home
***
CD-ROM
|
disk
|
FTP
|
other
***
search
/
Garbo
/
Garbo.cdr
/
mac
/
hypercrd
/
hc1_2_x
/
fretext1.sit
/
Welcome to Free Text v1.01
< prev
next >
Wrap
Text File
|
1990-04-13
|
4KB
|
104 lines
Greetings!
This is the master distribution disk for Free Text, the "real-time
high-bandwidth free-text information retrieval system". It contains:
Free Text Browser v1.01 - 65,497 bytes, created Sun, Sep 10, 1989,
modified Fri, Apr 13, 1990
Free Text Help/Services v1.01 - 180,714 bytes, created Sun, Jan 7, 1990,
modified Fri, Apr 13, 1990
This disk also includes a folder with backup copies of the Browser and
Help/Services stacks, a folder with the complete C source code for the
XFCNs used in the Free Text stacks, and this "Welcome to Free Text v1.01"
file itself.
Version 1.01 changes only three things from version 1.0:
- the Text View retrieval part of Free Text Browser now filters out
all control characters other than <return>, so highlighting for non-TEXT
files works more reliably;
- the source code for the Free Text external functions has been moved
out of the Help/Services stack and into a separate folder, for multiple
reasons;
- my 30kB essay on "Free Text Information Retrieval Philosophy",
written in March 1990, has been integrated into the Help/Services stack.
Please read the detailed documentation in Free Text Help/Services
for complete information about how to use Free Text most efficiently,
for the source code in C behind all the Free Text functions, for the
specific terms and conditions under which you can redistribute
Free Text (the GNU General Public License), for details of the lack
of warranty with which this software comes, etc.
I hope you enjoy using my programs, and that they serve you well!
Best,
Mark (^z) Zimmermann
Silver Spring, Maryland
February 1990
------------------------------------------------------
text of official Free Text release announcement follows:
FREE TEXT is a memory extender -- it lets you take multi-megabyte
collections of unstructured text and browse them rapidly and
effectively. It works by building a complete inverted index to every
word in the database, and then provides you with windows into the word
list (Index View), the instances of each word's occurrences (Context
View) and the full text around each instance (Text View). Free Text is
fast, both in indexing (I measured 17 Megabytes/hour on a IIcx recently,
indexing the King James Bible) and in searching/retrieving (limited by
HyperCard screen refresh delays). Index file overhead is reasonable,
approximately 80% for large files.
I've applied FREE TEXT to all sorts of big text collections -- archives
of my online correspondence, religious and literary works, technical
manuals (e.g., INSIDE MACINTOSH), etc. It's great for early stage
research, before one knows the precise questions to ask to a more
structured database. I describe FREE TEXT briefly as "the real-time
high-bandwidth free-text information retrieval toolkit".
FREE TEXT differs from more conventional inverted-index search systems
in its use of the point-and-click Mac user interface. It provides you
with word lists so that you can recognize what you want to retrieve and
click on it, rather than requiring you to remember and type in targets
for searches and retrieval commands. FREE TEXT is enhanced over my
earlier TEXAS and TEX systems in that the user interface is better, you
can now do full Boolean proximity search, multi-file browsing is easier,
and many less-visible improvements have been made behind the scenes, to
make FREE TEXT more powerful and extensible.
FREE TEXT is completely free software (in the GNU/Stallman sense). It
is written in C as external functions (XFCNs) behind a HyperCard front
end, which makes it easy to customize and modify as you please. FREE
TEXT includes full source code to all its XFCNs, along with extensive
help, a "guided tour" of its features, quick start summary information,
and other documentation and information to help users and developers.
Hope you enjoy using it! - ^z