[futurebasic] spelling checker?

Message: < previous - next > : Reply : Subscribe : Cleanse
Home   : September 2000 : Group Archive : Group : All Groups

From: lcs@... (Laurent SIEBENMANN)
Date: Thu, 21 Sep 2000 01:09:01 +0200 (MET DST)

Bill says:

 > This is ridiculous. My 32K "real world" test
 > file took 14 ticks on a G3/233.

You all were warned.  

Waverly, maybe you could adapt your and Bob's
solutions to my useful drag-and-drop design having
no 32K limit.

Let me point out that the industrial strength
version is nearly a spelling checker.  Indeed if we
start by building up a census for 100,000 pages of
well proofread books from Gutenberg, and next feed
in my last article, then the *new* entries my
article generates will be more-or-less the spelling
mistakes in my last article.  I think this is very
close to the design of early mainframe spelling
checkers. Note that our Macs are more powerful and
have more RAM than the mainframes that did such
spelling checks for me 10 years back.

When FB wins a silly test, it just looks silly. Only
when it wins in building useful and "enviable"
functionality, more programmers like you guys will
buy FB.

Cheers

Larry


PS. Waverly, you are still not being clear about the
distinctions between, hashing function, hash table,
and associative table.  Maybe when the dust settles
you could explain better...