Minutes of the FIGGY BAR RT Conference.

Date: 03/30/89   Time: 22:35EST

Attendees:
[[Gary] GARY-S]
[[len] NMORGENSTERN]
[[Kevin] APPERT]
[[jax is back] FIGCHAPTERS]
[[milt] M.FABERT]
[C.TING]
[D.BAKAS]
 
  Items discussed: learning script languages, UNIX, e-mailing for MM20 help,
                   stack operators NIP,TUCK,SWOOP,SLIP,SLIDE.., LOCALS,
                   Fred Olson's comparison of F-83 to BASIS7, RTX as a DMA
                   controller, fun at NASA, new book in works (?), fast
                   matrix multipliers on standard chips, bit banging on
                   Ethernet, why FIG-Forth won't die
 
Minutes:
<[len] NMORGENSTERN> Hi, Gary
<[Gary] GARY-S> hey len - welcome
<[len] NMORGENSTERN> I learned 3 languages over the weekend!...
<[len] NMORGENSTERN> Actually 2 were updates of old ones.
<[Gary] GARY-S> spanish,spanish, and forth
<[len] NMORGENSTERN> No. 1 was a new linked hypertext editor. One was the 
outliner..
<[len] NMORGENSTERN> maxthink, and the third was a new version of ACTA for the 
mac...
<[len] NMORGENSTERN> also an outliner.
<[len] NMORGENSTERN> My mind is going in circles!
<[Gary] GARY-S> I learned there is a vast difference in different 
implementations of UNIX
<APPERT> is here.
<[len] NMORGENSTERN> Hi Kevin.
<[Kevin] APPERT> hi len, gary
<[Gary] GARY-S> kev
<[len] NMORGENSTERN> ga Gary you were talking about Unix.
<[Kevin] APPERT> barf
<[Gary] GARY-S> len - I have been meaning to thank you for participating in 
Thursday FIGGY's...
<[Gary] GARY-S> I certainly don't reciprocate on Sunday...
<[len] NMORGENSTERN> That's ok. I do what I can.
<[Gary] GARY-S> I hope you don't mind the obvious disparity
<[len] NMORGENSTERN> I answer few letters, I don't do much about files.
<[Gary] GARY-S> I was honest going in - I can't be here Sundays
<[Gary] GARY-S> Kev - what did you ask ?
<[len] NMORGENSTERN> Why don't you like UNIX, Kevin? I have never used it
<[Kevin] APPERT> i was just barfing in response to the word EUNIX
<[len] NMORGENSTERN> A eunuch is a man who was cut out to be a bachelor.
<[Gary] GARY-S> I understand - either you accept it or you don't
<[Kevin] APPERT> It's a pile of little black boxes into which you cannot see.
<[jax is back] FIGCHAPTERS> hi
<[Kevin] APPERT> hi Jax
<[Gary] GARY-S> jaxon
<M.FABERT> is here.
<[jax is back] FIGCHAPTERS> got my car
<[Gary] GARY-S> and it is....
<[len] NMORGENSTERN> Aren't all languages black boxes, except Forth & assembler?
<[jax is back] FIGCHAPTERS> cheap! mazda 323, but brand new. Nice.
<[Kevin] APPERT> if you cannot hold it in your hand and understand it with your 
mind you cannot trust it and you dare not use it
<[Kevin] APPERT> congratulations, Jax. you deserve it
<[Gary] GARY-S> That leaves out a lot of ladies
<[jax is back] FIGCHAPTERS> brb ... baby  ... thanx, Kev.
<[len] NMORGENSTERN> :-) :-) :-) :-)
<[Kevin] APPERT> :) it's not considered apropriate to use other people and 
espicially ladies these days.
<[Gary] GARY-S> Len is this the comm prog you kept dumping macros with ?
<[len] NMORGENSTERN> Kevin, your comment reminds me of the "Dog's comment on 
life"
<[Kevin] APPERT> yes?
<[Gary] GARY-S> I was refering to understanding the ladies
<[len] NMORGENSTERN> If you can't eat it, and you can't [vulgar word for have 
intercourse with]
<[len] NMORGENSTERN> it, then piss on it!
<[Kevin] APPERT> I never have understood ladies.  I don't think it's a criteria.
<[Gary] GARY-S> milt - hows the NASA troops
<[len] NMORGENSTERN> Jax, Gary's notice was about "fancy" stack words...
<[milt] M.FABERT> Moving along I was at nasa HQ yesterday - what a zoo!
<C.TING> is here.
<[Gary] GARY-S> did you feed the animals ?
<[Kevin] APPERT> Ting!  hi
<[Gary] GARY-S> Dr. Ting ! nice surprise
<C.TING> hi, gary and kevin
<[milt] M.FABERT> yea.  software souflee(sp?)
<[Gary] GARY-S> for all unsuspecting - Thursday (non-guest Thursdays) is 
free-for-all ...
<[Gary] GARY-S> just jump in with comments
<[jax is back] FIGCHAPTERS> " a food fight of ideas"
<[Gary] GARY-S> my words (EXACTLY)
<[jax is back] FIGCHAPTERS> Happy New Year, Dr. T.
<[Gary] GARY-S> Jax - favor time
<C.TING> thanks, jax
<[Gary] GARY-S> Got a request
<[jax is back] FIGCHAPTERS> ..?
<[Gary] GARY-S> Next e-mail to Australia - ask Lance for some docs for MM20/MX20
<[jax is back] FIGCHAPTERS> is what an MM20?
<[Gary] GARY-S> MM20 = block f83 for CP/M, aussie style, MX = TEXT (F-PC like)
<[jax is back] FIGCHAPTERS> ok, leave me a note in mail.
<[Kevin] APPERT> assault rifles are being banned in some commutities in CA
<[Gary] GARY-S> I will be totally out of pocket for a couple of weeks and..
<[Gary] GARY-S> thought if you had fig chapter mailings..
<[Gary] GARY-S> you might ask.
<[len] NMORGENSTERN> Jax, Gary's notice was about "fancy" stack words...
<[jax is back] FIGCHAPTERS> next mailing is second week april, jaxthinx
<[len] NMORGENSTERN> since you advocated a set, why don't you defend the idea.
<[jax is back] FIGCHAPTERS> ok
<[jax is back] FIGCHAPTERS> I just noted ...
<[jax is back] FIGCHAPTERS> that we have certain shallow stack operations named 
in Forth ....
<[jax is back] FIGCHAPTERS> whereas complementary operations, somewhat less 
common but no more complex ...
<[jax is back] FIGCHAPTERS> are omitted, such as: ...
<[jax is back] FIGCHAPTERS> SWOOP ( n1 n2 --- n2 n1 n1 ) ...
<[jax is back] FIGCHAPTERS> SLIP ( n1 n2 --- n2 n2 n1) ...
<[jax is back] FIGCHAPTERS> SLIDE ( or UNDERDUP) ( n1 n2 --- n1 n1 n2) ...
<[jax is back] FIGCHAPTERS> and a coupla others. I use all these. SWOOP more 
common than others. ...
<[jax is back] FIGCHAPTERS> but for completeness sake, why not? But 
disregarding my "cutsie" names for them ...
<[jax is back] FIGCHAPTERS> BADEN advocated same, with more rational naming 
scheme ...
<[jax is back] FIGCHAPTERS> rejected in three minutes by TC, he says!
<[len] NMORGENSTERN> You have expressed on several occasions worry about 
keeping Forth simple.
<[len] NMORGENSTERN> Are you being inconsistent?
<[milt] M.FABERT> how about replacing all of the above with Locals?
<[jax is back] FIGCHAPTERS> Well ...
<[jax is back] FIGCHAPTERS> What ARE locals, after all, but stack items!!!!
<[jax is back] FIGCHAPTERS> Len ...
<[jax is back] FIGCHAPTERS> You may be right. But it seems like if you are 
going to stick weird things into Forth, which the TC seems to be intent on
doing ...
<C.TING> No more stack words.  I still cannot remember how ROT rotates.
<[jax is back] FIGCHAPTERS> you might as well complete the mathematical set of 
shallow stack operators ...
<[jax is back] FIGCHAPTERS> the selection we have chosen is arbitrary, and 
frankly ...
<[Gary] GARY-S> I don't like all the stuff being moved out of 'standard' into 
reference. To what end ?
<[milt] M.FABERT> But locals are easier to write and to read, jax.
<[Kevin] APPERT> I always try ROT myself to be sure it is still doing what I 
think it should, Ting
<[jax is back] FIGCHAPTERS> DX POP AX POP DX PUSH AX PUSH 1PUSH C; is faster 
that : SWOOP SWAP DUP ;
<[Gary] GARY-S> Has everyone read Fred Olson's comparison of F83 and BASIS 7 ?
<[len] NMORGENSTERN> Milt, give an example
<[jax is back] FIGCHAPTERS> But I tend to agree ...
<[Kevin] APPERT> is it in the BB, or files, Gary?
<[jax is back] FIGCHAPTERS> that FORTH should be left SIMPLE ...
<[Gary] GARY-S> both
<[jax is back] FIGCHAPTERS> IF they voted a standard that said ....
<[jax is back] FIGCHAPTERS> that the only standard word was FORTH ...
<[jax is back] FIGCHAPTERS> I would vote for it.
<[len] NMORGENSTERN> You can fight fire with fire
<[len] NMORGENSTERN> but you can't fight complexity with complexity.
<C.TING> Where is Olson's comparison? Gary?  Send me a copy!
<[milt] M.FABERT> : sum_of_squares ( a b --- a^2+b^2)
<[milt] M.FABERT> locals| b a |
<[milt] M.FABERT> a a *
<[milt] M.FABERT> b b *
<[Kevin] APPERT> that's almost as silly as this nonsense about ones compliment, 
Jax
<[milt] M.FABERT> +
<[milt] M.FABERT> ;
<[jax is back] FIGCHAPTERS> JForth has locals, Milt.
<[len] NMORGENSTERN> Thanx, Milt
<[jax is back] FIGCHAPTERS> Kevin, silly is as silly does.
<[jax is back] FIGCHAPTERS> I think that the standard is off track ...
<[Kevin] APPERT> I agree
<[jax is back] FIGCHAPTERS> for lack of technopeasant input.
<[milt] M.FABERT> What we need is for locals to be in a standard ...
<[milt] M.FABERT> multiforth has a great implementation ...
<[milt] M.FABERT> but I need then for the RTX2000.
<[len] NMORGENSTERN> I don't like the use of the term "local" for what you
<[len] NMORGENSTERN> have done, although it is on its way to acceptance.
<[len] NMORGENSTERN> I always think of locals as in Pascal & other languages
<[len] NMORGENSTERN> They can by anything: constants, variables, colon defs
<[len] NMORGENSTERN> which have a temporary name. When released, the original 
name
<[len] NMORGENSTERN> is restored.
<[milt] M.FABERT> a horse by any other name is fine by me.
<[len] NMORGENSTERN> I proposed a system at Austin last month.
<[len] NMORGENSTERN> A horse & a zebra are related but different.
<[milt] M.FABERT> don't let either NIP you!
<[jax is back] FIGCHAPTERS> What is new in the publishing biz, Dr. T?
<C.TING> May be we should see how RISC's do its windows.  It takes care of lots 
of locals.
<C.TING> F-PC is selling like hot cakes.
<[len] NMORGENSTERN> At the W. Coast Computer Faire, we were selling them
<[jax is back] FIGCHAPTERS> That's because it's cheaper to buy than to download!
<[jax is back] FIGCHAPTERS> ... :-)
<[len] NMORGENSTERN> so fast, we couldn't copy the disks fast enough.
<[Gary] GARY-S> Those interested in Fred Olson's comparison of F-83 to the 
current BASIS please read either Cat 14, Topic 2, Messages 35 through 43 -O
R- File 1638 in the library.
<[jax is back] FIGCHAPTERS> Do you ever publish books by other authors, Dr. T?
<[Gary] GARY-S> Did everyone get that ?
<[Kevin] APPERT>
<C.TING> The Indelko RTX kit is now very available.
<[Kevin] APPERT> a/sen 6 I missed your reply if you sent one, know anybody with 
f83 in rom for 8086 with source available?
<C.TING> Yes, I got the numbers .  Thanks, Gary.
<[Kevin] APPERT> ack, I got it, thanks Gary
<[milt] M.FABERT> dr. T, Have you done any more on parallel processors?
<[Gary] GARY-S> since your private send went public let's all here if the ROM 
F83 is available
<C.TING> Jax, Yes. I do like to publish other people's books.  If they allow me.
<[jax is back] FIGCHAPTERS> I had an idea for a monograph.
<C.TING> Milt, I am still in love with SIMD parallel structures. But no real 
activity.
<[jax is back] FIGCHAPTERS> It would be the sort of thing more interesting to 
Forthers than the general public.
<C.TING> Jax, I am interested.  I can do printing and selling.
<[Kevin] APPERT> what topic, Jax
<[jax is back] FIGCHAPTERS> Let me work on it. I am sure you will find it 
amusing at least ...
<[jax is back] FIGCHAPTERS> In the 1920's ...
<[jax is back] FIGCHAPTERS> Richard Reti wrote the definitive monograph ...
<[jax is back] FIGCHAPTERS> defining the place of Chess in the intellectual 
sphere ...
<[jax is back] FIGCHAPTERS> called "Modern Ideas In Chess" ...
<[jax is back] FIGCHAPTERS> Programming is a field intellectually intensive ...
<[jax is back] FIGCHAPTERS> but frequently viewed as culturally void ...
<D.BAKAS> is here.
<[jax is back] FIGCHAPTERS> I have not found it to be so void as the general 
public thinks ...
<[jax is back] FIGCHAPTERS> and I would like to try to assert ..
<[jax is back] FIGCHAPTERS> the place of Forth in the realm of creative 
endavour of humankind.
<C.TING> Milt, some friends are talking about a very fast matrix multiplier 
machine
<C.TING> are you interested?
<[milt] M.FABERT> sure.  Is it VHSIC based?
<C.TING> Jax, sounds great.  How many pages?
<[jax is back] FIGCHAPTERS> I envision about 120 to 150 laser set.
<C.TING> Milt, no.  Simple multiplier chips and memory.
<[milt] M.FABERT> What application do they have in mind?
<C.TING> Jax.  No problem.  Let me know when you have a draft.
<[jax is back] FIGCHAPTERS> Thanx, will do.
<C.TING> Milt, all scientific and engineering problems are cast in matrices.
<C.TING> Milt, a fast matrix multiplier will help solve most of these problems.
<[milt] M.FABERT> I know, I was just curious as whether it was customized for
<[milt] M.FABERT> a particular project.
<[milt] M.FABERT> It may be of some interest to the Global Change instruments ..
<[milt] M.FABERT> some of which have data rates on the order of a Giga bit/sec.
<C.TING> Jax, not yet.  Do it fast and sent 2 copies to the Copyright office in 
Library of congress.
<[jax is back] FIGCHAPTERS> Ok!
<[Kevin] APPERT> TRW has some interesting chips these days in the DSP area 
these days , FFT, convolvers, matrix mult., rect. -> polar coord transrer.
<C.TING> Kevin, these are things we caqn use a fast matrix multiplier for.
<[Kevin] APPERT> they were talking about putting a counter into the 
polar<->rect (biderectional) and using it to generate digital quadrature!  
seems l
ike there should be a simpler way!
<C.TING> The major problem is to get data in and out of this structure...
<C.TING> the classic von Neumann bottle neck.
<[Kevin] APPERT> you may very well wind up building it from TRW parts.
<C.TING> Kevin, I need to be in Lockheed to use TRW parts.
<[Kevin] APPERT> or get a govt grant.
<[Kevin] APPERT> if Lockheed can get Govt money, the criteria may not be all 
that strict!
<[Kevin] APPERT> :)
<C.TING> For experimental purposes, RTX is quite ok.
<[Kevin] APPERT> yes, I agree
<[Gary] GARY-S> Do they still operate the skunk works ?
<[milt] M.FABERT> Electronic Design had a good article on using the ...
<[Kevin] APPERT> the Kelly Johnson center for never mind is still in buisnesss
<[len] NMORGENSTERN> What's the skunk works?
<C.TING> No.  They are operating paper mills.
<[milt] M.FABERT> RTX as a DMA controller.  I adapted it to a design ...
<[milt] M.FABERT> I'm working on at Langley.
<[Kevin] APPERT> the closed door operation, Len, which produced the U2, the 
SR71, and other things
<[Kevin] APPERT> run by this fellow Johnson who actually got things done
<[len] NMORGENSTERN> Thanx
<[Gary] GARY-S> That should make for a pretty optimal DMA controller Milt !
<[Kevin] APPERT> the customer trusted him and he would not put up with 
unnescessary fiddle faddle, QA, paperwork, or delays from the customer or from
his own people
<[milt] M.FABERT> 20 mbytes/sec with no stack limitation!
<[Gary] GARY-S> What sort of bit rate can you absorb ?
<[Gary] GARY-S> thanks
<[Gary] GARY-S> Put that in your streamer and...
<[Gary] GARY-S> zooooooooooooooom
<C.TING> Yes.  You can use RTX to push lots of data around real fast.
<[len] NMORGENSTERN> Milt, you mean Mbytes! mbytes are millibytes.
<[milt] M.FABERT> ok ok
<[milt] M.FABERT> I get payed uDollars for doing it!
<[Gary] GARY-S> Sounds like a neat idea for maxtor, ... ????
<[Kevin] APPERT> maxtor isn't intersted in neat ideas
<[len] NMORGENSTERN> Femtodollars are even smaller
<[Gary] GARY-S> on board DMA ?
<[milt] M.FABERT> VME, inc. used it for the SCSI controller.
<[Gary] GARY-S> God - I hate their BUS
<[milt] M.FABERT> beats the microchannel.
<[Gary] GARY-S> true
<C.TING> The Forth Mafia in Matrox tried to sell the idea, but didn't get very 
far.
<[Kevin] APPERT> you could probably bit-bang an ethernet cable if there was 
anything usefull on an ethernet cable, ever
<[Gary] GARY-S> If all the collission detection didn't get in the way
<[milt] M.FABERT> Oh, come now I send circuit netlists to HHB's simulator
<[milt] M.FABERT> on a SUn using the ethernet.
<[Kevin] APPERT> protocol takes the $%&# thing from megahertz to kilohertz,it's 
not a solution to anything
<[Gary] GARY-S> How many other nodes - that's the criteria
<[milt] M.FABERT> 10 Mbits down to 1Mbit is more like it.
<[milt] M.FABERT> local clustering helps.
<[Gary] GARY-S> Then you get back to a bottle neck getting out of the local
<[Kevin] APPERT> I have seen a dec set-up transfer files at a wopping 10kbaud!  
bad implementatiion
<C.TING> By the way, I am updating the Sytems Guide to figForth.  Any body's 
interested?
<[Kevin] APPERT> I
<[len] NMORGENSTERN> Is FigForth still popular?
<[Gary] GARY-S> Used in a lot of labs still...
<[Kevin] APPERT> Charles Curley likes it (so do I! :)
<[Gary] GARY-S> heck Kitts peak like implementations are still in use
<C.TING> It simply refuses to die.  MVP keeps odering them and I have run out 
many times.
<[milt] M.FABERT> We're using FigForth on an HP1000 to control testing ...
<[len] NMORGENSTERN> That's very interesting. I thought it was kind of moribund.
<[Kevin] APPERT> Anybody else have the Kitt Peak Vax Forth (not an old-time 
implementation)
<[milt] M.FABERT> of an atmospheric monitoring instrument.
<[Gary] GARY-S> Why rewrite your code that has been running monitors for 
decades with no prob ?
<[Kevin] APPERT> I learned Forth from the Kitt Peak Primer  those were the days
<[Kevin] APPERT> I understand they ripped out a forth-based telescope 
something-or-other at Kitt Peak which had been working for these many years, to
replace it with degraded functionality
<C.TING> It looks like that figForth is still be best model if you have to put 
it on a uP all by your self.
<[Kevin] APPERT> or ROM it
<[len] NMORGENSTERN> What's up, Ting?
<[len] NMORGENSTERN> (I got it Microprocessor)
<C.TING> Well, I am transfering the Systems Guide to a Mac for laser printing.
<[Gary] GARY-S> There aren't a lot of ASCII keyboards with mu on them leonard :
-)
<[len] NMORGENSTERN> There are lots, but you can't transmit them by wire!
<[Kevin] APPERT> do you have it on IBMPC disks, or LSI11?
<C.TING> On Mac disks now.
<[Kevin] APPERT> ok, thought you might like to transfer  from decland.  let me 
know if ever you so wish
<[Kevin] APPERT> I think I got the last box of RX02 floppies from stationary 
stores :)
<[Gary] GARY-S> Jax - if you e-mail Australia - the problem with the lack of 
Docs is there seems to be no editor |-/
<C.TING> Thanks, Kevin.  I can always find help in your direction.
<[len] NMORGENSTERN> I have  Mac SE, Ting. Let me know if I can help.
<[jax is back] FIGCHAPTERS> ok, be sure to send me detailed email what you want,
 so i can forward it.
<[milt] M.FABERT> Sorry I didn't look you up while I was in Boulder, Jax ...
<C.TING> I was given a Mac IIx at the place I now work.  Very convenient.
<[milt] M.FABERT> but I couldn't resist skiing during my free time there.
<[len] NMORGENSTERN> Wow! wish I had one!
<[Gary] GARY-S> look at the messages on BB at Cat 1 Topic 5 messages 1,2,3 jax
<[Kevin] APPERT> how's the protien macro-assembler coming , Ting?
<[jax is back] FIGCHAPTERS> make it easy for me and I'll remember to do it, 
gars!
<[Kevin] APPERT> well fellow software samurai, I must be off to dinner.  fare 
you all well and good night.
<[len] NMORGENSTERN> Sorry, fellows, I have to leave. Very good conference!
<C.TING> I always have problem with commumication over RS232, with polyForth as 
always.
<[Kevin] APPERT> has left.
<[Gary] GARY-S> It's past last call on the FIGGY anyway - thanks for a 
stimulating evening folks - see you in about three weeks
<C.TING> thank you Gary.  It is a lively conference.
<C.TING> has left.
<[Gary] GARY-S> stay on and chat if you wish, but FIGGY is closed.
=== End of Steno notes. ===