Writing FORTH systems from scratch
Alaric B. Williams
Has anyone ever written any good books on this? Especially for
multiprogramming multithreading multiprocessor systems (yikes...)
This would mean starting by writing a cross compiler in C or the like
to produce the kernel code, then rewriting the compiler in FORTH...
the idea is that FORTH is the starting software of the target system.
Insider's tips on how to implement tricky things like assemblers etc.
would be nice.
Thanx,
ABW
I have become... Comfortably numb...
Alaric B. Williams Internet : ala...@abwillms.demon.co.uk
Gordon Charlton
ala...@abwillms.demon.co.uk (Alaric B. Williams) wrote:
}Has anyone ever written any good books on this? Especially for
}multiprogramming multithreading multiprocessor systems (yikes...)
}
}This would mean starting by writing a cross compiler in C or the like
}to produce the kernel code,
Or in Forth, of course.
A Forth meta-compiler (so-called - the name is questionable) does very much
the same as the Forth command line interpreter, with the difference that it
compiles not into the dictionary, but into a separate dictionary
constructed for the purpose. Kernel code is similarly using a Forth
meta-assembler, again in many ways similar to a regular Forth assembler.
The code assembled need not be that of the host machine, nor need the
dictionary structure be the same as that of the host Forth.
Some complications arise from the target dictionary not being executable,
and also from the replication of names. Similarly named words may occur in
the host Forth, in the metacompiler and in the target Forth.
Once sufficient words in the target Forth have been defined it is a simple
matter to transfer the image to the target machine, and compile additional
words in the conventional manner.
If you have the time to figure out the details it is an interesting
exercise, if you do not you can probably throw money at the problem and buy
one.
(Elizabeth, Stephen et al - care to mention the names of your companies?
;-)
I have yet to see a good book on the subject, but then I haven't read them
all.
Gordon.
Chris Jakeman
gor...@charlton.demon.co.uk "Gordon Charlton" writes:
> In article <8219019...@abwillms.demon.co.uk>,
> ala...@abwillms.demon.co.uk (Alaric B. Williams) wrote:
>
> }"Writing Forth From Scratch"
> }Has anyone ever written any good books on this? Especially for
> }multiprogramming multithreading multiprocessor systems (yikes...)
> }
> }This would mean starting by writing a cross compiler in C or the like
> }to produce the kernel code,
>
> Or in Forth, of course.
<snip>
> I have yet to see a good book on the subject, but then I haven't read them
> all.
FIG doesn't list any books on either building a Forth from scratch or
meta-compiling (using one Forth to build another).
An index to Forth Conferences was published in the Forth Dimensions magazine
and lists:
Forth Metacompiling - Jerry Boutelle FORML 1980 12 pages
A Simple Metacompiler - Guy Grotke FORML 1984 16 pages
Metacompilers Revisited - Guy Kelly FORML 1991 17 pages
A starting point for meta-compiling might be Pygmy. This is "a lean,
fast Forth with full source code. Incl full-screen editor, assembler and
meta-compiler". It's public domain for DOS by Frank Sergeant.
A starting point for Forth with a tiny kernel written in Assembler or C
might be my own Minimal ANS Forth (MAF). This has been written
specifically as a starting-point for anyone trying to write an
ANS-compliant Forth from scratch.
It's portable to any environment as all the decisions about implementation
(eg stack size, dictionary mechanism etc.) are hidden in the kernel.
E-mail me if you're interested.
Bye for now _
_______________________| |_____
Chris Jakeman / _Forth_Interest_Group_| |____/
/ /_ __ ______ _ _ | | __
at Peterborough / __/ / / / __ / | | | | | |/ /
(a cathedral city / / / / / /_/ / | \_| | | <
80 miles north of London) /_/ /_/ /___ / \____| |_|\_\
Where do you come from? / /
______________/ / United Kingdom
Voice +44 (0)1733 346477 /_______________/ Chapter
Alaric B. Williams
>In article <8219019...@abwillms.demon.co.uk>,
>ala...@abwillms.demon.co.uk (Alaric B. Williams) wrote:
>}Has anyone ever written any good books on this? Especially for
>}multiprogramming multithreading multiprocessor systems (yikes...)
>}
>}This would mean starting by writing a cross compiler in C or the like
>}to produce the kernel code,
>Or in Forth, of course.
>A Forth meta-compiler (so-called - the name is questionable) does very much
>the same as the Forth command line interpreter, with the difference that it
>compiles not into the dictionary, but into a separate dictionary
>constructed for the purpose. Kernel code is similarly using a Forth
>meta-assembler, again in many ways similar to a regular Forth assembler.
>The code assembled need not be that of the host machine, nor need the
>dictionary structure be the same as that of the host Forth.
Exactly.
>Some complications arise from the target dictionary not being executable,
>and also from the replication of names. Similarly named words may occur in
>the host Forth, in the metacompiler and in the target Forth.
Make a more rigid interface between the two would suffice IMHO...
>Once sufficient words in the target Forth have been defined it is a simple
>matter to transfer the image to the target machine, and compile additional
>words in the conventional manner.
Exactly... which would be in the 'minimal' CODE wordset?? A PD set of
ANS words which build up from a set of minimal words would be a good
idea. Of course file & terminal access would be outside the basic set,
but they're very sysdep anyway. The basic set would be common and
basic stuff like @, !, lit, etc.
>If you have the time to figure out the details it is an interesting
>exercise, if you do not you can probably throw money at the problem and buy
>one.
Heh...
>(Elizabeth, Stephen et al - care to mention the names of your companies?
>;-)
As it is a complex but reasonably 'common' topic - like any compiler
writing, except that's not quite the right term here - some form of
FAQ would be a good idea!
>I have yet to see a good book on the subject, but then I haven't read them
>all.
It's a sort of underrated topic, eh? I guess everyone just makes their
stab at it :-)
>Gordon.
Chris Jakeman
ala...@abwillms.demon.co.uk "Alaric B. Williams" writes:
> gor...@charlton.demon.co.uk (Gordon Charlton) wrote:
<snip>
> >Once sufficient words in the target Forth have been defined it is a simple
> >matter to transfer the image to the target machine, and compile additional
> >words in the conventional manner.
>
> Exactly... which would be in the 'minimal' CODE wordset?? A PD set of
> ANS words which build up from a set of minimal words would be a good
> idea. Of course file & terminal access would be outside the basic set,
> but they're very sysdep anyway. The basic set would be common and
> basic stuff like @, !, lit, etc.
You've given a good definition of MAF - Minimal ANS Forth.
It requires a minimum of start-up code from its host, specifically:
- 2 stacks
- Memory for data and dictionary
- An interpreter to read Forth definitions from a file
- A minimum dictionary containing:
EXECUTE
! @ C! C@ MOVE
PICK ROLL DROP DEPTH
>R R> R@ RDepth
CELLS CHARS ALIGNED
Header
>BODY ' Parse-Find
IMMEDIATE COMPILE, POSTPONE
Branch!
<:> <Does> <Variable> <Constant>
EXIT <Branch> <0Branch> <Literal>
Nand + NEGATE DNEGATE
KEY EMIT CR
TIB >IN
(ANS words are the ones in uppercase)
(Other minimum sets are possible - this works for MAF.)
These 43 words are enough to build all of the ANS Core word set (133 words).
The host can be written in any language from assembler to FORTRAN and
run on any operating system that has a screen, keyboard and access to the
start-up file of Forth definitions. (MAF's original host was another ANS
Forth system.)
MAF is public domain, just e-mail me for a copy.
Brad Rodriguez
>In article <AD234FF4...@charlton.demon.co.uk>
> gor...@charlton.demon.co.uk "Gordon Charlton" writes:
>
>> ala...@abwillms.demon.co.uk (Alaric B. Williams) wrote:
>>
>> }Has anyone ever written any good books on this? Especially for
>> }multiprogramming multithreading multiprocessor systems (yikes...)
>> }
>> }This would mean starting by writing a cross compiler in C or the like
>> }to produce the kernel code,
>FIG doesn't list any books on either building a Forth from scratch or
>meta-compiling (using one Forth to build another).
>
>An index to Forth Conferences was published in the Forth Dimensions magazine
>and lists:
> Forth Metacompiling - Jerry Boutelle FORML 1980 12 pages
> A Simple Metacompiler - Guy Grotke FORML 1984 16 pages
> Metacompilers Revisited - Guy Kelly FORML 1991 17 pages
Oh, now I'm offended, Chris. :-) Did you not see my three-part series
"Fundamentals of Metacompilation" in Forth Dimensions? Or the eight-part
"Moving Forth" in The Computer Journal, where I talk about writing a
minimal ANS Forth from scratch for three different CPUs? Granted, even
as multipart magazine articles they're a bit terse...someday When I Have
The Time, I should expand them into a book.
I've been meaning to look at your minimal ANS Forth MAF and compare it to
my CamelForth. It sounds like we had similar objectives.
If you go far enough back in Forth Dimensions, you'll find my
multiprogramming multiprocessing Forth for 68000s (easily adapted to
other CPUs). At the time (ca.1984), "multithreading" was unheard-of.
Soon (ha ha) I hope to port this to my multi-6809 system.
P.S. I missed the original post(s) to which you replied. Obviously my
netnews feed is either missing some articles, or is expiring them too
quickly for me to see them. So if I don't reply to a posting, please try
again...or better yet, send me email.
--
Brad Rodriguez b...@headwaters.com Computers on the Small Scale
Contributing Editor, The Computer Journal... http://www.psyber.com/~tcj
Director, Forth Interest Group......... http://www.taygeta.com/fig.html
=== 1996 Rochester Forth Conference: June 19-22 in Toronto, Ontario ===
= http://maccs.dcss.mcmaster.ca/~ns/96roch.html or email me for info =
Mike Coughlin
Chris Jakeman <cjak...@apvpeter.demon.co.uk> wrote:
>A starting point for meta-compiling might be Pygmy. This is "a lean,
>fast Forth with full source code. Incl full-screen editor, assembler and
>meta-compiler". It's public domain for DOS by Frank Sergeant.
Frank Sergeant, who has given everyone permission to copy it.
This is not the same as being in the public domain.
--
Michael Coughlin mi...@gnu.ai.mit.edu Cambridge, MA USA
Jim Schneider
>Chris Jakeman <cjak...@apvpeter.demon.co.uk> wrote:
>>In article <AD234FF4...@charlton.demon.co.uk>
>> gor...@charlton.demon.co.uk "Gordon Charlton" writes:
>>
>>> In article <8219019...@abwillms.demon.co.uk>,
>>> ala...@abwillms.demon.co.uk (Alaric B. Williams) wrote:
>>>
>>> }"Writing Forth From Scratch"
>>> }Has anyone ever written any good books on this?
And if I can through in my $0.02 worth, there were a bunch of articles on
Meta/Cross compiling in Dr. Dobb's about a decade ago. They were collected
into a two-volume set called "Dr. Dobb's Toolbook of FORTH". I believe the
Meta/Cross compiling articles are in the second volume.
(I just grabbed it off of my shelf, and yes, the articles are in the second
volume.) I don't know if the books are in print now or not. They were
published by M&T Books, with volume one editted by Marlin Oeuverson, and
volume two editted by "The Editors of Dr. Dobb's Journal". The ISBN for
volume two: 0-934375-41-0. The copyright date is 1987, so the articles are
probably way out of date.
For those of you who *MUST* *HAVE* *THESE* *BOOKS*, M&T Books list their
address as:
M&T Publishing, Inc., 501 Galveston Drive, Redwood City, CA 94063
and the phone number I have is:
(800) 533-4372 (or (800) 356-2002 in California).
Also, for the real archeaologists in the crowd, I still have a copy of the
version 1.0 Nautilus Systems FIG FORTH cross-compiler, with a copyright date
of 1981. I don't know where you could get a copy of it. (Mine's not for
sale.)
Disclaimers:
1) No, I don't work for M&T Books, I just happen to have a book they
published that I thought you might like to read.
2) The books in question are almost ten years old. It's anyones guess
as to if they are still in print.
3) Since I got the address and phone number from the book I mentioned,
it's anyones guess as to if M&T will still be there.
Allen David Boozer
>In article <4dqsi1$l...@news.worldlinx.com> Brad Rodriguez <b...@headwaters.com> writes:
>>Chris Jakeman <cjak...@apvpeter.demon.co.uk> wrote:
>>>In article <AD234FF4...@charlton.demon.co.uk>
>>> gor...@charlton.demon.co.uk "Gordon Charlton" writes:
>>>
>>>> In article <8219019...@abwillms.demon.co.uk>,
>>>> ala...@abwillms.demon.co.uk (Alaric B. Williams) wrote:
>>>>
>>>> }"Writing Forth From Scratch"
>>>> }Has anyone ever written any good books on this?
><a whole bunch of stuff deleted>
>
>And if I can through in my $0.02 worth, there were a bunch of articles on
>Meta/Cross compiling in Dr. Dobb's about a decade ago. They were collected
>into a two-volume set called "Dr. Dobb's Toolbook of FORTH". I believe the
>Meta/Cross compiling articles are in the second volume.
Could you post the references to the original magazine articles? Thanks!
I recently wrote my own 80x86 FORTH system from scratch. I wrote the first
version in assembly language, using "DW Word1, Word2, ..." to manually
compile forth words. I used macros to automatically link for words into the
dictionary structure, define constants & variables, and to simulate a
5-element stack to simplify conditionals & loop structures. Once a minimal
FORTH kernel was running, I wrote a crude metacompiler & converted the
assembly code to FORTH source. The metacompiler sped up the development
process enormously - in retrospect, I think I would have been better off
writing the metacompiler on an existing FORTH system, and then using it to
compile my own kernel. I am currently working on porting the FORTH system
to my TI-85 calculator (z80 based) using an improved version of the
metacompiler. I have spent a fair amount of time trying to make my
metacompiler pretty, and am still not very satisfied with it. It seems this
is one area where the extensibility of FORTH is a disadvantage. The whole
concept of a metacompiler does not apply to, say, C - you just run the
(n+1)st generation source code through the nth generation compiler.
Anyway, I would be very interested in a discussion on implementing a
metacompiler - has anyone come up with a reasonably elegant solution to
the problem?
- David Boozer (ad...@virginia.edu)
Chris Jakeman
mi...@gnu.ai.mit.edu "Mike Coughlin" writes:
> In article <821991...@apvpeter.demon.co.uk>,
> Chris Jakeman <cjak...@apvpeter.demon.co.uk> wrote:
> >A starting point for meta-compiling might be Pygmy. This is "a lean,
> >fast Forth with full source code. Incl full-screen editor, assembler and
> >meta-compiler". It's public domain for DOS by Frank Sergeant.
>
> Pygmy is NOT public domain. It is copyrighted by its author,
> Frank Sergeant, who has given everyone permission to copy it.
> This is not the same as being in the public domain.
Sorry Frank (and thanks Mike for pointing this out).
Can you say exactly what is meant by 'public domain' and how this differs
from the Pygmy distribution terms?
Frank Sergeant
basically, Chris Jakeman <cjak...@apvpeter.demon.co.uk>,
while suggesting Pygmy as a starting point for information
about metacompiling, said
> It's public domain for DOS by Frank Sergeant.
As far as I can tell, that posting never made it to my ISP,
so the first I knew of it was when mi...@gnu.ai.mit.edu
"Mike Coughlin" said
> Pygmy is NOT public domain. It is copyrighted by its author,
> Frank Sergeant, who has given everyone permission to copy it.
> This is not the same as being in the public domain.
Then, Chris said
> Sorry Frank (and thanks Mike for pointing this out).
> Can you say exactly what is meant by 'public domain'
> and how this differs from the Pygmy distribution terms?
I would like to thank both Chris and Mike.
To answer Chris's question, my position on Pygmy is that
is copyrighted by me and distributed as shareware. However,
it is shareware that does not require any payment to me.
Of course, that does not mean that money is unwelcome! As
an encouragement to "register" (i.e. pay me some money),
I offer a bonus disk. (The basic shareware Pygmy is freely
re-distributable, but the bonus disk is not.) Anyone is
free to use Pygmy to produce products for private, commercial,
whatever purposes. Pygmy is available on finer sites
everywhere (e.g. ftp.coast.net/simtel/msdos/forth/pygmy14.zip
or some such).
Basically, I don't want to give up any of my rights, whatever
they may be, yet I don't want anyone to feel reluctant to
communicate with me about Pygmy due to feeling that he was
required/expected to "register" Pygmy but hasn't.
Pygmy is currently at version 1.4, in spite of plans/hopes
of releasing a version 1.5. I _still_ hope to do it, but
can no longer guess when.
In an attempt to solve my newsfeed problem, I have switched
to getting the newsgroups from Eskimo North instead of my
local ISP, Axiom. I'll still try posting via Axiom.
Oops, I didn't exactly answer Chris's question(s).
> Can you say exactly what is meant by 'public domain'
but I'll leave that to others.
-- Frank (31 Jan 96)
serg...@axiom.net (current, but Axiom's on probabtion)
py...@pobox.com (permanent)
Anton Ertl
|> In article <4e2v5s$h...@GRAPEVINE.LCS.MIT.EDU>
|> mi...@gnu.ai.mit.edu "Mike Coughlin" writes:
|> > Frank Sergeant, who has given everyone permission to copy it.
|> Can you say exactly what is meant by 'public domain' and how this differs
|> from the Pygmy distribution terms?
Public domain means "no copyright".
- anton
--
M. Anton Ertl Some things have to be seen to be believed
an...@mips.complang.tuwien.ac.at Most things have to be believed to be seen
http://www.complang.tuwien.ac.at/anton/home.html
Joel Rubin
an...@a0.complang.tuwien.ac.at says...
or what can be done with derived works. For example, if you write a
program with the GNU C/C++ compiler, it becomes subject to the GNU
general license. Among other things, I think that means you have to
distribute the source code.
Kodogr
: Julian, and anyone else still on GEnie,
: If you haven't logged on in the last 24 hours you probably won't be
: aware that
: GEnie has just doubled their rates. To cancel, call 1 800 638-9636.
: You'll need your
: user ID.
: L. Greg Lisle, PE Forth Toolsmith L.G....@ieee.org
Their Forth area is pretty dead. I quit GEnie because GE is a major
corporate pollution monger. Lets hope that GEnie dies.
LgLisle
LgLisle
I classify this material as freeware.
Classifications of Software and/or other material.
I believe these are reasonably accurate definitions, but
there may be legal ramifications I am unaware of.
Commercial:
This material is copyrighted, and the author/creator retains
all rights to distribution, replication and use. A license is
given to those who buy the material. The license will specify
how, or if, the material may be copied and or distributed. The
license is usually quite restrictive.
Demoware:
This material is similar to Commercial, but the creator distributes
it at no charge. It is usually of limited functionality, or a
time limit may apply.
Shareware:
This material is copyrighted by the author/creator, but a license
is given to freely copy and distribute it. There are occasionally
restrictions on commercial distribution. Anyone who chooses to use
the material is required to pay a license fee and abide by any usage
restrictions applied by the author.
Freeware:
The author/creator retains ownership of the material, but gives
permission to copy, distribute and use it free to anyone. There
is usually a restriction on modification of the material, and
the creator usually requires that any ownership information be
kept with the material. The creator may at any time revoke
the license to distribute it.
Public Domain:
Material placed in the public domain is totally unprotected.
Anyone may take, copy and use it anyway they please. It is
community property. Once material is placed in the public domain
it can not be taken back.
Jim Schneider
Actually, your program is *not* subject to the GPL unless you explicitly
distribute it (the programs) under the terms of the GPL. Go read the
document. In a nutshell, if your program contains substantial portions
that are derived from a covered work, then your program is also under
the GPL. If you just use a tool that is covered by the GPL in
compiling/editting/linking/whatever your program, it is not covered by
the GPL. There is a rather tricky interpretation, however: If you
create a parser using bison (a yacc compatible compiler-compiler which
is covered by the GPL), the resulting C source code (usually in the
file fname.tab.c) is covered by the GPL, although your original bison
source isn't.
The GPL describes this very clearly.
Mike Coughlin
Joel Rubin <jmr...@ix.netcom.com> wrote:
>Copyrighted freeware can have restrictions as to how it is distributed
>or what can be done with derived works. For example, if you write a
>program with the GNU C/C++ compiler, it becomes subject to the GNU
>general license. Among other things, I think that means you have to
>distribute the source code.
of law can try copyright case when so few people can understand
copyrights. You do not have to distrubute the source code for a program
you write and publish if you use the GNU C/C++ compiler to compile it.
The GNU C/C++ compilier is published as source code. If you copy and
use any of this source code in your new program, then you must
distribute all of your new source code with the same GNU copyright.
This is intended to let people who want to share their programs keep
those who will not share their work in a similar way from using their
published source code.
The GNU copyright can apply to any program, but it gets
particularly confusing when applied to compiliers and Forth systems.
The code from the libraries and Forth kernel gets mixed up with any new
program that anybody writes. There is a provision in the GNU copyright
to cover this situation, but I don't understand it well enough to write
about it.
There are many different aspects to copyrights, freeware,
shareware, public domain, etc. You must find out exactly what
an author intends to be done with his work before you use it.
Don't guess that he thinks the way you do.
john wavrik
My understanding is that if you put your work in the public
domain, then you cannot restrict the use of this work or
require compensation (royalties, etc).
HOWEVER work put in the public domain is still your intellectual
property. One thing that someone else cannot do is claim that
the work is theirs.
[Old classical music is in the public domain. You can perform
a Mozart piece without paying royalties -- but you cannot claim
that you wrote it!]
jjwa...@ucsd.edu John J Wavrik
Dept of Math - UCSD
Brad Rodriguez
>Their Forth area is pretty dead. I quit GEnie because GE is a major
>corporate pollution monger. Lets hope that GEnie dies.
I may have neglected to report this earlier, but at the November meeting,
the FIG Board of Directors voted to terminate FIG's presence on GEnie.
Quite simply, all the action has moved to the Internet, and despite their
claims, GEnie is not a viable presence on the Internet right now. I
cancelled my account last week; for the prior two weeks, I saw no
messages posted to the GEnie Forth Roundtable.
Had GEnie kept up with the times, things might have been different. The
last bulletin I received announced a change of ownership of the GEnie
service; I'm curious to see what the new owners do (besides raising
rates).
--
Brad Rodriguez b...@headwaters.com Computers on the Small Scale
Contributing Editor, The Computer Journal... http://www.psyber.com/~tcj
Director, Forth Interest Group........... http://www.forth.org/fig.html
>>> 1996 Rochester Forth Conference: June 19-22 in Toronto, Ontario <<<
>> PAPERS WANTED! See http://maccs.dcss.mcmaster.ca/~ns/96roch.html <<
Keith Wootten
ala...@abwillms.demon.co.uk "Alaric B. Williams" writes:
>
> Has anyone ever written any good books on this? Especially for
> multiprogramming multithreading multiprocessor systems (yikes...)
[snip]
'Fig Forth - A Systems Guide' by Dr Ting is an excellent (if
elderly) how-to-write-a-Forth book. I don't know where to
get it though ... maybe Fig?
--
Keith Wootten
Mike Coughlin
Public domain means that the original author has no rights
under the copyright laws. Usually this means that the material
is very old (50 years after the death of the author, or 75 years
after the publication if the author is a corporation; check with
your lawer to make sure these numbers are correct). It used to be
required in the United States, that a copyright notice be attached
to published works; otherwise the works would be in the public
domain. In order to do things the way the rest of the world did
things (and produce more income for lawyers thru lawsuits), the
US law was changed. Now everything published is copyrighted,
unless the author specifically states that his work is public
domain. If you never intend to sue anybody no matter what they
do with what you have written, then say it is public domain.
Otherwise say nothing and let them worry.
I wonder what the courts will decide about copyrights on
usenet news articles.
Paul E. Bennett
mi...@gnu.ai.mit.edu "Mike Coughlin" writes:
> ......................................................... the
> US law was changed. Now everything published is copyrighted,
> unless the author specifically states that his work is public
> domain. If you never intend to sue anybody no matter what they
> do with what you have written, then say it is public domain.
> Otherwise say nothing and let them worry.
>
> I wonder what the courts will decide about copyrights on
> usenet news articles.
>
It sounds like you already answered that question yourself. If anyone thinks
otherwise I will have to ammend my sig. I try not to put sensitive information
up here anyway.
--
Paul E. Bennett
p...@transcontech.co.uk
Going Forth Safely
Chris Jakeman
Ke...@wootten.demon.co.uk "Keith Wootten" writes:
Free from the FIG UK library - membership is 10 pounds a year.
Contact Dr. Doug Neale on 0181 542 2747.
LgLisle
Sergeant) writes:
>Pygmy is currently at version 1.4, in spite of plans/hopes
>of releasing a version 1.5. I _still_ hope to do it, but
>can no longer guess when.
Uhm, Frank? Pygmy users?
Is there anything that can be done to encourage the release of v1.5?
What
are the obstacles? Are there volunteers to help?
Vernon M. Tallman
might be in v1.5?
Vern Tallman