[bionet.software] Browsing through terabytes.

HARPER%FINFUN@ICNUCEVM.CNUCE.CNR.IT ("Robert Harper Finland", CSC) (06/25/91)

   If you want to sell an idea who do you go to? To the big bosses
   who hold the purse strings, or the grass roots users who utilise
   the network. I think I prefer the latter.

   << Now we prepare the ground >> ( An introduction to WAIS )

   Excepts from BYTE May 1991.  "Browsing through terabytes"

   " WAIS is an engineering effort spearheaded by Thinking machines,
   the manufacture of the Connection Machine, a massively parrallel
   super computer."

   "The Libary of congress archives roughly 25 terabytes in its
   collection. To browse through this volume on your own would be
   nearly impossible"

   " WAIS can distill the contents of archives into neatly
   manageable and browsable folders"

   " With Relevance feedback WAIS can retreive documents with
   greater ease and speed "

   " The WAIS protocol also lets you use English-language-style
   query lexicon instead of cryptic SQL or forth generation
   languages"

   "The DowQUEST runs on Connection machine... has 1 gigabyte of
   text derived from over 400 sources... the search time with a
   hundred word query composed in typed english... is less than half
   a second."

   "The projections for the connection Machine system indicate that
   when it is scaled to a 1 terabyte database with a ten word query
   obtaining an answer within 10 sec or less is highly probable."

   Rob "I have seen the future and it is big" Harper