EuroForth 2008 announcement and other EuroForth stuff
Anton Ertl
already have your 2008 calendars, so you can immediately note the
dates:
May-June : Registration becomes available
June 27: Deadline for draft papers (academic stream)
August 12: Notification of acceptance of academic stream papers
August 19: Registration deadline
September 15: Deadline for camera-ready paper submission
(academic and industrial stream)
September 25-26: Forth200x meeting
September 26-28: EuroForth 2008
The home page of EuroForth 2008 is
<http://www.complang.tuwien.ac.at/anton/euroforth/ef08.html>; the Call
for Papers is already available
<http://www.complang.tuwien.ac.at/anton/euroforth/ef08/cfp.html>.
Also, there is finally a EuroForth 2007 homepage:
<http://www.complang.tuwien.ac.at/anton/euroforth/ef07.html>.
Also new: the 2007 Proceedings and Papers page:
<http://www.complang.tuwien.ac.at/anton/euroforth2007/papers/>.
And there are Euroforth 2007 photos by Bernd Paysan and (new) me on
<http://www.forth-ev.de/gallery/euroforth2007>.
Finally, the EuroForth homepage has moved: it is currently residing at
<http://www.complang.tuwien.ac.at/anton/euroforth/>, and has a
permanent home at <http://www.euroforth.org/>; this site and the old
site currently redirect to the current page. Thanks to Peter Knaggs
for maintaining the web site up to 2007, for providing the content and
for the redirections.
- anton
--
M. Anton Ertl http://www.complang.tuwien.ac.at/anton/home.html
comp.lang.forth FAQs: http://www.complang.tuwien.ac.at/forth/faq/toc.html
New standard: http://www.forth200x.org/forth200x.html
EuroForth 2008: http://www.complang.tuwien.ac.at/anton/euroforth/ef08.html
Alex McDonald
wrote:
I noted with interest; "Klaus Schleisiek: Dying Forth". Would a
summary of his argument be possible?
--
Regards
Alex McDonald
Slava Pestov
> I noted with interest; "Klaus Schleisiek: Dying Forth". Would a
> summary of his argument be possible?
Possibly something along the lines of http://www.everything2.com/index.pl?node_id=1391352
Slava
Marcel Hendrix
> On Dec 12, 1:28 am, Alex McDonald <b...@rivadpm.com> wrote:
>> I noted with interest; "Klaus Schleisiek: Dying Forth". Would a
>> summary of his argument be possible?
( read this one first )
> Possibly something along the lines of http://www.everything2.com/index.pl?node_id=1391352
E.g.
The "Forth is dying" troll is now a classic. Every time an article is
posted in the news about Forth, hordes of trolls inevitably post the
text of "Forth is dying" on CLF. Amazingly, even now it still gets
people responding to it. The text has evolved somewhat over the ages,
but the current common version is as follows:
It is official; DDJ now confirms: Forth is dying. One more crippling
bombshell hit the already beleaguered Forth community when IDC confirmed
that Forth market share has dropped yet again, now down to less than a
fraction of 1 percent of all embedded systems. Coming close on the
heels of a recent DDJ survey which plainly states that Forth has
lost more market share, this news serves to reinforce what we've
known all along. Forth is collapsing in complete disarray, as
fittingly exemplified by failing dead last in the recent Dr. Dobb's
Journal comprehensive embedded systems test.
You don't need to be a Kreskin to predict Forth's future. The hand
writing is on the wall: Forth faces a bleak future. In fact there
won't be any future at all for Forth because Forth is dying. Things
are looking very bad for Forth. As many of us are already aware,
Forth continues to lose market share. Red ink flows like a river of
blood.
LSE64 is the most endangered of them all, having lost 93% of its
core developers. The sudden and unpleasant departures of long time
LSE64 developers Jeff Fox, Mike Coughlin and Pete Werty only
serve to underscore the point more clearly. There can no longer be
any doubt: LSE64 is dying.
Let's keep to the facts and look at the numbers.
Gforth leader Anton states that there are 7000 users of Gforth. How
many users of Gforth are there? Let's see. The number of Gforth
versus LSE64 posts on Usenet is roughly in ratio of 5 to 1.
Therefore there are about 7000/5 = 1400 LSE64 users. Forth/OS
posts on Usenet are about half of the volume of LSE64 posts.
Therefore there are about 700 users of Forth/OS. A recent article put
Gforth at about 80 percent of the Forth market. Therefore there are
(7000+1400+700)*4 = 36400 Gforth users. This is consistent with the
number of Gforth Usenet posts.
Due to the troubles of Silicon Valley, abysmal sales and so on,
LSE64 went out of business and was taken over by IntellasSys who
sell another troubled OS. Now IntellaSys is also dead, its corpse
turned over to yet another charnel house.
All major surveys show that Forth has steadily declined in market
share. Forth is very sick and its long term survival prospects are
very dim. If Forth is to survive at all it will be among OS
dilettante dabblers. Forth continues to decay. Nothing short of a
cockeyed miracle could save Forth from its fate at this point in
time. For all practical purposes, Forth is dead.
Fact: Forth is dying
There are a number of reasons why this seems to be such a successful
troll:
1. Target audience: Forth users are often very proud of the
fact they use Forth, so there is always someone who is more than
willing to defend it.
2. Easily knocked down "facts". The paragraph estimating the number
of Forth users for example is very clearly absurd. The fact that
many of the claims are easy to rebut increases the likelihood
someone will respond.
3. Incorrect claims. For example, the DDJ survey actually shows
that a large portion of the most reliable machines in deep space are
running Forth.
4. It portrays opinions as facts. The tone is authoritative, which
further encourages people to reply to correct the "mistakes". The
fact that it is clearly absurd makes it even funnier when people reply :o)
BSD is dying has spawned a number of children. The text to Mike
Coughlin's explanation of why he left Forth, and "What we can learn
from Forth" are examples.
Overall, its impressive that after so long it still gets people
responding, but it is kind of irritating that a large portion of
posts in all Forth articles are now Forth is dying crapfloods.
Alex McDonald
> Slava Pestov <sl...@jedit.org> wrote Re: EuroForth 2008 announcement and other EuroForth stuff
>
> > On Dec 12, 1:28 am, Alex McDonald <b...@rivadpm.com> wrote:
> >> I noted with interest; "Klaus Schleisiek: Dying Forth". Would a
> >> summary of his argument be possible?
>
> ( read this one first )
>
My apologies if I have stood on one of your bunions.
This was one of two discussions that did not result from a workshop,
and had no paper associated with it. Having met Klaus at euroForth
2006, I was interested in his opinion.
--
Regards
Alex McDonald
Anton Ertl
[Very nice parody of market share press releases]
>There are a number of reasons why this seems to be such a successful
>troll:
> 1. Target audience: Forth users are often very proud of the
> fact they use Forth, so there is always someone who is more than
> willing to defend it.
Actually, you see such articles in many newsgroups for minority
languages. Last I looked (several years ago), most articles in
comp.lang.modula-2 were "Modula-2 is dying" articles.
>Overall, its impressive that after so long it still gets people
>responding, but it is kind of irritating that a large portion of
>posts in all Forth articles are now Forth is dying crapfloods.
My impression is that the "Forth is dying" theme is less frequent in
comp.lang.forth than it was 10-15 years ago. Most comp.lang.forth
regulars have seen the topic often enough that they are through with
it, and most newbies know that Forth is a minority language and don't
have delusions of grandeur. The only one who has tried to play this
troll in the last few years is John Doty; he's an old Forther, but
relatively new to c.l.f (at least he was not here when "Forth is
dying" postings were en vogue), and he thinks that he can hype LSE64
by playing that troll.
Anton Ertl
>On Dec 9, 6:08 pm, an...@mips.complang.tuwien.ac.at (Anton Ertl)
>wrote:
>> <http://www.complang.tuwien.ac.at/anton/euroforth/ef07.html>.
>>
>> Also new: the 2007 Proceedings and Papers page:
>> <http://www.complang.tuwien.ac.at/anton/euroforth2007/papers/>.
everthing!]
>I noted with interest; "Klaus Schleisiek: Dying Forth". Would a
>summary of his argument be possible?
I don't remember much (I did not find the topic that interesting), so
here's just my impression: Klaus has finally noticed that Forth is no
longer as popular as in the 80s, especially when looking at events
like EuroForth, so his introduction to the workshop sounded much like
one of the "Forth is dyning" postings in c.l.f from the '90s. He
suggested having a nice funeral or something. IIRC the workshop then
mainly discussed some ways of making Forth events and publications
like EuroForth more popular.
Alex McDonald
wrote:
> Alex McDonald <b...@rivadpm.com> writes:
> >On Dec 9, 6:08 pm, an...@mips.complang.tuwien.ac.at (Anton Ertl)
> >wrote:
> >> Also, there is finally a EuroForth 2007 homepage:
> >> <http://www.complang.tuwien.ac.at/anton/euroforth/ef07.html>.
>
> >> Also new: the 2007 Proceedings and Papers page:
> >> <http://www.complang.tuwien.ac.at/anton/euroforth2007/papers/>.
>
> [Fullquote shortened to the probably relevant part; please don't quote
> everthing!]
I'm sorry, a habit born of other fora and a necessary defence when
replying to certain posters on clf.
>
> >I noted with interest; "Klaus Schleisiek: Dying Forth". Would a
> >summary of his argument be possible?
>
> I don't remember much (I did not find the topic that interesting), so
> here's just my impression: Klaus has finally noticed that Forth is no
> longer as popular as in the 80s, especially when looking at events
> like EuroForth, so his introduction to the workshop sounded much like
> one of the "Forth is dyning" postings in c.l.f from the '90s. He
> suggested having a nice funeral or something. IIRC the workshop then
> mainly discussed some ways of making Forth events and publications
> like EuroForth more popular.
>
> - anton
> --
> M. Anton Ertl http://www.complang.tuwien.ac.at/anton/home.html
> comp.lang.forth FAQs:http://www.complang.tuwien.ac.at/forth/faq/toc.html
> New standard:http://www.forth200x.org/forth200x.html
> EuroForth 2008:http://www.complang.tuwien.ac.at/anton/euroforth/ef08.html
Oh, ok. I thought it might have had more substance and analysis than
the average wake. Thanks for the update anyhow.
--
Regards
Alex McDonald
John Doty
> The only one who has tried to play this
> troll in the last few years is John Doty; he's an old Forther, but
> relatively new to c.l.f (at least he was not here when "Forth is
> dying" postings were en vogue), and he thinks that he can hype LSE64
> by playing that troll.
Not at all. LSE64 is what I've got: it is not really what I want. It
makes money for me by being a useful tool, but it's pretty crude. I
don't sell it: you want it, you can have it (GPL). It's better for my
purposes than Standard Forth, but that, of course, is faint praise indeed.
What I want is a 21st century Forth capable of bringing back old
Forthers who've given up, and recruiting the youngsters who use things
like Python in embedded systems because there's nothing better that
makes sense to them. Standard Forth represents what this community once
embraced but now has thoroughly rejected. LSE64 is at best a small
exploratory step toward something more usable. ColorForth is
fascinating, but not, I think, something that can be embraced by a
significant user community.
There's enough brain power in this group to create a user-oriented 21st
century Forth, but it requires major excavation of heads from sand.
--
John Doty, Noqsi Aerospace, Ltd.
http://www.noqsi.com/
--
Specialization is for robots.
Bernd Paysan
> I don't remember much (I did not find the topic that interesting), so
> here's just my impression: Klaus has finally noticed that Forth is no
> longer as popular as in the 80s, especially when looking at events
> like EuroForth,
And it's even less popular than in the 90s...
> so his introduction to the workshop sounded much like
> one of the "Forth is dyning" postings in c.l.f from the '90s. He
> suggested having a nice funeral or something. IIRC the workshop then
> mainly discussed some ways of making Forth events and publications
> like EuroForth more popular.
As a summary, one could say that the events are organized in a less
enthusiastic and energetic way as they were in the days when Forth took up
(even though the persons are the same), so the solution found was to put
the message-spreading work and similar organization work on more shoulders.
--
Bernd Paysan
"If you want it done right, you have to do it yourself"
http://www.jwdt.com/~paysan/
John Doty
> Alex McDonald <bl...@rivadpm.com> writes:
>> On Dec 9, 6:08 pm, an...@mips.complang.tuwien.ac.at (Anton Ertl)
>> wrote:
>>> Also, there is finally a EuroForth 2007 homepage:
>>> <http://www.complang.tuwien.ac.at/anton/euroforth/ef07.html>.
>>>
>>> Also new: the 2007 Proceedings and Papers page:
>>> <http://www.complang.tuwien.ac.at/anton/euroforth2007/papers/>.
> [Fullquote shortened to the probably relevant part; please don't quote
> everthing!]
>
>> I noted with interest; "Klaus Schleisiek: Dying Forth". Would a
>> summary of his argument be possible?
>
> I don't remember much (I did not find the topic that interesting), so
> here's just my impression: Klaus has finally noticed that Forth is no
> longer as popular as in the 80s, especially when looking at events
> like EuroForth, so his introduction to the workshop sounded much like
> one of the "Forth is dyning" postings in c.l.f from the '90s. He
> suggested having a nice funeral or something. IIRC the workshop then
> mainly discussed some ways of making Forth events and publications
> like EuroForth more popular.
Forth events are irrelevant: there will always be True Believers who
will come to them. The measure is to look for Forth where people are
engaged in solving problems. Good luck finding it unless you know
beforehand where to look...
Bernd Paysan
> Forth events are irrelevant: there will always be True Believers who
> will come to them.
But less and less.
> The measure is to look for Forth where people are
> engaged in solving problems. Good luck finding it unless you know
> beforehand where to look...
Yes, tricky ;-). But it's not that we don't have any connections at all -
it's just spread over the True Believers.
Ian Osgood
wrote:
I do my bit by contributing nicely-factored Forth code to language
comparison web sites like Rosetta Code and pushing Forth in embedded
contexts like my rocketry club. I also speak up for Open Firmware when
it makes sense (I sure wish I could work in OFW rather than EFI). I am
indirectly supporting Forth (and other lovely languages like Python
and Squeak) by donating to One Laptop per Child and telling people
that Forth was used to accelerate its hardware implementation. I've
also spoken to a few classes of undergraduate CS students.
What are other folks doing to promote Forth?
Ian
P.S. to John Doty: I believe Factor is the "Forth for Python young
turks" that you are looking for. Factor is trying hard to broaden its
applicability and come with "batteries included".
John Doty
> P.S. to John Doty: I believe Factor is the "Forth for Python young
> turks" that you are looking for. Factor is trying hard to broaden its
> applicability and come with "batteries included".
I don't think so. That's more a Forth descendant for computer
scientists, not physical scientists. That's not a criticism: if Forth's
ideas are to thrive in the 21st century, I think more than one
descendant will be required. Hurray for Factor: keep the ideas alive and
growing! But Factor is not the descendant I'm looking for. One language
does not fit all applications, or even all parts of one application.
van...@vsta.org
> Ian Osgood wrote:
>> P.S. to John Doty: I believe Factor is the "Forth for Python young
>> turks" that you are looking for. Factor is trying hard to broaden its
>> applicability and come with "batteries included".
> I don't think so. That's more a Forth descendant for computer
> scientists, not physical scientists. That's not a criticism: if Forth's
> ideas are to thrive in the 21st century, I think more than one
> descendant will be required. Hurray for Factor: keep the ideas alive and
> growing! But Factor is not the descendant I'm looking for. One language
> does not fit all applications, or even all parts of one application.
I just recently downloaded Factor and played with it some. I may not agree
with every design decision you've made, but you've obviously put some
excellent work into creating a post-Forth type of language. Kudos.
Andy Valencia
John Doty
Huh? Not to me, kudos to Slava and collaborators! Keep Forth a living
idea, not an endangered species in a preserve. But I need a different
species in the Forth genus.
Peter Knaggs
> I noted with interest; "Klaus Schleisiek: Dying Forth". Would a
> summary of his argument be possible?
Here is a copy of the notes which where distributed at the time:
* issue of Marketing
* use channels that young people use
* os-news site
* retro forums
* robotics forums
* write articles for magazines
* polished/web 2.0 euroforth website
* send information on euroforth to many forums/newsgroups/websites/
* regulary (every 1-2 month)
* make sure people see we're alive
* organizer will contact a group of outreach people which will do the
marketing in their area/community
* setup a hardcore euro-forth mailing list from participants
of the past 5 years (Ulli)
* get papers in other journals / magazins (robotics ...)
* getting the young students / information into the academic world
* communities we can contact
* robotics (hobbyist)
* commercial forth users (stephen pelc)
** retro computer (Carsten Strotmann)
* OpenFirmware community (Stefan Reinauer, Carsten Strotmann)
** OLPC (One Laptop Per Child -- OpenFirmware)
* LinuxBIOS people
* Apple PPC / SUN Sparc / IBM Cell people ??
* euro forth email list (130 Emails)
* setup mailman (or similar) for mass-mailings
* how important is the 4th day?
* advanced forth trainings?
* wikipedia forth entry?
* wikipedia euro-forth entry? (Bill, Peter, rest of the group)
znmeb
Well ...
1. I just got my OLPC machine today, and I've loaded gForth 0.6.2 on
it. As far as I'm concerned, a machine with "only" 256 MB of RAM needs
Forth, and a machine that runs at "only" 433 MHz needs the assembler
in Forth. :) Besides, I really really really don't want to learn
Python.
2. I really don't know what to make of Factor. Some people I really
respect from the Ruby community, namely Zed Shaw and Pat Eyler, have
taken up Factor. On the other hand, I don't know what advantage Factor
has over ANS Forth or the coming revised Forth standard. In other
words, what's broken in Forth that Factor fixes?
3. What am I doing to promote Forth? Not much, really. When someone
asks me what language they should learn to give them a new perspective
on programming, I usually say that there are really only two languages
-- Forth and Lisp/Scheme. And most everyone has tried Lisp/Scheme. :)
Mark W. Humphries
> 3. What am I doing to promote Forth? Not much, really. When someone
> asks me what language they should learn to give them a new perspective
> on programming, I usually say that there are really only two languages
> -- Forth and Lisp/Scheme. And most everyone has tried Lisp/Scheme. :)
I add Prolog and J/APL to that list.
The Beez'
idea for this newsgroup. Instead of bashing each others heads in all
the time for (sometimes) the most trivial of discussions, we could do
what other .lang groups do best: exchange ideas and code and help
newbies.
Has anyone ever though what an impression it makes when one gets here?
We're talking ABOUT a language instead of doing it. We're so 1337. We
claim that Forth is 'an amplifier', not for occasional programmers AND
we're proud to be 'reasonably polite nowadays' when it comes to help
newbies with "stupid" questions. When I started to make an easy Forth
with - god forbid - a MANUAL people were shaking their heads. 'If
you're not smart enough to figure out the compiler source code, Forth
is not for you'. Now that is a way to attract a crowd. Wondering why
Forth is dying?
We're probably the only crowd who have exiled their creator, since 'he
obviously doesn't understand his own language'. Variations are
blasphemy - unless you've moved so far away from ANS Forth that it
becomes harmless. No, they're only a reason for another rant or
useless discussion. LSE64, StrongForth, 4tH to name a few..
And what about Forth itself? LOCALs, FILEs, quoted strings, strong
typing - I wonder why I moved from C to Forth in the first place. If
the API isn't a copy, the symbols are. Why not give the FILE wordset
their proper names, like PRINTF, FOPEN, FCLOSE?
Webpages still look like they're created by a junior student, nothing
has changed since. It makes quite a professional impression - whatever
"valid" argument you have for them. And that's the first thing people
see of Forth. If I were them I'd rather move to Befunge.
Hans Bezemer
P.S. That still won't keep me from writing, using and deploying 4tH
programs. So IMHO Forth is not that dead.
van...@vsta.org
> ... When someone
> asks me what language they should learn to give them a new perspective
> on programming, I usually say that there are really only two languages
> -- Forth and Lisp/Scheme. And most everyone has tried Lisp/Scheme. :)
Try FP, or maybe APL. The old saw is that you can write Fortran-style code
in ANY programming language... but I think FP is the exception. Its
Wikipedia entry is at:
http://en.wikipedia.org/wiki/FP_programming_language
Regards,
Andy Valencia
znmeb
Well ... the original APL was certainly unique, and was in fact a
genuine innovation in programming languages. But as far as I'm
concerned Prolog is a Lisp dialect. :)
I think J is also gone from the face of the Earth. There is something
called K which is not IIRC open source, and a dialect of APL called
APlus that is open source but not all that active.
znmeb
> We're probably the only crowd who have exiled their creator, since 'he
> obviously doesn't understand his own language'. Variations are
> blasphemy - unless you've moved so far away from ANS Forth that it
> becomes harmless. No, they're only a reason for another rant or
> useless discussion. LSE64, StrongForth, 4tH to name a few..
I'm not sure I understand what you're saying. In what way have we
"exiled" Charles Moore?
znmeb
Interesting that you should pick FP as the language you can't write
FORTRAN in, since John Backus was the inventor of both languages. But
I have little use for the modern crop of functional languages except
for Erlang. I think Erlang got functional programming right, and for
the right reasons -- to make an industrial strength platform for
highly concurrent and highly reliable applications. Most of the other
functional programming languages seem more academic, cryptic,
theoretical and devoid of any notion whatsoever of marketing. :)
Marcel Hendrix
[..]
> Webpages still look like they're created by a junior student, nothing
> has changed since. It makes quite a professional impression - whatever
> "valid" argument you have for them.
That's some complicated logic you have there. Because it's a pet
peeve of mine, I'll react to your most important comment.
Most 'modern' webpages are truly horrible. Try to make the
window you're viewing them in smaller. See what happens when
you turn off Java, Active-X, or Flash. See what happens when
people with certain disabilities want to make use of pages
(by turning off font and color changes). On more and more
webpages it is actually impossible to cut text by highlighting
and using the mouse.
Standards for the layout of scientific papers and books seem to be
going down too.
-marcel
Mark W. Humphries
integrate constructs and approaches from other languages whenever you
deem them useful for the particular application you're working on.
The Beez'
> I'm not sure I understand what you're saying. In what way have we
> "exiled" Charles Moore?
ANS Forth on more than one occasion. But how seriously do we take him
anymore? Mostly by turning his language into something he obviously
disgusts.
Hans Bezemer
van...@vsta.org
>> Try FP, ...
> ... But
> I have little use for the modern crop of functional languages except
> for Erlang.
If the criteria is "getting a job done", then I agree (I have quite a bit of
respect for Erlang). But the question was about getting a new perspective
on programming. I remember first encountering FP as an undergrad, and it
certainly turned my view of programming sideways. (Maybe I never recovered;
Forth and then FP, oh my. :->)
Andy Valencia
Anton Ertl
>2. I really don't know what to make of Factor. Some people I really
>respect from the Ruby community, namely Zed Shaw and Pat Eyler, have
>taken up Factor. On the other hand, I don't know what advantage Factor
>has over ANS Forth or the coming revised Forth standard. In other
>words, what's broken in Forth that Factor fixes?
Factor is a higher-level language. The system does a number of things
for the programmer that a Forth system does not, in particular memory
management and keeping track of types.
Is Forth broken? I don't think so. They are just different languages
that can be useful in different circumstances. I guess Factor is
better than Forth in cases where Lisp, Python or Ruby would also be
better; Forth is better than Factor in cases where C or assembly
language would also be better. And then there is quite a wide range
of problems where you can use either Factor or Forth well.
Anton Ertl
>> On Dec 25, 12:50 pm, znmeb <zzn...@gmail.com> wrote:
>> >I usually say that there are really only two languages
>> > -- Forth and Lisp/Scheme. And most everyone has tried Lisp/Scheme. :)
>But as far as I'm
>concerned Prolog is a Lisp dialect. :)
Well, reading your older message, you would say so, but Prolog differs
from Lisp a lot. I would say it differs more from Lisp than Python or
Factor differ from Lisp. Basically the only thing it has in common
with Lisp is dynamic type checking and automatic storage reclamation.
One other language you could include in your canon is a lazy
functional language such as Haskell.
Anton Ertl
>When I started to make an easy Forth
>with - god forbid - a MANUAL people were shaking their heads. 'If
>you're not smart enough to figure out the compiler source code, Forth
>is not for you'. Now that is a way to attract a crowd. Wondering why
>Forth is dying?
Nice flame bait. I'll bite: Forth is alive and kicking; nobody
forbids making a Forth system with a manual, and indeed, quite a
number of Forth systems come with manuals, and it seems to me that
they are the more popular ones.
>We're probably the only crowd who have exiled their creator
You later clarified that you were referring to the fact that Chuck
Moore chose to drop out of the ANS Forth standardization effort.
That's actually pretty common. Even Bjarne Stroustroup, who AFAIK
participated more than most language designers in the standardization
effort, refers to the C++ standards committee as "They/we"
<http://www.research.att.com/~bs/bs_faq.html#spoil-C++>. I know about
a few others:
Pascal: At the time when Pascal was stanardized (1978-1983), Wirth had
already moved on to Modula and Modula-2. I guess the difference to
Chuck Moore is that the name of his newer languages do not contain
"Pascal".
Modula-2: At the time when Modula-2 was stanardized (1987-1996), Wirth
had already moved on to Oberon. I guess the difference to Chuck Moore
is that the name of his newer language does not contain "Modula".
Prolog: ISO Prolog is based on Edinburgh Prolog, not Marseille Prolog,
Prolog-2, or Prolog-3 (the systems that Alain Colmerauer worked on).
C: Dennis Ritchie refers to the C99 standards committee as "they", and
is quite critical of this standard
<http://www.itworld.com/Comp/3380/lw-12-ritchie/>.
>since 'he
>obviously doesn't understand his own language'.
I think he has stated himself that he is not interested in a standard;
he does not need one.
William James
> Most 'modern' webpages are truly horrible. Try to make the
> window you're viewing them in smaller. See what happens when
> you turn off Java, Active-X, or Flash.
Yes. The mindless masses think that a webpage should be
extremely cluttered and flashy. If it's as plain and
simple and easy to handle as a page in a book, then it's
considered pitifully primitive and unsophisticated.
Make sure it has plenty of frames, blinking text,
animations, drop-down menus, pop-up windows, background
pictures that make the text unreadable, etc. The mind
must be stunned and stupified or actual thinking may
occur.
John Doty
It's not the mindless masses that are responsible: it's the people who
design web pages to SELL, SELL, SELL! They'd turn the whole world into
an imitation of Las Vegas or Shinjuku if they could. With computer power
and no legal limitations on bad web aesthetics, it's much easier to do
this with Flash than with neon.
The destructive effect Powerpoint has had on presentations is related.
It's so easy to tart up ideas there that people do that and forget that
the object is to communicate.
The Beez'
> Nice flame bait.
Sorry, I don't need the attention. It's just an opinion, whether you
agree with it or not. Don't call me names; it's not nice.
> You later clarified that you were referring to the fact that Chuck
> Moore chose to drop out of the ANS Forth standardization effort.
> That's actually pretty common. Even Bjarne Stroustroup, who AFAIK
> participated more than most language designers in the standardization
> effort, refers to the C++ standards committee as "They/we"
> <http://www.research.att.com/~bs/bs_faq.html#spoil-C++>. I know about
> a few others:
"Did the ANSI/ISO standards committee spoil C++?
No. THEY/WE DID A GOOD JOB. You can quibble with details (and I do,
sometimes loudly), but I'm happy with the language and the new
standard library. ISO C++ is a better and more coherent language than
earlier versions of C++. You can write much more elegant and
maintainable C++ programs today than was possible when the standards
process started. The new standard library is also a real boon."
> C: Dennis Ritchie refers to the C99 standards committee as "they", and
> is quite critical of this standard
> <http://www.itworld.com/Comp/3380/lw-12-ritchie/>.
"I'm less ecstatic about the C99 standard, but don't denounce it. THEY
DID A PRETTY GOOD JOB; C does have to evolve. I was not involved with
its work, but was given opportunities to snipe or contribute earlier.
So I won't do much second-guessing after the fact."
> I think he has stated himself that he is not interested in a standard;
> he does not need one.
True, but he still has ideas about it. Compare the above with:
"I had reservations about ANSI. I worried that it would be a disaster
and not merely a dubious advantage. All of my fears of the standard
and none of the advantages of the standard have come to pass. Any
spirit of innovation has been thoroughly quelched. Underground Forths
are still needed. I said I thought the standard should be a
publication standard but they wanted an execution standard."
"I'm rewriting the standard to suite my taste and I hope it will be
much simpler and it will convey much of the same information and some
that was not previously available. Another thing I don't like is that
the internals are exposed. There is far too much discussion of words
that an application programmer will never use. Particularly if that
application programmer doesn't decide to write their own interpreter.
I don't like people writing their own inpreters because that means
that as a user I have to learn a different interpreter. Forth has a
perfectly good interpreter if you will just accept the word order. So
words like WORD STATE and HOLD and another dozen or two, not that they
shouldn't be in the standard, they shouldn't be in the CORE wordset.
They should be de-emphasized by the standard or there should be
another standard appropriate for application programmers and this one
is for systems programmers.
And then the double words. Get rid of all the double words. We don't
need to do 64 bit arithmetic. Machines now are 32 bits or at least 20
bits and in no case are double precision arithmetic relevant. And it
it complexifies things horribly. If you look at these arithmetic
operators: UM*/MOD or something, I mean what in God's name does that
do and why should I care? And you shouldn't. It can be gone. U/MOD is
too complex. / certainly. */ yes. MOD maybe, but that is about the end
of it.
That includes words like 2OVER 2SWAP 2@ and 2! You oughtn't use such
words. They make your stack more complicated. They are making
presumptions about the layout of data in memory which you probably
shouldn't be doing. It is just too complicated.
If you want an extension wordset with all these double arithmetic
operators fine. If you want an extension wordset with all the internal
necessary to write your own interpreter fine, but not the CORE
wordset."
"The standard has everything built into the CORE wordset if it were
factored a little bit better it would be easier to understand and
provide better versitility. ---"
I even got a webpage for you: http://www.ultratechnology.com/moore4th.htm
But you probably know that one by now.
Hans Bezemer
Bernd Paysan
> On 25 dec, 08:21, znmeb <zzn...@gmail.com> wrote:
>> I'm not sure I understand what you're saying. In what way have we
>> "exiled" Charles Moore?
> Be real: CM was on the ANS Forth committee and has vented his ideas on
> ANS Forth on more than one occasion. But how seriously do we take him
> anymore?
I think Chuck's mistake in that time (early 90s) was to go to the sourceless
system, OK. He has recovered from that mistake, and got to ColorForth by
that detour. I'd say Chuck exiled himself, by abandoning Forth as a
language and using it only as an instruction set. He's back now, at least
mostly (ColorForth still is obviously influenced by the "Forth as
instruction set" idea, but it's also a language with text source).
> Mostly by turning his language into something he obviously disgusts.
I think it's more than just the language. He disgusts the environment -
bloated PCs with bloated instruction sets, bloated operating systems, and
running bloated Forth systems on top of all that bloat.
Marcel Hendrix
[..]
> True, but he still has ideas about it. Compare the above with:
> "I had reservations about ANSI. I worried that it would be a disaster
> and not merely a dubious advantage. All of my fears of the standard
> and none of the advantages of the standard have come to pass. Any
> spirit of innovation has been thoroughly quelched. Underground Forths
> are still needed. I said I thought the standard should be a
> publication standard but they wanted an execution standard."
[..]
> And then the double words. Get rid of all the double words. We don't
[..]
( http://findarticles.com/p/articles/mi_m0EIN/is_2006_May_1/ai_n16135032 )
--quote Intellasys blurb--
Moore will describe the SEA Platform in detail and how it promises
to enable a host of highly compact embedded applications, including
distributed digital media processing, which must operate at high
speed and low power. He will also discuss the attributes of his
firm's proprietary VentureForth(TM) programming language, a RISC
version of the well-established Forth software he created nearly
three decades ago.
As the native machine code for the SEA Platform chips, VentureForth
provides simple but elegant instructions, including full
18 x 18 bit multiplies with 36-bit results. This powerful version
^^^^^^^ ^^^^^^
of ANSI Forth frees designers from laboring over thousands of lines
of assembly code.
--end quote--
This doesn't seem to say that ANSI Forth is bad, nor that doubles
should be abandoned.
-marcel
Andrew Haley
> [ CM said ] I said I thought the standard should be a publication
> standard but they wanted an execution standard."
Does anyone here understand how this would work? How is it possible
to make a standard suitable for the publication of code but not its
execution?
Andrew.
The Beez'
> "The Beez'" <hans...@bigfoot.com> writes Re: Promoting Forth (Re: EuroForth 2008 workshop "Forth is dying")
36 != 64.
Hans
astrobe
> I think Chuck's mistake in that time (early 90s) was to go to the sourceless
> system, OK. He has recovered from that mistake, and got to ColorForth by
> that detour.
So it wasn't a mistake, it was an experiment. He didn't "got
Colorforth by that detour"; Sourceless programming was a dead end, but
by experimenting with it, he found something usable and efficient.
> I'd say Chuck exiled himself, by abandoning Forth as a
> language and using it only as an instruction set. He's back now, at least
> mostly (ColorForth still is obviously influenced by the "Forth as
> instruction set" idea, but it's also a language with text source).
>
I don't think so. According to himself, all the software problems wre
solved, the remaining problems were in hardware, so he decided to fix
that problems.
IMHO, part of the frustation people feel about Moore is that he is
virtually no more a software guy, he became a hardware guy. The ideas
an improvements he is working on are less and less ogten appropriate
for PC-based Forth.
He didn't exile himself, or he has not been exiled by us, he just
moved.
And true, he is back, but with SEAForth sticks.
> > Mostly by turning his language into something he obviously disgusts.
>
> I think it's more than just the language.
Yes, it's the fundamental idea of a standard Forth that attempts to be
compatible with most of what have been done in the last 10 years. For
a man always looking forward it couldn't possibly work.
Amicalement,
Astrobe
Coos Haak
36 18 2* =
--
Coos
CHForth, 16 bit DOS applications
http://home.hccnet.nl/j.j.haak/forth.html
Anton Ertl
Well, one answer would be that the standard is for a non-executable
language, e.g., a specification language.
However, since Forth is an executable language, my guess is that he
had something like Algol 60 in mind: Algol 60 has no machine-readable
source code representation, so the representation used by one compiler
usually could not be compiled by a different compiler. It also did
not have any standard I/O, so one could not write any standard
programs that did anything different from an empty program when
executed.
As a result, Algol 60 as standardized was only useful for publication:
You don't need I/O to explain the code for most algorithms; and the
on-paper representation was more standardized and human readers are
more forgiving of little differences than compilers.
Hmm, there are those who say that the standard's purpose is not to
port programs, but to port programmers, so at least some did not want
an execution standard, either.
I am firmly in the camp that wants an execution standard. I find it
great that I can run the same program unchanged on four different
Forth systems, like I did yesterday.
foxchip
> So it wasn't a mistake, it was an experiment. He didn't "got
> Colorforth by that detour"; Sourceless programming was a dead end, but
> by experimenting with it, he found something usable and efficient.
Sourceless programming also led to the aha experiment and has
led to novel ways of generating code on new targets. The experiments
in blurring the line between source and object code, when there
are links between and editor and compiler have led to useful
knowledge.
Also the sourceless era was somewhat of a protest against the
exclusion from ANS. He 'abandoned' Forth somewhat in protest to ANS
having excluded him.
> I don't think so. According to himself, all the software problems wre
> solved, the remaining problems were in hardware, so he decided to fix
> that problems.
Yes, 1980. Chuck was saying that the 'software problem' was solved
by using Forth in a one-language environment. All the remaining
problems were due to hardware not being designed to assist in these
sort of simple solutions.
> IMHO, part of the frustation people feel about Moore is that he is
> virtually no more a software guy, he became a hardware guy. The ideas
> an improvements he is working on are less and less ogten appropriate
> for PC-based Forth.
Certainly he had been a valuable software resource at Forth Inc. and
was seen as moving to hardware when he made what he considered a
lateral move between Forth Inc. and Novix since both companies were
owned by John Golden at the time. This was the beginning of the
great split.
Chuck continued to work with Forth software, but to design hardware.
Forth Inc. abandoned the Forth OS side of Forth on the PC and followed
much of the world into Bill's Windows market environment. This was
perhaps a bigger split between Chuck's ideas of Forth as a one-
language
environment designed to avoid all those multi-lingual problems and
Forth as one of many langauges in a vast and pervasive multi-language
environment.
> He didn't exile himself, or he has not been exiled by us, he just
> moved.
> And true, he is back, but with SEAForth sticks.
>
> > > Mostly by turning his language into something he obviously disgusts.
>
> > I think it's more than just the language.
>
> Yes, it's the fundamental idea of a standard Forth that attempts to be
> compatible with most of what have been done in the last 10 years. For
> a man always looking forward it couldn't possibly work.
He had supported the idea of an ANS standard, but felt it should be
more of a publication standard and open to the future rather than
fixating on the past. Once he was was rejected from the process
he questioned the direction of the process.
He suggested what he had done in the 80s but it was rejected as not
what the vendors were selling at the time or what they intended to
sell in the future. By the time the standard was ratified in 94 it
has standardized on what Chuck had been doing twenty years earlier
it had rejected ideas that he introduced in cmForth and machineForth.
It was already almost fifteen years behind the improvements he had
been
making in the implementation of the improved virtual machine and
simplfied compiler that led to colorforth.
Best Wishes
John Rible
> ...
>
> Certainly he had been a valuable software resource at Forth Inc. and
> was seen as moving to hardware when he made what he considered a
> lateral move between Forth Inc. and Novix since both companies were
> owned by John Golden at the time.
That was John Peers, a consummate seller of sizzle.
> ...
Aleksej Saushev
> Op Wed, 26 Dec 2007 02:28:12 -0800 (PST) schreef The Beez':
>
>> 36 != 64.
>>
> 36 18 2* =
It doesn't show anything, except one can conclude that you're real Forth geek.
--
CKOPO CE3OH...
Bruce McFarling
36 18 2* = .
-1 OK is what gforth tells me ... I expect this is saying that in an
18bit Forth, a 36bit result on the stack occupies two cells ... that
is, is a double result?
However, if the result is going to be store, I expect that rather than
storing it with 2! ... it would be stored with !A+ !A
Brad Eckert
> By the time the standard was ratified in 94 it
> has standardized on what Chuck had been doing twenty years earlier
> it had rejected ideas that he introduced in cmForth and machineForth.
> It was already almost fifteen years behind the improvements he had
> been
> making in the implementation of the improved virtual machine and
> simplfied compiler that led to colorforth.
I would have liked to see Chuck go off and publish his own standard.
He's too nice a guy to have done that at the time, but things have
cooled off since then. I suppose colorForth is the closest thing we
have to Chuck's vision of Forth, but it's a Forth tailored to his own
requirements. If he were to specify a standard for the Forth community
at large it might be different. For example, it's hard to find an off
the shelf text editor that supports colorForth source code.
What would colorForth be without colors? I mean, if it were
implemented to compile text source files instead of blocks and omitted
magenta variables and had a way to do without tagging keywords with
colors (say, with [ and ]) would it still have some of the
productivity of colorForth?
What is it about colorForth that makes it more productive than ANS?
Brad
znmeb
wrote:
Well ... I would if I *liked* lazy languages. But I don't. :)
Bernd Paysan
But since SEAForth is a 18 bit machine, 36 bit are doubles there. On a 16
bit Forth, doubles are 32 bit, and on a 64 bit Forth doubles are 128 bit.
Elizabeth D Rather
Why? That cell size is hardly specific to addresses. On an 18-bit
Forth, single integers will be 18 bits, and doubles 36 bits. This is a
great example of why the 2-prefix and D-prefix operators are far more
useful than operators with specific bit-ranges.
Cheers,
Elizabeth
--
==================================================
Elizabeth D. Rather (US & Canada) 800-55-FORTH
FORTH Inc. +1 310-491-3356
5155 W. Rosecrans Ave. #1018 Fax: +1 310-978-9454
Hawthorne, CA 90250
http://www.forth.com
"Forth-based products and Services for real-time
applications since 1973."
==================================================
Bruce McFarling
> Bruce McFarling wrote:
> > 36 18 2* = .
> > -1 OK is what gforth tells me ... I expect this is saying that in an
> > 18bit Forth, a 36bit result on the stack occupies two cells ... that
> > is, is a double result?
> > However, if the result is going to be store, I expect that rather than
> > storing it with 2! ... it would be stored with !A+ !A
> Why? That cell size is hardly specific to addresses. On an 18-bit
> Forth, single integers will be 18 bits, and doubles 36 bits. This is a
> great example of why the 2-prefix and D-prefix operators are far more
> useful than operators with specific bit-ranges.
That's just the impression I got from the last Forth sources I saw
from Chuck Moore.
Aleksej Saushev
> Aleksej Saushev wrote:
>
>> Coos Haak <chf...@hccnet.nl> writes:
>>
>>> Op Wed, 26 Dec 2007 02:28:12 -0800 (PST) schreef The Beez':
>>>
>>>> 36 != 64.
>>>>
>>> 36 18 2* =
>>
>> It doesn't show anything, except one can conclude that you're real Forth
>> geek.
>
> But since SEAForth is a 18 bit machine, 36 bit are doubles there. On a 16
> bit Forth, doubles are 32 bit, and on a 64 bit Forth doubles are 128 bit.
I meant, it doesn't print result.
--
CKOPO CE3OH...
Coos Haak
If you had one smidgen of Forth knowledge, you used a decimal point after
the expression. It would print alright. Don''t flatter yourself with being
a Forth nitwit. (Dutch: weet niet)
As the othere respondents said: 36 bit is a double on SEAForth.
Bruce McFarling
> I meant, it doesn't print result.
The result is on the stack, of course. If you want it printed, you
would do:
. <return-key>
... which doesn't seem too strenuous. But maybe you'd want to use it
for something else, like defining the constant TRUE on a system that
doesn't have one ...
DECIMAL
36 18 2* = CONSTANT TRUE
Aleksej Saushev
> On Dec 29, 2:18 pm, Aleksej Saushev <a...@inbox.ru> wrote:
>> I meant, it doesn't print result.
>
> The result is on the stack, of course. If you want it printed, you
> would do:
>
> . <return-key>
Oh, really?
> ... which doesn't seem too strenuous. But maybe you'd want to use it
> for something else, like defining the constant TRUE on a system that
> doesn't have one ...
>
> DECIMAL
> 36 18 2* = CONSTANT TRUE
Yeah! I can even tell
FALSE CONSTANT TRUE
It may complain or not, but it will do.
--
CKOPO CE3OH...
foxchip
> I would have liked to see Chuck go off and publish his own standard.
He said he would have liked a publication standard. I doubt if it
would have covered ANS extensions. The core would be reduced.
I think that means it might have been closer to a programmer's
reference card with agreed upon meanings for a list of words. His
idea was to just be able to post code that people could understand.
Since he does provide access below standard @ and ! I would expect
!+ etc. I would expect -IF and its associated high level constructs
in addition to IF and variations. 20 year old stuff instead of the
40 year old stuff. Smaller than ANS but with some critical
additions and updates to very old words like BLOCK.
He really hasn't talked about what he would offer since the
went the colorforth route. So he might be inclined to define
some colors or tagged classes of words. If so I would expect
that he would suggest standardizing on red for define, green
for compiler, yellow for interpret, magenta for variable etc.
But I don't recall him ever talking about any ideas about
colorforth standards.
Other people have tried things that didn't quite match what
Chuck was doing. Shawn made macro wordlist words like IF yellow
while Chuck makes them green even though they get sort of yellow
like behavior when compiled in green. I suspect that Chuck would
suggest the more traditional approach that he used for green
and yellow. On a black background the colors associated with
standard Forth words would be
( white
: red
] green
[ yellow
variable magenta
postpone cyan
green [ yellow stack-action green ] literal
Since he doesn't use wordlists I wouldn't expect any of that
stuff. Simple methods to keep things simple. The same name
might appear in several apps at once, but in a single-
language environment he doesn't need wordlists like ANS.
> He's too nice a guy to have done that at the time, but things have
> cooled off since then.
If publishing a standard makes one a bad guy then perhaps so. ;-o
> I suppose colorForth is the closest thing we have to Chuck's
> vision of Forth,
It is part of his vision, the Pentium details are a bootstrapping
distraction. His vision involves hardware and software and how
the two interact with human programmers. He says you can't
discuss software without a hardware context.
> but it's a Forth tailored to his own requirements.
He said it was as general purpose as any Forth. He defines general
purpose in terms of potential apps not in terms of general purpose
OS. What you are saying is the opposite of what he said.
> If he were to specify a standard for the Forth community
> at large it might be different. For example, it's hard to find an off
> the shelf text editor that supports colorForth source code.
It is also hard to find colorforth on most shelves.
It is not hard to find a colorforth editor in colorforth. ;-)
Of course I always say that Forth programmers are programmers and
can write the editor they need. Editors are pretty trivial
things.
> What would colorForth be without colors?
Updated virtual machine, updated compiler, tagged code. The
tags mean something to the compiler and something to the editor.
> I mean, if it were
> implemented to compile text source files instead of blocks and omitted
> magenta variables and had a way to do without tagging keywords with
> colors (say, with [ and ]) would it still have some of the
> productivity of colorForth?
>
> What is it about colorForth that makes it more productive than ANS?
There are a number of things colorforth does to try to be more
productive. I could certainly enumerate them but I think your
question was rhetorical. And in the end the assumed context around
the word 'productive' makes all the difference.
Chuck said Forth was created to escape the multi-language nightmare
and present a simple one-language OS, compiler, tools and application
environment to let the programmer focus on a programming task with
fewer distractions present in multi-language environments. Chuck
has always had a vision of Forth as a simple one-language environment
and for half the history of Forth a vision of it providing easy to
program in Forth hardware designs too. They can be smaller, simpler,
cheaper, and lower power than hardware to tackle more difficult
multi-language problems just as Forth software can be smaller etc.
The multi-language environments have evolved into having many more
languages than when Chuck called it the multi-language nightmare
fourty years ago. Using a tool designed to escape that box is
probably not productive to solve the multi-lingual problems
that come up inside that box. If one assumes that context in
the definition of 'productive' then Chuck's idea of avoiding
all that doesn't apply.
Other people use ANS Forth inside the box, or even colorforth
inside the box. I use ANS Forth, machineForth and colorforth.
I use colorforth stand-alone and colorforth hosted where I
can use OS file access words etc. I find using colorforth
to be quite different than using other things.
I was thinking the other day that what I happened to be doing
felt a lot more like Tetris than thoughtful programming. Maybe
I had just gone through the stage of thinking about the coding
options in the system long enough that it moved to a more
reflexive level of operation in my thought but it was also
about simple code and the colored interface.
After writing and testing some code it became a visually apparent
to me that I could take this colored block and turn it sideways
and it would cancel out some other colored blocks and I could
delete it and have a smaller pile of blocks at the bottom of
the screen. And that made it visually obvious that I could do
it again and cancel out some more code and have a smaller pile
of colored code blocks accumulated inside this section of the
application. And I might have done it again before I moved on.
I noted that it was more like playing Tetris than going through
the kind of thought processes that I go through in machineForth
or ANS Forth programming. While I am describing how it felt
I am sure that a catscan would have shown that the process
was more like playing Tetris than programming, at least at
that time I described it feeling like Tetris.
Can fun be measured? Maybe with a catscan it can detected but
probably not measured. I don't know if you consider 'fun' to
be part of the spec for Forth and related to productivity.
But I will report that I find playing Tetris with colorforth
code to be a lot of fun. I don't know in which environment
I am most productive.
Best Wishes