DLI social science team

DLI Social Science Team Home Page

send comments or questions to: l-neuma1@uiuc.edu

DLI Undergrad Focus Group Summary

Bishop, Ignacio, Neumann, Star 1/5/95

Note: This summary was prepared by writing up Bishop's notes and checking them against the summaries prepared by Ignacio and Neumann to achieve completeness and consistency. All individual comments recorded in writing during the focus group are included in this summary. Comments reflect the language of the participants as much as possible. Notes in square brackets reflect researchers' thoughts.

Date: 11/30/94, 7:00-9:00 p.m.

Attendees:

There were two male and two female participants; three students were in Computer Science, one in Physics.

Moderator comments:

Discussion began late, due to late arrivals. More direct questions were required to stimulate discussion. Students were very uneven in their information "sophistication," or at least in the verbal expression of basic terms and concepts (e.g., one participant did not seem able to distinguish between books and journals while another tossed out the term "thesaurus"). Nonetheless, they clearly articulated their preferences and behaviors related to information seeking and use. They did not seem particularly interested in the DL project until the end of the discussion, when they really got into their wishlist for system features. It seemed easier for them to imagine revolutionary changes than fairly simple ones. For example, they said the computer should be able to "just read your mind" so you don't have to worry about using the right search terms yet they seemed astounded at the suggestion of fulltext display of articles. Convenience seemed uppermost in their minds as what the DL could offer.

Purpose of Journal Use

--Get more up-to-date info on topic in rapidly changing field from a credible source; you know the authors are respected
--Up-to-date, but language is hard. Journals help me figure out how to apply textbook stuff, present real-world applications; are more interesting than texts and present different voices
--Provide more in-depth info than texts
--Use journals to write papers on what interests us in the field, to identify possible career options
--For research, use journals to find out what others are doing, get ideas
--Don't use to answer specific questions; if you have a question, you ask your TA and they answer... don't send you to journals
--No time to browse journals just for general interest or enjoyment; might browse a little if trying to narrow topic
--When browsing for info on programming languages might find one reference and browse the area [is searching for material on programming languages different or does this just reflect an individual difference?]
--Don't browse because can't get into stacks in main library
--One time I just browsed available issues of a journal and found something neat
--Rarely use journals; courses have textbooks and how-to manuals for computer systems and programs
--Had some visiting research profs in a course once and thought about using journals to look up relevant lit (by that prof, or in that research area)
--Use journals more often for courses outside engineering Nature of Journal Use
--Read beginning of article; use abstracts in online system
--Start with InfoTrac to see brief summary
--Look at first paragraph; sometimes the titles don't make sense and abstracts help, but I usually still don't know if I need the paper from the abstract
--Start with intro to get idea of what paper is about, then skim or jump to particular section.
--Skim if looking for ideas, supporting statement. First paragraph gives sense of what it's about.
--Refs helpful, but library doesn't have all of the sources cited
--ToC and index helpful for longer documents [I think he really meant books]
--Simple pictures catch my eye and prompt me to read the article
--Some tables get me to read the article; words are boring, but pictures get my interest

Problems

--Library just doesn't have the journals I need
--Journals are scattered all over campus
--Article I need is always ripped out
--Online collections are too vast; we don't have all that stuff
--You have to phrase things just right or you won't get what you want
--Always a wait at the InfoTrac terminals... so you can't interactively search, get and read articles, search some more
--Profs won't let us write papers from the abstracts alone
--I'd take local, irrelevant article over remote, relevant one
--Never have time for interlibrary loan
--If I need 5 cites, I make sure I get a list of about 20

[In general, students seemed overwhelmed by library system and have yet to develop "tacit" knowledge: don't know differences between databases, how to find other resources on campus, how to narrow search.]

Other Info Needs

--I mostly use books; go to that place on the shelf to find more and also use the index to get info
--I use books on programming languages. Journals offer articles on latest developments though, like Mosaic. Also 'tricks of the trade,' and what's coming in the future
--I use software reference manuals
--I'm unsure of my major, so I'd like autobiographies of physcists
--I read NCSA, CS alumni newsletters to see what grads are doing, find out about interesting stuff on campus to help in career choice. Also need outside context to see if what my prof is saying is really important [Others didn't seem so keen on newsletters. Some use email to keep up with departmental info. Several used Mosaic.]
--I want a book that actually tells me HOW to write C++
--Easy to feel lost, don't know purpose of what you're studying
--Skim newsgroup headings, but don't get much out of them [so why use?]
--Ref books are obsolete
--Ref books can be just as confusing, though you need common knowledge
--Ref books not detailed enough for papers
--Use computer use cheat sheets; quick guide to what key to push
--Don't scan shelves because don't have stacks access. If looking for starting point, this doesn't work
--I get to one book I'm after and take it; don't look around the shelf
--In bound journals, sometimes scan nearby issues
--[Personal collection?] Yes, books for class and enjoyment. Keep old texts in case needed later or to help friend
--Keep different reference pamphlets, like using email
--Profs give out ready-made notes; would be good if library kept a copy in case you lose yours
--Have CS folder for info on programs, class notes, lab notes (e.g., instructions for using Kermit)
--Class notes from prof very helpful
--Trouble finding things? I'm very organized [other students jumped in saying they WEREN'T]
--"Electronic slides" related to class available sometimes
--Use some stuff in Novanet

Ideal Digital Library Features

--All the books actually there
--Access to all material from one screen; not separate systems for separate stuff
--Source to explain subject matter of journal [explaining tacit knowledge--prof who knew which Physical Review source held what]
--Should include course notes on reserve
--Graphics and touch screen
--User friendly/idiot proof
--I'd be afraid to touch such a great system; it'd be intimidating
--Want to search for my term in all journals and books
--Special little libraries for certain topics: CS->programming languages->C++->then shelf or section where you can get it
--Multiple copies of heavily used items
--Don't want to have to know commands; you shouldn't need any prior knowledge of the system
--Help balloons; online help very important; shouldn't have to leave terminal to get help
--Want list of quick tips next to terminals for searching, e.g., which function keys do what.
--Lots of graphics
--An index: general to subtopic to subtopic to what you want; shouldn't have to type words
--If you have to use words, have the thesaurus right there... I spend lots of time asking everyone in the library "do you know another word for this?"
--Should be automatic... you type in word and it suggests related words
--Yeah, my last search was hard. I typed in herpes and noticed that the article was labelled herpes simplex, so I tried that
--Need an acronym list... don't even know what AIDS stands for [general agreement]
--You should be able to type in a subject and it takes into account all other possible words and pulls up those, too, in all articles
--What if you get 400 hits? Look at first couple and get ideas how to refine [i.e., build in examples of how to narrow/expand searches
--Want to skip around in numbers... just get tired of going in order
--Window that presents "another index" that separates out the most relevant and shows which related to which topics
--Should be able to add and take away sources on the fly [recombining booleans]; e.g., Asia and certain Prime Minister, certain date
--Ease of access from home
--Prompt for local, national, worldwide location of info [assuming that fulltext not available online]
--Don't just limit search terms to subjects; also dates
--Want lots of ways to access the same article because everyone doesn't think the same way; aside from topic, might want to access by where and when it happened, which people were involved
--Be able to search by topic, then get ToC
--If there really were a DL, I'd be using it a lot.... it can be scary; I don't want to be in a box, surrounded by gloominess. Make the physical site homey.
--Need to be able to make copies easily; everyone should have access to a printer, so you can print and read at your leisure
--Need good chairs and monitors
--Have plenty of terminals
--[What if you could actually get the article?] That'd be great! [general agreement]
--First you should pop up the title, who wrote it, index and abstract. Then, if you want it, pop up the full text. And then print, but not that kind of all inky printer
--Or download to disk
--Should pop up option for how you want to search and how display
--Zoom in on article, then back to rest of the set; working forward and back
--Search by age range, e.g., for paper on pregnancies
--Author not so important
--Be able to say "this," not "this" in one search
--Voice controlled input, touch screen would be ideal
--Paper better for making notes on the document