[net.text] Needed: Text retrieval system for LARGE book index

ir524@sdcc6.UUCP (ir524) (08/16/85)

I am looking for software for retrieving entries from a book index.
Ideally it would run on both mini and microcomputers. Public domain and
commercial packages are both of interest.

Our research project has published a series of 96 large books, which
were not individually indexed.  We are now halfway through the input
stage of a comprhensive index effort.  We expect the total volume to
exceed 500,000 entries.  There is a lot of interest in making the
resulting data files available to our research community for computer
searching.  The preferred model is to package the data for micros, but
if that proves infeasable we can provide access via remote access to a
large VAX system.

So far my knowledge of systems suitable for this application is limited
to text retrieval systems and traditional database systems.  Since each
entry in the index is essentially a 'document' from the text retrieval
point of view this might cause some problems.  Traditional database
systems I am familiar with are not well suited to variable length text
fields where essentially every word needs to be an access key.
 
Any suggestions would be appreciated.  Please call collect if that is
more convenient than email.

UUCP  ihnp4\   ----!sdcsvax!sdcc6!peterw
      akgua  \     Peter Woodbury
      decvax  /    Deep Sea Drilling Project (A031)
      dcdwest/     Scripps Inst. of Oceanography
      ucbvax/      UCSD, La Jolla CA 92093
      ------       619-452-3526