[70] in Information Retrieval
adventures in libraries
daemon@ATHENA.MIT.EDU (Tom Owens)
Fri Mar 6 09:08:22 1992
To: elibdev@MIT.EDU
Date: Fri, 06 Mar 92 09:06:53 EST
From: Tom Owens <owens@Athena.MIT.EDU>
Regarding Bill's trip into the libraries:
>I sat down at one of the PC's in Barker, but the first one I tried,
>alleging to be a connection to GEAC was hung. The second one gave me
>thousands of entries on my keyword DISPLAY. I found two mechanical
>engineering theses that almost looked relevant until I discovered they
>were for telesensory displays for MIT's robot submarine.
I assume you mean the title keyword index. As the caretaker of Barton,
I readily admit the keyword indexes, lacking boolean which we have no
resources for, are not very useful. With boolean, which Geac provides,
they'd be darned useful. (see below)
>What I learn from this is that, although Mercury is not well regarded,
>even by its developers, it has merit. The merit comes from having many
>databases available and being able to search on multiple keywords.
Well, I think it would be fair to say that bad access to data is better than
no access to data, at least for the sophisticated. It is a tautology, and
therfore essentially empty, to say good access to data is better than bad
access to data, but, hell I say five empty things a day, so this is a
healthy start.
Keyword searching is
incredibly powerful, but the naive have trouble understanding it and I
strongly believe it should be added in addition to, not instead of, the kind
of phrase searching Barton provides.
All the studies, pre- and post-automated library catalogues, tell us that
80% of library searches is for a "known item." The easiest and fastest access
to a known item is through phrase searching. In addition, phrase searching
tends to provide user feedback which allows the user to adjust his "system
concept." A naive user doesn't get much feedback from "S1 506 entries"
but quite a bit from
"Display Systems, Aeronautic
Display Systems, Commercial
Display Systems, Information
Display Systems, etc"
>Now all I have to do is find out how to get access to the actual texts:
>CHI + GI 1987 Conference: Human Factors in Computing Systems
> and Graphics Interface; Toronto, Ont., Canada; 5-9 April 1987;
>[I think MIT has this one somewhere]
>Proceedings of the Human Factors Society 31st Annual Meeting:
> Rising to New Heights with Technology; New York, NY, USA; 19-
> 23 Oct. 1987;
>Proc. S.I.D. (USA); vol.30, no.1; 1989; pp. 23-8
>[I wonder what S.I.D is...]
Well, authorities on CD-Barton tells me that SID is the Society for Information
Display. I found, I believe, all these sources pretty quickly on CD-Barton.
Proceedings of the SID Technical Session is at the Barker Library. Volume
10 is 1969, so volume 30 1989 makes it seem likely this is the guy you want.
Human Factors Society, Proceedings of the Annual Meeting, is at Barker,
call number TA166.H8, from 1972 on.
CHI + GI Conference proceedings, 1987 & continued, is at Barker, call number
T385.M36.
Without having your citation, I can't be sure these are the guys you want,
but they look likely.
>Stay tuned for "An odyssey through interlibrary loan" :-)
With any luck, you won't have to do that.
--
Tom Owens
MIT Library Systems Office
617-253-1618
owens@athena.mit.edu
