Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!samsung!usc!zaphod.mps.ohio-state.edu!unix.cis.pitt.edu!pitt!willett!ForthNet
From: ForthNet@willett.pgh.pa.us (ForthNet articles from GEnie)
Newsgroups: comp.lang.forth
Subject: User Interface
Message-ID: <2191.UUL1.3#5129@willett.pgh.pa.us>
Date: 2 Jan 91 12:38:34 GMT
Organization: String, Scotch tape, and Paperclips.  (in Pgh, PA)
Lines: 76


 Date: 12-30-90 (13:02)              Number: 733 of 744 (Echo)
   To: GARY SMITH                    Refer#: 716
 From: RAY DUNCAN                      Read: NO
 Subj: USER INTERFACE                Status: PUBLIC MESSAGE
 Conf: FORTH (58)                 Read Type: GENERAL (+)

   Mitch Bradley writes:

   >I do know that ... I used a version of LMI Forth
   >that did not compile from text files...

 In this you are certainly correct, and I guess I misunderstood some
 aspects of your original message.  I did not notice that you were
 drawing a distinction between file-based Forth systems and Forth
 systems that could compile ordinary text files.

 All of our systems since the original Z-80 FORTH for CP/M 2.2 in 1980
 have been file-based, and have included a complete set of words that
 allow an application to open, create, read, and write any arbitrary
 file in the host file system.  However, for a very long time we did
 not support compilation of Forth source code from ordinary text
 files.  The interpreter/compiler always assumed that source code took
 the form of BLOCKs stored in a simple random-access text file, 1024
 bytes per record, and it always had a "current" block file which was
 the target of Forth commands such as BLOCK, FLUSH, and so on. The
 user could change the "current" block file with the USING command.

 The ability to compile Forth source code from conventional ASCII
 text files was not added to LMI systems until circa 1986 (might be a
 year earlier or later; I don't have version information handy).  We
 use the command INCLUDE for this type of compilation (although I
 don't claim that we thought up the term INCLUDE).  Our INCLUDE can
 compile from either block files or text files and supports nested
 compilation (the number of nesting levels is user configurable).

 I must say that although our systems have supported compilation from
 text files for some years, I still find myself using blocks in files
 99% of the time.  I like the flexibility and the "random-access"
 character of blocks, I like using my own full-screen block editor
 which I can modify at will instead of an external text editor, and I
 like the ability to put a simple BLOCK/BUFFER/UPDATE/SAVE-BUFFERS/
 EMPTY-BUFFERS/FLUSH interface in a ROM'd target system and use the PC
 as a BLOCK server over a serial link.   I'll be the first to admit
 that this is simply a matter of taste, and that if I invested the
 time to write a text file editor in Forth and make our text file
 compilation facilities more powerful (e.g. the ability to "mark" a
 piece of text within a file and then compile it selectively) I might
 find Forth code in text files a more friendly environment.

 Your point about the lack of standardization in Forth file interfaces
 is certainly well taken.  Our original file access word set was not
 very well thought out, and we didn't arrive at our current stdio-type
 interface until about 1986 after we had done a couple of custom ports
 to UNIX systems and saw the need for a file interface that would map
 easily onto DOS, UNIX, OS/2, Macintosh, Atari ST, and whatever else.
 I guess this was about the same time frame as you started promoting
 this concept at FORML etc. but since I don't attend Forth hobbyist
 events I didn't become aware of your work immediately.  (By this I do
 not mean to imply that you are only a Forth hobbyist, but only that
 the papers at FORML and similar events often show a shocking
 ignorance of what people are doing with Forth in commercial
 systems.)

 In another message, you mentioned that the Forth Vendors Group
 Floating Point Standard has not been heard of lately.  This is true
 but I believe (or at least hope) that the word set defined in that
 document is still widely used.  We use it, anyway.

 NET/Mail : LMI Forth Board, Los Angeles, CA (213) 306-3530
 <<<>>>
-----
This message came from GEnie via willett.  You cannot Reply to the author
using email.  Please post a follow-up article, or use any instructions
the author may have included (USMail addresses, telephone #, whatever).
Report problems to: dwp@willett.pgh.pa.us or uunet!willett!dwp