[comp.cog-eng] Issues regarding the storage of information

efam@nvuxr (elliott familant) (11/08/90)

I am interested in why users store various kinds of information.  Now
I'm not talking about the storage of information that is directly
related to some task they are working on, but information that they have
some discretion over whether to sre it or not.  There are those of us
who are information pack rats who keep things around (I hypothesize) in
case we ever need it.  Does anyone know of research that has
systematically looked at this behavior.  I am interested in why it
occurs and what effects it might have on the ablility of users to manage
the information they control.  My own literature search has so far
turned up zilch.  

Elliott Familant

eichmann@cs.wvu.wvnet.edu (David Eichmann) (11/09/90)

efam@nvuxr (elliott familant) writes:

>                                              There are those of us
>who are information pack rats who keep things around (I hypothesize) in
>case we ever need it.  Does anyone know of research that has
>systematically looked at this behavior.  I am interested in why it
>occurs and what effects it might have on the ablility of users to manage
>the information they control.

   Does 85mb make me a pack rat?  I've resorted to my own series of keyword
retrieval systems for old news articles, etc.

version 1 - news articles saved in a directory, retrieval by fgrep'ing
            (Slooow, and compound queries a drag)

version 2 - Hypercard stack (interesting, but compound queries a drag,
            plus reuse of embedded stuff even more of a drag)

version 3 - hierarchy of directories (generated by nn saves) with a cron-
            driven script that adds interesting words to an Ingres database.
            Retrieval is through a query program (lex/yacc) supporting
            compound queries and a variety of actions (display, print, etc.)
            (This is a real space hog - the database is as big as the file
            directory is, but I've got fast compound queries and wildcarding.)

This current version provides everything I want in retrieval of stuff from
a News archive hierarchy, but it doesn't address another major want -
retrieval of various code fragments from places like comp.sources.unix or
anonymous ftp sites.  

- Dave
===
David Eichmann
Dept. of Statistics and Computer Science
West Virginia University                  Phone: (304) 293-3607
Morgantown, WV  26506                     Email: eichmann@a.cs.wvu.wvnet.edu