2 (1 "2"), once version 3 (1 "3.0") and seventh version 4 (1 "4.**", 1 "4.0", 1 "4.01 for Windows", 1 "4.03", 2 "4.04", 1 "4.5 PR1"). Of two persons answered "netscape navigator" one give the version 3. One person use "netscape navigator or communicator" with version "3.1". 8 persons "Internet Explorer": 3 persons version 3 (1 "3", 1 "3.0", 1 "3.01") and 5 persons version 4 (2 "4", 2 "4.0", 2 "4.01"). Another person use Netscape and also Explorer ("Netscape (MS-Explorer at home)" version "most recent").
See below under "Browsers" the actually used browsers.

PART II OF THE QUESTIONNAIRE

1. What Use Would You Make of a One-Stop Information System as Described in the Introduction?

5 persons filled "1" by the part "other" and they filled the text field with following answers:
"Finding publications that refer to a given one", "development interfaces, OPACs, ...", "access to fulltext scources of works by pre 1648 mathematicians [;-)]", "Information delivery for the users of the library", "search items connected to a specific problem".

2. What Category is Important for Your Search? What Would be Your Preferred Search Field?

4 persons filled "1" for the selection by other classification wrote to the text field:
"Computer", "Artificial Intelligence Subjects, Automated Facial Expression Analysis", "Full text of abstracts" and "? something to be developed". Also 4 persons filled "2" for the selection by other classification wrote to the text field: "DDC", "Lattice theory", "CDD" and "dewey".

11 persons, who filled "1" for other fields to be included, answered in the text field:
"combined title/keywords, perhaps abstract", "Conference title", "full text", "International Standard Number", "it is essential to have an analogue of the ``anywhere'' option in MathSciNet (I am not sure if this is covered by the ``keyword'' above, since ``keyword'' may refer to specifically selected keywords --- an important option too)",
"LABORATORY OR ADRESS", "language", "Refers to ...", "series", "subject headings", "words occurring in the title or abstract".
Further 5 text field answers: "a (rough) classification by type (survey, ...)" ("2"), "Citations" ("2"), "dept., institute" ("3"), "series/collection" ("3") and "Anywhere" ("5").

3. What Form of Search and Navigation Would You Prefer / Should be Selected?

9 persons answered for other browsing field:
"author & keyword" ("1"), "Journal Title" ("1"), "series" ("2"), "subject headings" ("2"), "Subject" ("2"), "Type of publications (Journal)" ("2"), "date " ("3"), "journals & series" ("3") and "grammatical stemmig of keywords" ("5").

4. What Information Sources Would Be Important For You?

(Only 14 persons is answer the first part of the question, 118 did not) 7 persons answered "1", one person "2" and one with "3" for other information sources gave the following answers:
"E-books", "Encyclopedie integral text", "Publishers' catalogues", 2 "reviews", "technical reports", "Thesis (PhD, diploma)", "citations" ("2") and "address directory" ("3").

7. Please Indicate Your preferences Regarding Result Representation

10 persons filled out the following suggestions:
" depend on the search field", "by classification", "I prefer to choose, it depends of the research", "Journal Title", "journal", "relevance (according to some - possibly configureable - ranking method)", "relevance: number of times the search condition is satisfied for the document", "selectable", "this depends!" and "user-specified".

10. If You Are in Favour of Mulitlingual Features, Where Do You Expect Them?

The text field other expected multilingual features was filled out by 2 participants:
"everything" and "documents should be multilingual!".

11. If You Are in Favor of Multilinual Features, In What Language Do You Expect Them?

Other expected languages are: "Latin", "Belgian", "altAltaVist" (funny answer without comment), "Russian, Chinese, Japanese, Korean, Arabic, ...", "Norwegian", twice "Russian" and twice "English".

12. If you use some of the existing search engines, which one do you prefer?

57 persons of 132 made no comments to the preferred search engines. The other answers are: 1 "alta vista, fireball, paperball", 1 "aLTA VISTA, YAHOO LYCOS", 1 "Altaviasta, SCI", 1 "altavista yahoo", 31 "Altavista" (in different spellings), 1 "AltaVista, MathSciNet", 1 "altavista,yahoo", 1 "www.altavista.digital.com", 1 "available engines in Netscape", 1 "google.stanford.edu", 1 "I use Altavista or Excite", 2 "Infoseek", 1 "Infoseek, Yahoo, Altavista", 1 "MATH-NET", 2 "MATH Database" (in different spelling), 1 "MathDataBase and MathSciNet", 1 "MathSci, Libris", 1 "MathSciNet and INSPEC", 7 "MathSciNet" (in different spelling), 1 "metacrawler/fermavista", 1 "no preference", 1 "noone", 1 "SavvySearch, ", 1 "sleuth", 1 "Webferret", 1 "What I get from my webmaster", 1 "YAHOO-ALTA VISTA", 5 "Yahoo", 3 "Zentralblatt" and 2 ("Zentralblatt/MATH" "e-math, ZBl").

Why?
78 persons did not answered the question. The other answers are: ", quick and relevant results of search", "1. Our department has access to it. 2. The choice of different kinds of display (html for viewing, ps for printing etc.) is very convenient. 3. The choice \"Global index contains the words\", to search for words occurring in title, abstract etc.", "Best match", "Clear Classification", "comparison of different search results is possible ", "comprehensive, good experience", "Cover my field of interest", "easily understandable GUI and easy to use way of displaying search results", "Easy access", "Easy to use, fast", "easy to use, good performances", "easy to use, high number of hits", "fast and accurate", "Fast and quite large", "Get rid of Internet garbage", "Gives very relevant and complete information", "Good menu and reach search features. Good possibilities to refine your search.", "Got used to; big number of resources", "habit? SCI due to its relevance Zentralblatt presentation is fine as it is (a good template)", "I am somewhat familiar with it", "I have got familiar with it - just that!", "I use a couple of engines, but don't know by head which is which.", "It's a meta search", "it's fast enough, and the first one that met my expectations", "It retrieves lots of relevant resources", "It seems simpler, because it presents the result of the search without trying to classify them or to select them. It is quick and easy to be used.", "It seems to produce more hits relevant to my query.", "It usually produces what I need in a short time", "It was the first search engine I found... And it is easy and decent. ", "I find library catalogues and bibliographical databases the things I need to access the most (I am still very much into PRINTED journals, and for preprints I rely mostly on ``old boys (and girls) network'' of transmitting information between the experts, plus a subscription some of LANL electronic preprint servers). MathSciNet has a wealth of information conveniently accessible, and the subscription price is NOT TOO HIGH (though it is not cheap).", "Very clean and reliable information, covering large area. It might well be used as a starting point, I consider it very good. What I miss in MathSciNet is: - document retrieval (it is now only very limited) - `inverted' citation search: finding those publications citing a given one, and sort them according to relevance (perhaps measured by a number of secondary publications considered relevant by the user)", "The both engines give me clear and comprehensive bibliographic information. The problem is that they are too slow, do not cover preprints, and do not allow me to get full size documents (my Institute has no subscription for document delivery services).", "most used to it", "My institution has a licence", "No particular reason.", "number of hits velocity", "Possibility to refine the search, free selection of matches to show (first 10 or 21-30, last 10, ...)", "powerful", "proximity operators plus comfortable boolean searches", "rapidity", "relevance of results", "Seems most complete", "Seems to have a high success rate, i.e. the first page of results typically points to the info I was looking for.Manuelk ", "Simplicity and a wide range of results. Although sometimes it is easier to use some kind of classification tree as e.g. in Yahoo.", "speed and wide range", "The pertinence of answers of each search engine", "very complete", "very extendable", "Very fast.", "What do you exactly mean by search engine? A mathematical search engine or any search engine?", "wide-ranging, reliable, fast", "YAHOO / POUR LES RECHERCHES SUR LES SITES FRANCAIS ALTA VISTA / BEAUCOUP PLUS DE REPONSES POUR LES SITES ETRANGERS", and "You can find most things with it, though not very selective."

13. What search engine has for you the most effective interface design?

87 empty answers, 3 with quotation marks, 2 with "No idea", and "none that I know". The other answers: 16 "AltaVista" (in different spellings), 1 "clear, flexibility", 1 "dogpile", 1 "duke e-print", 1 "Hotbot", 1 "http://ftpsearch.ntnu.no/", 1 "Infoseek", 1 "MATH Database (Zentralblatt)", 1 "Math reviews", 1 "MathDataBase and MathSciNet", 1 "MathSci", 4 "MathSciNet ", 2 "metacrawler" (in different spellings), 1 "noone", 1 "Northern Light", 1 "sleuth and gerhard", 1 "the above ", 1 "the same", 1 "xxx e-print archive (xxx.lanl.gov)", 1 "Zentralblatt is quite good. Math Reviews is customizable" and 1 "Zentralblatt".

What makes it most effective in your eyes?

102 persons did not answer the question. the other make this comments: "Choice of fields and of options", "Customizable search criteria.", "Early on, they showed how to insert HTML to _your_ homepage so that you could search from \"the comfort of your homepage\". That is truly excellent.", "Fast, no ads, fast.", "first a brief result of what I am looking for, then if they are relevant I would like to see details", "Gerhard"habit", "graphical refinement", "It is well designed", "It offers many possible type of searching easily to be choosed from the home page.", "It works intuitively.", "I can't tell. I think that the most important feature of a search engine is flexibility. There should be methods for browsing, quick search, extensive search (including the capability of a command language),... Perhaps the different layouts shoul different 'home pages'. The default should probably be a page for simple search containing in its upper part links to the more complicated layouts." , "That's a long story. Short answer is - everithing (the reason of such effectiveniss is that these engenes are based on strong traditions of Zentralblatt and Mathematical Reviews)." "Not ideal but see above.", "Only the information needed can be searched; possibility to get lists of the database", "searches different databases automatically", "See comment on question 12.", "simple + help for more advanced queries", "Simple menu and no preparatory selection needed", "Simple", "Simplicity and lack of advertisment. (At least on telia.com there is only one adv. image, I'm not sure how does US version look like)", "Simplicity", "simplicity", "simplicity", "the same", "True boolean searh possible", "used to it", "very good menu driven search facilities", "Very simple to use one-line search method." and "we can use the notion of phrase ".

14. Are there special functions that you want to see incorporated into the EULER engine?

105 person made no comments to the question. The other:

"? semantic proximity, if possible",
"a method to print the result in a convenient format (like ps in Zentralblatt/MATH)",
"As little `ANDs', `OR's, truncation and such technical approaches to searching, as possible. Nobody knows what those are.",
"Boolean operators and verbal research ",
"Contact info for authors when available.",
"Direct online ordering with different payment options (not only by credit card)",
"Facilities to reach a contact person in case one needs the full text or the paper version of a document one has found. E-mail address were the best link.",
"Geographical coperture concerning resources OPacs",
"Good online help, especially on Boolean operators",
"I believe that it is convenient to have a possibility to select several items from results of (possibly) several search procedures and to save or print out them.",
"1. The subscription price should be kept to a minimum. 2. Refined search capabilities on the output of the initial search are ESSENTIAL (a serious drawback of MathSciNet is their absence). 3. For bibliographical databases, it is important to have the information about translations, if any (often missing in MathSciNet and Math Reviews)." ,
"kind of mathematical taxonomy tree preferably better organized than AMS classification.",
"many links to other information provider/systems, with a short description of what one would have to expect from the other provider/system.",
"navigation with hypertext links (authors, keywords)in the results pages.",
"NO",
"No, but show me how to start my searches from my own homepage! (See Q13.)",
"No.",
"not loosing myself while search for information through web, unfortunately this is the case noadays",
"Overviews of Textbooks in Mathematics for Undergraduates, in particular new stuff. Listed by type of course e.g. \"Basic Linear Algebra\",
\"Mathematics for Biochemistry\",
(or whatever)",
"personnel electronic shelf "Please take into account the upcoming MathML standard for display of mathematics via the web. I believe MathML will greatly facilitate the display *and* content-cataloging of math on the web. It will eventually allow us to read mathematics on the web directly via browsers.", "possibility to get lists of the database; possibility to refine a search",
"See point 12"",
"Something similar to MathSciNet would be fine.",
"The above will be sufficient"
and "tree of categories, chains of information organised in something like tutorials (e.g. \"Everything you want to know about trigonometry)".