Wolfram and "general 2D simulators"
George Maydwell
From page 928 of ANKOS: "But until after my work in the 1980's
simulators for more general 2D cellular automata were rare."
I'd say such simulators are still rare, although post 1980's
developments such as the internet and platform standardization on
Java/Windows have improved the situation. In future generations
readers of Mr. Wolfram's tome may not realize that computers
themselves were rare until the late 1980's at earliest.
Wolfram seems to be implying that his work has been a significant
influence on the development of general 2D CA simulators. Does anyone
know of anything he might have done to give credence to this claim?
For the record, I've written my share of "more general 2D" simulators,
some of which are pretty general. The Cellular Automata Machines book
by Toffoli and Margolus (published perhaps coincidentally in the
1980's) has always been my inspiration. Wolfram has not influenced me
at all.
Did I somehow miss a flood of "general 2D cellular automata" software
in the 1980's, resulting from the influence of Wolfram's work? Or is
Wolfram perhaps referring to general 2D cellular automata support in
Mathematica, a program I can't afford to use to play with cellular
automata?
George Maydwell
--
Modern Cellular Automata: www.collidoscope.com/modernca
Collidoscope Hexagonal Screensaver: www.collidoscope.com
Erik Max Francis
> Wolfram seems to be implying that his work has been a significant
> influence on the development of general 2D CA simulators. Does anyone
> know of anything he might have done to give credence to this claim?
I shudder to think at the possibility that Wolfram is talking in
self-aggrandizing terms without backing up the claim! Who'd a-thunk it?
:-)
> For the record, I've written my share of "more general 2D" simulators,
> some of which are pretty general. The Cellular Automata Machines book
> by Toffoli and Margolus (published perhaps coincidentally in the
> 1980's) has always been my inspiration. Wolfram has not influenced me
> at all.
I've also written several, also long before I'd ever heard of Wolfram.
Quite frankly, cellular automata simulators _themselves_ were not common
until later; their modest increase in popularity can simply be explained
by more programs of all kinds being written. I don't see any reason to
believe that the increase in CA simulators, much less general ones, had
anything to do with Wolfram.
--
Erik Max Francis / m...@alcyone.com / http://www.alcyone.com/max/
__ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/ \ See the son in your bad day / Smell the flowers in the valley
\__/ Chante Moore
Bosskey.net: Aliens vs. Predator 2 / http://www.bosskey.net/avp2/
A personal guide to Aliens vs. Predator 2.
Tim Tyler
: From page 928 of ANKOS: "But until after my work in the 1980's
: simulators for more general 2D cellular automata were rare."
[...]
: Wolfram seems to be implying that his work has been a significant
: influence on the development of general 2D CA simulators. Does anyone
: know of anything he might have done to give credence to this claim?
You could read it like that - but it seems like more a statement of
chronology than anything else.
You could also interpret it as: I wrote all those papers about
1D CA before there were really any proper tools to explore
2D CA with.
--
__________
|im |yler http://timtyler.org/ t...@tt1.org
George Maydwell
>George Maydwell <geo...@collidoscope.com> wrote:
>
>: From page 928 of ANKOS: "But until after my work in the 1980's
>: simulators for more general 2D cellular automata were rare."
>
>[...]
>
>: Wolfram seems to be implying that his work has been a significant
>: influence on the development of general 2D CA simulators. Does anyone
>: know of anything he might have done to give credence to this claim?
>
>You could read it like that - but it seems like more a statement of
>chronology than anything else.
The quote seems ambiguous at best. A clear statement of chronology
would read simply "But when I did my work in the 1980's simulators for
more general 2D cellular automata were rare". I apologize for not
previously noting that the quote was plucked from the history section
of the notes for chapter five, "Two Dimensions And Beyond".
From the history of cellular automata section of the notes for chapter
two of ANKOS: "The publication of my first paper on cellular automata
in 1983 (see page 881) led to a rapid increase of interest in the
field, and over the years since then a steadily increasing number of
papers (as indicated by the number of source documents in the Science
Citation Index shown below) have been published on cellular automata -
almost all following the directions I defined". Besides not parsing
well, this new quote seems more than a statement of chronology.
Could it indeed be true that almost all articles published about
cellular automata followed directions Wolfram defined? Did he write
down a set of instructions for writing cellular automata papers? Fun
aside, can anyone familiar with the literature independently verify
the "almost all following the directions I defined" phrase?
>
>You could also interpret it as: I wrote all those papers about
>1D CA before there were really any proper tools to explore
>2D CA with.
Except for that the very next sentence says "A sequence of hardware
simulators were nevertheless built starting in the mid-1970s by
Tommaso Toffoli and later Norman Margolus". Why does Wolfram use the
word "nevertheless" here? He seems to be implying that researchers in
the 1970's should have known better than to pursue an avenue of
investigation because of knowledge discovered in the 1980's.
Richard Fateman
PhD dissertation the universality of various 2D cellular automaton.
There was considerable activity in writing programs for the game of
Life, prior to that. I think that anyone with even a passing
interest in CA would very rapidly decide that
a computer simulation would be helpful, before or after 1980.
In 1970, Stephen Wolfram was 11 years old.
I think that the quoted statement is just a minor example
of SW's monumental immodesty.
By the way, does anyone else think he looks sort of like
George Costanza (from Seinfeld TV show)?
http://www.movieprop.com/tvandmovie/Seinfeld/george.htm
Tim Tyler
: In 1970, Roger Banks (Edwin Roger Banks) proved in his MIT
: PhD dissertation the universality of various 2D cellular automaton.
: There was considerable activity in writing programs for the game of
: Life, prior to that.
Hmm - not by very much. The Game of Life was invented in 1970, and
I believe the interest in it mainly followed after the Scientific American
(October 1970) article on it by Martin Gardner:
Of course this is still 12 years before Wolfram's first paper on the subject.
Stephen Harris
"George Maydwell" <geo...@collidoscope.com> wrote in message
news:3d3bd175...@news.prodigy.net...
> On Sat, 20 Jul 2002 19:53:49 GMT, Tim Tyler <t...@tt1.org> wrote:
>
> >George Maydwell <geo...@collidoscope.com> wrote:
> >
> From the history of cellular automata section of the notes for chapter
> two of ANKOS: "The publication of my first paper on cellular automata
> in 1983 (see page 881) led to a rapid increase of interest in the
> field, and over the years since then a steadily increasing number of
> papers (as indicated by the number of source documents in the Science
> Citation Index shown below) have been published on cellular automata -
> almost all following the directions I defined".
This is correct and if anything is understated. Before he wrote his
classifying paper the field of cellular automata was dormant. Wolram
single-handedly revitalized this area. Nearly everybody cited his paper
and used his reference scheme. Fredkin was known for supporting
the universe as a digital computer even then, but Wolfram was Mr. CA.
I started reading occasionally about CA in 1987. The other big
name was Conway and the Life game that Gardner promoted in
Scientific American which held a lasting recognition.
If you remove Conway, not all the other people added together would
equal Wolfram's influence because the influence of the old names had
disappeared...CA was a nearly dead field. For the last 20 years most
educated people could match CA to Wolfram not to anyone else.
And that is where public funding comes from and attracting new players
to the game. Not the few people who would know who Grothendieck is
and who would also know von Neumann's side interest.
>Besides not parsing
> well, this new quote seems more than a statement of chronology.
> Could it indeed be true that almost all articles published about
> cellular automata followed directions Wolfram defined? Did he write
> down a set of instructions for writing cellular automata papers? Fun
> aside, can anyone familiar with the literature independently verify
> the "almost all following the directions I defined" phrase?
>
No, they used his 4-part classification scheme which was standard.
This is obvious if you know any history of CA. I am aware that you
produce working CA. What gets me is you guys who claim that
Wolfram is arrogant. How much arrogance does it take for you to
criticize Wolfram who made CA history? You who know less
about this than some 6th graders! They could answer this question:
GM wrote:
"Fun aside, can anyone familiar with the literature independently verify
the "almost all following the directions I defined" phrase?"
Your reading comprehension is poor. Sentences need to be understood
in context. Wolfram means that he moved CA in a new direction or new
approach with his 4 part classification scheme. This 4 part scheme is
the thing Wolfram is most famous for. The people who wrote about
CA nearly all adopted this 4 part classification scheme he invented.
No, Wolfram did not write a bunch of rules that everybody had to
follow in order to write a CA paper. How come you could think of
this as a logical meaning? I mean you had to stretch so far to come
up with this. And obviously you know nothing about the history of
CA or Wolfram's contributions. So where did you get the nerve
to create a fictional interpretation of history? The question is rhetorical.
You joined the bandwagon of the other arrogant, envious wannabes.
Please feel free to killfile my posts if you dont like this one. I dont
want to waste my time on your ilk anyway. No wonder Wolfram
didn't want to wallow in the marketplace.
James Dolan
stephen harris <stephen....@worldnet.att.net> wrote:
|For the last 20 years most educated people could match CA to Wolfram
|not to anyone else.
you are a liar or an idiot.
--
[e-mail address jdo...@math.ucr.edu]
George Maydwell
Sorry for the misunderstanding. This statement of mine was intended as
a joke.
>up with this. And obviously you know nothing about the history of
>CA or Wolfram's contributions.
I freely admit to my own ignorance. Why else would I ask questions?
Thanks for sharing your own opinions and valuable historical
perspective here in this newsgroup.
>So where did you get the nerve
>to create a fictional interpretation of history? The question is rhetorical.
>You joined the bandwagon of the other arrogant, envious wannabes.
You joke or troll here, right? You are of course quite welcome to
consider me arrogant (although I find this assertion ironic in a
thread about Wolfram) but I dispute creating a fictional
interpretation of history or being an envious wannabe. Wolfram writes
papers and books, I create eye candy. I'm pretty far from being an
envious Wolfram wannabe. Asking questions about what Wolfram has
written hardly constitutes creating a fictional interpretation of
history. Nerve is needed here only to maintain a veneer of civility
when responding to the posts of others.
Tim Tyler
: Could it indeed be true that almost all articles published about
: cellular automata followed directions Wolfram defined? Did he write
: down a set of instructions for writing cellular automata papers? Fun
: aside, can anyone familiar with the literature independently verify
: the "almost all following the directions I defined" phrase?
Wolfram's papers from the era might easily be characterised as
"having a theme".
In particular, you can regard:
http://www.stephenwolfram.com/publications/articles/ca/85-twenty/
...as laying out a research program.
Whether people followed it is another question.
For example, Wolfram didn't have much to say about self-reproduction.
Codd, Langton, von Neumann et al hardly seem to intersect with Wolfram's
universe at all.
Nor did Wolfram have much to say about biological modelling and artificial
life. Spatialised prisoner dilemma work, Avida, and a range of models of
viscous populations:
(e.g. http://www.zoo.cam.ac.uk/zoostaff/matt/Paper10.html)
...don't seem to do so much as throw a sideways glance to Wolfram.
The Game Of Life has remained popular. I doubt enthusiasts view
their work as owing much debt to Wolfram.
Also, relatively few people seem to have become very interested in the
1D CAs that facinated Wolfram. We have 2D hardware, 2D displays and 2D
retians - and live in a 3D world. The overall impression is that 1D CAs
are generally regarded as of reduced interest.
I'd identify the academic nature of Wolfram's work as contributing to
the general lack of interest. Almost anyone is immediately intrigued by
Conway's work. By contrast, it takes a special sort of person to see the
appeal of Wolfram's 1D CAs, and the academic papers he wrote about them.
I think Wolfram himself says at one point that he was disappointed with
how little his "program" was followed up on by others - and that if he
wanted to see things get done he was obviously going to have to do it
himself.
Tim Tyler
: "George Maydwell" <geo...@collidoscope.com> wrote in message
:> On Sat, 20 Jul 2002 19:53:49 GMT, Tim Tyler <t...@tt1.org> wrote:
:> >George Maydwell <geo...@collidoscope.com> wrote:
:> From the history of cellular automata section of the notes for chapter
:> two of ANKOS: "The publication of my first paper on cellular automata
:> in 1983 (see page 881) led to a rapid increase of interest in the
:> field, and over the years since then a steadily increasing number of
:> papers (as indicated by the number of source documents in the Science
:> Citation Index shown below) have been published on cellular automata -
:> almost all following the directions I defined".
: This is correct and if anything is understated. Before he wrote his
: classifying paper the field of cellular automata was dormant.
That does not appear to be very accurate.
We have Fredkin since the year dot, The Theory of self-reproducing
Machines in 1966, Codd in 1968, Conway in 1970, Zhabotinsky in 1970,
Burks from 1970, Kimura from 1976-1979, Toffoli, papers spanning
1977-1984, Conway, Guy and Berlekamp's famous proof from 1982,
Self-replicating loops from Christopher Langton in 1984, Margolus
in 1984. Interest in cellular automata was on the up.
Mandelbrot published in 1982 - and soon it seemed that chaotic
systems were suddenly fashionable.
There *was* an explosion of interest in CAs in the early 1980s - but the
factor was rather obviously cheap home computers - and not Wolfram.
: If you remove Conway, not all the other people added together would
: equal Wolfram's influence because the influence of the old names had
: disappeared... CA was a nearly dead field.
Old names like von Neumann, I suppose - who practically invented the field.
I'm not sure how von-Neuman's influence could ever die away - while
the field still exists.
:> Besides not parsing well, this new quote seems more than a statement
:> of chronology. Could it indeed be true that almost all articles
:> published about cellular automata followed directions Wolfram
:> defined? Did he write down a set of instructions for writing cellular
:> automata papers? [...]
: No, they used his 4-part classification scheme which was standard.
: This is obvious if you know any history of CA.
This classification scheme has been subject to criticism:
``Wolfram's classification may then be worse than useless: it may be
ill-defined, since there is no quick test for distinguishing class 4
rules from other rules.''
- http://www.ics.uci.edu/~eppstein/ca/wolfram.html
``The fundamental difficulty here is that the "Wolfram Classes" are not
well defined, even by Wolfram himself. Taken at their face value, it has
been proven that they can't be; in articles by Culick III, as I recall.''
Harold V. McIntosh's comment, from http://cafaq.com/classify/
Essentially, I agree with the critics. Classifying behaviour based
on the befaviour of typical random configurations is not very useful -
since significant properties (such as universal computation and universal
construction) depend on the details of the behaviour of particular
structures - and not on what happens during the evolution from "typical"
random starting configurations.
: What gets me is you guys who claim that Wolfram is arrogant. How much
: arrogance does it take for you to criticize Wolfram who made CA
: history? You who know less about this than some 6th graders!
Unfortunately, Wolfram seems to have provided no shortage of ammunition.
For example, I'm sure many sixth-graders would be quite capable of scoring
a few cheap hits. Something about how commercial interests peverted
the process of scientific discovery, maybe.
Stephen Harris
"James Dolan" <jdo...@math-lw-n01.math.ucr.edu> wrote in message
news:ahd190$606$1...@glue.ucr.edu...
> in article <1gn_8.42183$iu6.2...@bgtnsc04-news.ops.worldnet.att.net>,
> stephen harris <stephen....@worldnet.att.net> wrote:
>
> |For the last 20 years most educated people could match CA to Wolfram
> |not to anyone else.
>
> you are a liar or an idiot.
>
>
college degree in various fields such as English, Psychology, Sociology
Anthropology, etc., peripheral science majors outside of math and physics
a multiple choice question like: Who is known for his substantial
achievements
in the field of cellular automata?
a. Gacs
b. Fredkin
c. Wolfram
d. Von Neumann
And I think at least 75% would choose Wolfram. I think next would
be Von Neumann because that would be a best guess. I think Gacs
who has probably done the most work would probably be near last.
In an article or paper about CA, Wolfram is nearly always mentioned.
Did you interpret "educated" to mean just math majors?
Do you think "educated" just means people who know the right answer?
If you think Wolfram is not who most educated people would choose
then who is your candidate? I've lazily followed this field for the last
15 years, have you? Just what name have you seen more frequently
than Wolfram's? You are yet just another pedantic academic who
lacks critical thinking skills, such as pointing out imaginary errors in
reviews.
Stephen Harris
"Tim Tyler" <t...@tt1.org> wrote in message news:GzLAK...@bath.ac.uk...
Von Neumann devoted the last few years of his life to automata. And he made
significant advancements with some help from Ulam. But he is more known
for his contributions to quantum mechanics, game theory and computer design.
He was fairly prolific.
You named less than a dozen individuals. That is not a thriving research
area.
Cellular automata is not mentioned in the index of Godel, Escher and Bach.
It is somewhat like fractals. Fractals had a old origin but Mandlebrot was
sleeping in his car when he made that breakthrough and he was racing against
time to publish ahead of his fellow researchers.
I did not say Wolfram's categorizing scheme was perfect. I said it was
widely
adopted; I read about it several times. I think the real proof of what I am
saying is in the citations. I will try to find something factual. Not sure
how.
On page 878 the graph looks pretty flat between 1974 to 1986.
NKS page 878: "Yet by the late 1970s, despite all these different directions
research on systems equivalent to cellular automata had largely petered out.
That this should have happened just around the time computers were first
becoming widely available for exploratory work is ironic."
I've read a similar summary of CA before and not by Wolfram.
Were you reading papers on CA in 1984?
Tim Tyler
: Von Neumann devoted the last few years of his life to automata. And he made
: significant advancements with some help from Ulam. But he is more known
: for his contributions to quantum mechanics, game theory and computer design.
: He was fairly prolific.
Yes - a master.
: You named less than a dozen individuals. That is not a thriving research
: area.
That may have something to do with the fact that the message was a usenet post.
http://cafaq.com/bibliography/ seems to have about a hundred papers from
the 1970s alone.
: Cellular automata is not mentioned in the index of Godel, Escher and Bach.
A book with little or nothing to do with cellalar automata, if I'm not
mistaken.
: I did not say Wolfram's categorizing scheme was perfect. I said it was
: widely adopted; [...]
I think you may be correct there - but it may be that a major
influence of it has been to muddy thinking on the subject.
: On page 878 the graph looks pretty flat between 1974 to 1986.
: NKS page 878: "Yet by the late 1970s, despite all these different directions
: research on systems equivalent to cellular automata had largely petered out.
: That this should have happened just around the time computers were first
: becoming widely available for exploratory work is ironic."
Ah - you cite Wolfram in your support. I think the problem is that
there's an assertion that Wolfram's history of the field diverges
from that of other observers in the importance of the role attributed
to Wolfram. Citing wolfram in support of this thesis leaves something
to be desired.
: I've read a similar summary of CA before and not by Wolfram.
A citation might help there. It's not always easy to remember who wrote
what in the face of revisionist historians.
: Were you reading papers on CA in 1984?
I wrote some cellular automata simulations in about 1982 - but I don't
recall where I came across the idea. These were in 6502 machine code
on my BBC computer.
As far as I can remember, the first book I read about CAs was
Poundstone's "The Recursive Universe". I read a paperback
version around the time it came out - which was in 1987.
Stephen Harris
> : "George Maydwell" <geo...@collidoscope.com> wrote in message
> :> On Sat, 20 Jul 2002 19:53:49 GMT, Tim Tyler <t...@tt1.org> wrote:
> :> >George Maydwell <geo...@collidoscope.com> wrote:
>
> :> From the history of cellular automata section of the notes for chapter
> :> two of ANKOS: "The publication of my first paper on cellular automata
> :> in 1983 (see page 881) led to a rapid increase of interest in the
> :> field, and over the years since then a steadily increasing number of
> :> papers (as indicated by the number of source documents in the Science
> :> Citation Index shown below) have been published on cellular automata -
> :> almost all following the directions I defined".
>
> : This is correct and if anything is understated. Before he wrote his
> : classifying paper the field of cellular automata was dormant.
>
> That does not appear to be very accurate.
>
> We have Fredkin since the year dot, The Theory of self-reproducing
> Machines in 1966, Codd in 1968, Conway in 1970, Zhabotinsky in 1970,
> Burks from 1970, Kimura from 1976-1979, Toffoli, papers spanning
> 1977-1984, Conway, Guy and Berlekamp's famous proof from 1982,
> Self-replicating loops from Christopher Langton in 1984, Margolus
> in 1984. Interest in cellular automata was on the up.
>
Well you mention Toffoli, and I am going to quote from him a criticism
that appears to dispute your optimism and the rosy view of von Neumann.
http://pm1.bu.edu/~tt/publ.html Page 2 of occam.ps
Tommaso Toffoli, "Occam, Turing, von Neumann, Jaynes:
How much can you get for how little?
(A conceptual introduction to cellular automata)" [ps],
The Interjournal (October 1994); adapted from ACRI'94:
Automi Cellulari per la Ricerca e l'Industria, Rende (CS), Italy,
(Sept.~29-30, 1994).
"Indeed, one may say that von Neumann's is a model
of "life" only in the same sense that chess is
a model of "war": a board game that qualitatively
captures certain essential feedback loops of a
certain situation.
I'm convinced that the very shortcuts that made
von Neumann's project feasible were in part responsible
for shunting cellular automata onto a dead track for a
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
couple of decades. Von Neumann's model of life was very
qualitative---almost metaphorical; many must have felt
that cellular automata could be good only for models of
that kind; perhaps their proper place was, together
with Fibonacci's rabbits (another "highlevel" model
of reproduction), in a "recreational biology" column."
SH: It seems to me that a researcher that you have qualified
is in an excellent position to evaluate the slump of CA and to
be aware of some limiting influences introduced by von Newumann.
Recall, that his focus was actually on self-reproduction and CA
served as a means to that end. Toffoli said "shunting CA onto a
dead track" which is even stronger than Wolfram's "petered out".
> Mandelbrot published in 1982 - and soon it seemed that chaotic
> systems were suddenly fashionable.
>
> There *was* an explosion of interest in CAs in the early 1980s - but the
> factor was rather obviously cheap home computers - and not Wolfram.
>
I dont think so. Home computers were slow with limited memory until
the early '90s. I think your statement is causally off by about 9 years.
> : If you remove Conway, not all the other people added together would
> : equal Wolfram's influence because the influence of the old names had
> : disappeared... CA was a nearly dead field.
>
> Old names like von Neumann, I suppose - who practically invented the
field.
>
> I'm not sure how von-Neuman's influence could ever die away - while
> the field still exists.
>
I dont deny he did important foundational work. But isn't it apparent
to you who really cares about this, who made this their life work rather
than as a vehicle to explore self-reproduction (IMO this really has to do
with the possibility of a sentient AI ala Hofstadter) at the end of his
career
which according to Toffoli had a detrimental influence on CA history.
Regards,
Stephen
Stephen Harris
"Tim Tyler" <t...@tt1.org> wrote in message news:GzLKo...@bath.ac.uk...
I liked that book which I read maybe because you mentioned it.
Well, I provided the Toffoli citation in my other post.
http://pm1.bu.edu/~tt/publ.html Page 2 of occam.ps
"I'm convinced that the very shortcuts that made
von Neumann's project feasible were in part responsible
for shunting cellular automata onto a dead track for a
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
couple of decades."
Wolfram said late 1970's so two decades would be
approximately 1960 to 1980. Von Neumann wrote
"The Theory of Self-Reproducing Automata" in 1966.
Regards,
Stephen
Erik Max Francis
> I think that if you asked a thousand people who had earned a 4 year
> college degree in various fields such as English, Psychology,
> Sociology
> Anthropology, etc., peripheral science majors outside of math and
> physics
> a multiple choice question like: Who is known for his substantial
> achievements
> in the field of cellular automata?
>
> a. Gacs
> b. Fredkin
> c. Wolfram
> d. Von Neumann
>
> And I think at least 75% would choose Wolfram.
Well first of all, I very seriously doubt 75% people with 4-year degrees
non-maths, non-physics majors would be able to recall _any_ names
related to cellular automata. I'd wager that of those who can actually
remember a name, Conway would be a lot more forthcoming than Wolfram.
Gordon D. Pusch
> On Sat, 20 Jul 2002 19:53:49 GMT, Tim Tyler <t...@tt1.org> wrote:
>
> >George Maydwell <geo...@collidoscope.com> wrote:
> >
> >: From page 928 of ANKOS: "But until after my work in the 1980's
> >: simulators for more general 2D cellular automata were rare."
> >
> >[...]
> >
> >: Wolfram seems to be implying that his work has been a significant
> >: influence on the development of general 2D CA simulators. Does anyone
> >: know of anything he might have done to give credence to this claim?
> >
> >You could read it like that - but it seems like more a statement of
> >chronology than anything else.
>
> The quote seems ambiguous at best. A clear statement of chronology
> would read simply "But when I did my work in the 1980's simulators for
> more general 2D cellular automata were rare".
It has already been established that Wolfram has little interest in either
being historically accurate, or in following even the most basic standards
of scholarly attribution.
-- Gordon D. Pusch
perl -e '$_ = "gdpusch\@NO.xnet.SPAM.com\n"; s/NO\.//; s/SPAM\.//; print;'
Stephen Harris
"Erik Max Francis" <m...@alcyone.com> wrote in message
news:3D3AFB90...@alcyone.com...
> Stephen Harris wrote:
>
> > I think that if you asked a thousand people who had earned a 4 year
> > college degree in various fields such as English, Psychology,
> > Sociology
> > Anthropology, etc., peripheral science majors outside of math and
> > physics
> > a multiple choice question like: Who is known for his substantial
> > achievements
> > in the field of cellular automata?
> >
> > a. Gacs
> > b. Fredkin
> > c. Wolfram
> > d. Von Neumann
> >
> > And I think at least 75% would choose Wolfram.
>
> Well first of all, I very seriously doubt 75% people with 4-year degrees
> non-maths, non-physics majors would be able to recall _any_ names
> related to cellular automata. I'd wager that of those who can actually
> remember a name, Conway would be a lot more forthcoming than Wolfram.
>
I mentioned Conway as a major competitor but he is known for "Life".
But this interest is rather specific---I cannot recall Conway being
mentioned
as a founding father of cellular automata; Wolfram is frequently by the
press.
There is a thread on this newsgroup about -why Wolfram gets so much
media attention-Wolfram has been in the news continually for years and
"Life" and Conway have pretty much out of the news. I think twenty
years ago before Mathematica you would have been right. Ask some
of your literate not computer scientist friends: Have you heard of
Stephen Wolfram; have you heard of the Game of Life? (Conway)
Without clues I think the answers will at least 75% favor Wolfram.
Ulam is deemed the originator of the idea for von Neumann using CA
for demonstrating self-reproduction. Von Neumann was working on
his book about self-reproducing automata in 1957 but died. It was
completed nine years later (1966) and von Neumann does not get
much credit for that. Also there is the opinion of Toffoli, supposedly
an expert, that von Neumann may well have been detrimental to the
development and growth of CA as a field. (quoted elsewhere)
Wolfram has devoted most of his life including building Mathematica
to pursue his interest in cellular automata. Von Neumann just towards
the end of his life and VN is better known for other explorations.
This paper by Mitchell is a historical survey and von Neumann is
mentioned a bunch but so is Wolfram. Other CA papers overwhelming
cite Wolfram over von Neumann. Usually his classification scheme.
http://www.santafe.edu/~mm/paper-abstracts.html#ca-review
From page 4 of Melanie Mitchell's review of CA.
"Von Neumann's Self-Reproducing Cellular Automaton
The original concept of cellular automata is most
strongly associated with the great scientist and
mathematician John von Neumann. According to the
history recounted by Burks (1966, 1970b), von Neumann
was deeply interested in connections between biology
and the (then) new science of computational devices,
\automata theory." Paramount in his mind was the
biological phenomenon of self-reproduction, and he
had posed a fundamental question: What kind of logical
organization is suĆcient for an automaton to be able
to reproduce itself?" The idea of using cellular automata
as a framework for answering this question was suggested
to von Neumann by Stanislaw Ulam (Burks, 1970b). Thus,
the original concept of cellular automata can be credited
to Ulam, while early development of the concept can be
credited to von Neumann."
The origin may be credited with von Neumann. But according
to Toffoli: "I'm convinced that the very shortcuts that made
von Neumann's project feasible were in part responsible
for shunting cellular automata onto a dead track for a
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
couple of decades."
I think Wolfram is correct in claiming he revitalized the
growth of CA papers as depicted by the chart on page
878 of NKS. Books on fractals hardly mentioned CA
and home computers were in infancy as was the internet.
Yes they were contributing factors. But I have read before
Wolfram's book was published about the importance of
having a classification scheme provided by Wolfram and
the impetus it gave to the field. Wolfram has kept CA in
the limelight for years. I mean non-techies who decide
about funding. Fredkin has also helped with this but he
focused more on words like discrete and digital computer.
Fredkin just doesn't have the press following that Wolfram
does. And the media's obsession with a NKS.
The book was a best seller before it was published. Like
Goedel, Escher and Bach it wasn't made into a best seller
by the academic community, but by literate educated
thinking people. And they knew the book was about CA
even if they didn't have much idea what CA really was.
Yes it has to do with marketing. Mandlebrot slept in his
car and current CA researchers sleep in beds. I believe
that his book is a best seller because the educated public
link Wolfram to CA, Mathematica, expertise and success.
You can believe otherwise, I'm moving on.
I find Wolfram arrogant, but based on actual accomplishments.
He is scrupulously intellectually honest. I find this challenged
by people who claim he has no integrity. That classifies them.
I do not find Wolfram rewriting history when he says the
CA field had "petered out". Instead I find him in conflict with
proponents of a mythologized history of CA.
Finally, when I mentioned the Illiad as an example of the
difference between translation and transformation you
responded that the copyright had expired. Which is why
I picked the Illiad. I was talking about seperate instances
of a class versus a different class and what made proofs
similar proving some idea and what made other proofs
different which still proved the same idea. So I removed
the issue of copyright because it was irrelevant which you
reinstated like I didn't know the Illiad was 2,000 years
old when there were no copyrights and that current law
was life of author plus 70 years I think now. I found it
frustrating that you invented that reply because you did
not understand what I wrote--because you are another
arrogrant person who thinks they can scan messages
that other people have given a lot of thougt to and churn
out a plethora of insightful replies. You didn't understand
what I wrote so you made up something. Which is what
pissed me off with George Maydwell. He didn't know
what Wolfram meant so he fabricated some meaning
which was pretty ludicrous.
Wolfram has pride in his intellectual honesty. It seems
that Tim Tyler shares that quality. Most of the rest of
the detractors are full of ignorance, spite and envy.
Whether this is true or not does not really matter since
this presupposes valuing intellectual honesty.
I dont believe Conway's book (books) have ever been
a best seller before publication. Most people will really
get their money's worth, their is a lot of education. The
people who are better educated but thought they might
get the secret to the universe published in a book are
getting exactly what they deserve. Consider it a cheap
lesson ($50 with tax) about gullibility. The superior most
intelligent people have an ego to match and that produces
a naivete impervious to self-examination: suckers for cons
As a smart person it took me a long time to learn one
crucial lesson. A smart person nearly always assumes they
have sufficient information to make an intelligent decision
when in fact they nearly always need to investigate more.
Your Grandiose servant,
Stephen
Erik Max Francis
> I mentioned Conway as a major competitor but he is known for "Life".
> But this interest is rather specific---I cannot recall Conway being
> mentioned
> as a founding father of cellular automata; Wolfram is frequently by
> the
> press.
Your hypothetical question didn't deal with the facts, it dealt with
peoples' impressions -- and the impressions of non-maths, non-physics
degree holders, people who are not likely to know much, if anything,
about cellular automata.
It's possible that the recent press that Wolfram has garnered himself
would put Wolfram higher up in such an opinion poll. But I bet more
people polled would remember Conway's name more readily than Wolfram's,
regardless of Conway's contributions in comparison to Wolfram's.
I bet half the people, at least, would have no idea what a cellular
automaton was. It's not exactly something you need to come across when
earning a four-year degree, even one in maths or physics. I would wager
the more clever test takers might come up with the name von Neumann,
since his name conjures up images of automata in general, again
regardless of his actual contributions.
An opinion poll is not a very good way to test who is the most important
person in a field.
Kent Paul Dolan
> I dont think so. Home computers were slow with limited memory until
> the early '90s. I think your statement is causally off by about 9 years.
Nope, got to side with Tim on this one. I for one was a follower of a
then-horde of thousands of Apple][+ owners writing implementations of
Conway's Game of Life in 1984, and the Life newsletter was most active
in that time frame.
While home computers _then_ look dog slow _now_, _then_ they looked
like speed demons compared to the usual one day turnaround on an
office business computer, and everyone who could was programming what
George Maydall just characterized as "eye candy". I know _I_ was, on
an IBM PC luggable computer I had at work, even in 1983, as well as my
first home computer in 1984.
xanthian.
Stephen Harris
"Kent Paul Dolan" <xant...@well.com> wrote in message
news:a3eaa964.02072...@posting.google.com...
Hello Kent,
I found your post interesting and I wanted to ask about Conway and Life
anyway.
Since you snipped the basis for my comment, I will reinsert it:
TT wrote:
There *was* an explosion of interest in CAs in the early 1980s - but the
factor was rather obviously cheap home computers - and not Wolfram.
SH wrote:
: If you remove Conway, not all the other people added together would
: equal Wolfram's influence because the influence of the old names had
: disappeared... CA was a nearly dead field.
In 1981 or 1982 I bought a new Vic 20. It had no floppy or hard drives.
It has an audio tape deck drive! Maybe 64k or 128 of memory. Now
they did do programming---but running a CA simulation--wouldn't that
take hours and hours?
I can see a few diehard enthusiasts dedicating themselves to this type
of exploration...but were talking 'an explosion of interest in CA due
to cheap home computers'.
Now the IBM XT came out in 1983 with 10/20 mb hard drive and
and less than 1mb of memory at 4.77 hurtz. This cost over $5000!
The Apple Mac cost $2495! in 1984. Now how many of these
people who could afford these prices for home computers also
wrote papers that caused the Index of CA papers to start its
upward trend? I mean people who explored on their 'cheap home
computers' rather than using their computer access at the university.
You would think most papers in Scientific Index for CA would
could from academia with university perhaps corporate computers.
Home computers: These early computers were pretty arcane and
CAs a fairly esoteric field. Since these are home computers how
many people would then submit academic papers to be published
in place of their university computers.
Do you see what I'm saying? You would need a huge influx of
home computer users to produce academica papers that would
not have otherwise been produced. As difficult as this was for
the early cheaper computers to do this type of programming
(in a little known field) and then later the more capable computers
were not cheap, they were expensive and thereby limited. Did
most of the people who would become Life nerds have $2500
or more to plonk out for a more capable computer? And then
because of their home experiments write papers for the CA journals?
It was 1993 when I paid over $1800 for a 4mb mem, 33mz at Turbo
speed with a 100mb hard drive. This was after a price drop. Their
were local bulletin boards where you could tap into for Fidonet.
The Internet was still hardly available for the home computer user
but just about ready to prosper. I think you could do emails.
So let me ask you some questions about your post. Was the Life
newsletter snail mail? Did they have an easy way to intercommunicate?
You wrote that their were thousands of Apple I computer users.
Are you also saying there were thousands of people using their
Apple I for programming the Game of Life? Or do you mean perhaps
a few hundred did this?
If a few hundred did this, how many wrote papers because they had
home computers. If they would have written papers at their day jobs
in academia anyway, this would not produce an upswing of CA papers
in the Scientific Index under discussion on page 878 NKS. And just
after these fairly cheap (I got my Vic 20 for $99) home computers
the price went way up for home computers. Much more affordable
to business types than Game of Life hackers. And some of these
hackers already had access to University computers. The more capable
computers (not so hard to program or take as long to run programs)
cost more.
So:
There *was* an explosion of interest in CAs in the early 1980s - but the
factor was rather obviously cheap home computers - and not Wolfram.
I find this logic fallible. The very early cheap computers were hard to
program. Later the computers that were easier to program were expensive.
And most of the home users who were not also academics with access to
univeristy computers, were unlikely to submit papers which would register
on the academic database of Scientific Index for CA papers. It seems more
likely to me that an academic paper (Wolfram around 83 or 84) which was
read by other academics (availability limited) who actually do the causal
production of papers registered on the Scientific Index is the most likely
factor. The "cheap home computer" bunch which is a dubiously distinguished
is a remote causal influence for the output were looking for. I think it is
a
logical fallacy like false attribution of underlying cause. Something that
happened
at the same time but did not causally produce "explosion of interest in
CAs".
You or Tim are really gonna have to go some to convince me that cheap home
computers is more than a very lightly weighted causal factor in the
explosion of
CA and that a seminal paper (classification scheme) is not a much larger
factor
since it directly impacts the people who do write papers or who are in a
close
academic area of interest. Something which sparked the interest for other
papers.
I think even now that maybe one in 1,000 of cheap home computer users who
has access to lots of CA programs, usenet, forums etc. submits a paper for
publication that will be registered by the Scientific Index for CA papers. I
dont
think an academic who would write a paper but who also owns a cheap home
computer should be but in the cheap home computer paper publisher because
they would publish anyway. I think the situation was worse 20 years ago.
: If you remove Conway, not all the other people added together would
: equal Wolfram's influence because the influence of the old names had
: disappeared... CA was a nearly dead field.
Erik Max Francis suggested Conway as a larger influence on keeping
the interest in CA maintained and flourishing. That thought had occurred
to me also which is why I worded the above quote that way.
But I dont find Conway given much credit in the literature as a founding
father of CA or even keeping CA interest alive. Is that because though
Life is a type of CA, Conway's interest was focused on Life only with
just its rules, rather than cellular automata as a general field of
interest?
I mean usually Conway is associated with the Game of Life and CA
might be mentioned as an afterthought. Was he so singular with Life?
(in this area I know he has myriad geometric interests--tiling.) Yes
he is given plenty of credit for Life, but life is not promoted as a CA,
and not even always mentioned as a CA. I dont think that a reader
with a good general education would know that Life was a CA even
if they knew John Conway invented Life. Do you agree with that?
I have read histories about the early computer days when the hardware
assembly nerds would make today's hackers look like beach pansies.
I almost orderd an Altair but was saved by Bailey's, Bushmills and coffee.
The Well of Souls,
Stephen
Tim Tyler
: "Kent Paul Dolan" <xant...@well.com> wrote:
:> "Stephen Harris" <stephen....@worldnet.att.net> wrote:
:> > TT wrote:
:> > > There *was* an explosion of interest in CAs in the early 1980s -
:> > > but the factor was rather obviously cheap home computers - and not
:> > > Wolfram.
:> > I dont think so. Home computers were slow with limited memory until
:> > the early '90s. I think your statement is causally off by about 9 years.
:>
:> Nope, got to side with Tim on this one. I for one was a follower of a
:> then-horde of thousands of Apple][+ owners writing implementations of
:> Conway's Game of Life in 1984, and the Life newsletter was most active
:> in that time frame.
:>
:> While home computers _then_ look dog slow _now_, _then_ they looked
:> like speed demons compared to the usual one day turnaround on an
:> office business computer, and everyone who could was programming what
:> George Maydall just characterized as "eye candy". I know _I_ was, on
:> an IBM PC luggable computer I had at work, even in 1983, as well as my
:> first home computer in 1984.
: In 1981 or 1982 I bought a new Vic 20. It had no floppy or hard drives.
: It has an audio tape deck drive! Maybe 64k or 128 of memory. Now
: they did do programming---but running a CA simulation--wouldn't that
: take hours and hours?
The Vic 20 had 5K of RAM by default - though you could expand it to
32kb.
In 1982 I had a 32K BBC B computer that cost me about 350 UKP.
No doubt like many others, one of the first things I did on it,
was write a GOL simulation.
My first simulations ran in teletext mode - using "O" for on and " "
for off.
I'd guess I got about ten franes a second on a 80x25 display.
Later on, I believe I took advantage of the Teletext graphics -
to produce a display of 160x75 - no doubt with a drop in speed.
The BBC had "hi-res" graphics of 640x512 - but life in those modes would
indeed have been too slow.
You could get a lot of the feel for life out of a small display at 10fps -
it was certainly a lot more interesting than working with stones on a go
board at any rate.
Home computers probably also fuelled interest in fractals - which
also became popular at about the same time.
: You would think most papers in Scientific Index for CA would
: could from academia with university perhaps corporate computers.
Having a computer on your desk still helps. Running CA simulations
via a dumb terminal over a network is not very satisfying.
I expect interest in computer science generally increased dramatically
in the early 1980s. Computer magazines were on the news stands for the
first time - and it was clear that money could be made in the field by
individual pioneers.
: I think it is a logical fallacy like false attribution of underlying
: cause.
So - you think I'm doing it - and Wolfram isn't?
Between 1980 and 1985 the number of people who could or had run a CA
simulation probably increased by a factor in the thousands.
I imagine they would typically have been running Conway's Game of Life.
Few of them would have even heard of Wolfram initially.
Kent Paul Dolan
Steve Harris wrote:
[Steve's post is too disjointed and silly to be worth answering
directly; told that the early-to-mid 1980's indeed contained a raft of
hobbiests doing CA on newly available personal computers with color
graphics displays, he still wants to argue a participant out of
believing his own life history.]
? : You would think most papers in Scientific Index for CA would
? : could from academia with university perhaps corporate computers.
Home computer hobbiests are not typically the source of professional
papers, nor indeed are most computer users. I've been a programmer
since 1961, and never yet had a professional paper published in a
publically available journal (though several for in-house
publication), nor felt the urge to write one. [Though I have
co-authored three international computer standards.] Papers are the
product of academia and some few folks lucky enough to be styled
corporate scientists, but rarely emerge from "the trenches" in my
experience.
? Having a computer on your desk still helps. Running CA simulations
? via a dumb terminal over a network is not very satisfying.
Indeed, it was very much the hands on approach that made GOL
programming popular, and, frankly, it was one of the easiest, least
talent requiring things you could do with a computer of the capability
available then, granted you sometimes had to leave something running
over-night to see how it came out.
Writing the GOL in a 4 Kbyte usable address space is a whole different
animal from writing a useful text editor in that much space. The
folks who could do the latter left me awestruck.
? I expect interest in computer science generally increased
dramatically
? in the early 1980s. Computer magazines were on the news stands for
the
? first time - and it was clear that money could be made in the field
by
? individual pioneers.
Yep, I paid for my $3500 box with some moonlighting efforts using that
same box for a county mental health clinic.
? Between 1980 and 1985 the number of people who could or had run a CA
? simulation probably increased by a factor in the thousands.
And the number who'd written one, ditto.
? Few of them would have even heard of Wolfram initially.
Right. Outside the community of his software customers, most people
who know of him now hadn't heard of Wolfram until a couple of months
ago; I know I knew nothing about him in 1985, (when I bought the Amiga
for which someone wrote a GOL that ran full screen pixel-per-cell
updating at the frame refresh rate), and probably not until about 1995
or so, despite that I kept up fairly well with the trade literature.
xanthian.
[Oh, and the Game of Life newsletter was indeed hardcopy, Steve.]
Stephen Harris
"Kent Paul Dolan" <xant...@well.com> wrote in message
news:a3eaa964.02072...@posting.google.com...
> Tim Tyler ?t...@tt1.org? wrote:
>
> Steve Harris wrote:
>
> [Steve's post is too disjointed and silly to be worth answering
> directly; told that the early-to-mid 1980's indeed contained a raft of
> hobbiests doing CA on newly available personal computers with color
> graphics displays, he still wants to argue a participant out of
> believing his own life history.]
>
Wolfram said that Scientific Index of CA papers increased dramatically
about the time of his early paper around 1983.
Did Tim or you write a paper back then? If you didn't then it did
not get counted in the Scientific Index of Ca papers. This has nothing
to do with arguing out of your experience of playing with Life.
If you or other hobbyists did not write papers then the increase
of CA papers did not come from hobyyists. It came from people
who do write scientific papers.
I mean this is just repeating what you say below. (cheaper) home
computers helped hobbyists write programs not write papers.
The papers came from academics or scientists who did not
need cheap home computers to run their programs, though this
may have marginally helped them.
I disputed this remark by Tim Tyler:
:There *was* an explosion of interest in CAs in the early 1980s -
: but the factor was rather obviously cheap home computers -
:and not Wolfram.
Wolfram wrote an important paper about classification which is
cited all the time and which was adopted by many of the other
CA researchers who do read papers and write them.
It is these other researches who get tallied on the Scientific
Index. Wolfram said that because the Scientific Index went up
after his paper was read (by the people who do write papers)
that this indicated his paper (and him by default were _influential_).
I didn't deny your experience but that your experience produced
the sharp increase of CA papers among the scientific community.
It seems to me your statement agrees with that:
KPD wrote:
"...Papers are the product of academia and some few folks lucky
enough to be styled corporate scientists, but rarely emerge from
"the trenches" in my experience."
That means you are saying the early hobbyists of Life in the
"trenches" rarely produced CA papers(maybe Tim did) so
their efforts did not account for the rise of CA papers which
the Scientific Index measures. At most they were very weak
cause(hobbyists on cheap home computers) and as a cause
more likely to surface several years later.
This gives Wolfram's claim more credibility because the advent
of cheap home computers had very little impact on the measuring
device: The Scientific Index of CA papers which surged after
Wolfram published. The cheaper home computers and Wolfram's
paper happened about the same time. Which was the largest
contributing factor? Hobbyists programming on cheap computers
who rarely write papers or Wolfram who even now is still cited
frequently for that paper by the people who write papers---
academics, which at that time Wolfram was an academic.
I wrote the sentence below and forgot the word "come" after
would/could which was so hard to figure out that my post
was silly and disjointed and an attack on or an attempt to deny
hobbyist programming experiences rather than explaing why
hobbyists did not generate the increase of Scientifc Index CA
papers which Wolfram was alluding to that was part of the thread.
Does that make you silly and illogical because you wrote a
post which supported my position? Back then kids who would
be going to college could not afford those Macs. And I think
the people who could afford Macs were business types with a
strong interest in Photoshop, by in large. You hobbyists were
closer to the olden meaning of hacker which were in short supply
and who wrote even fewer papers.
That is not an attack of your experience. I pointed out that there
weren't that many of you and so did not account for the rise in
Scientific Index CA papers. The rest of your post just affirms
that you were out of touch with the element that did read
scientific CA papers. You knew Conway's name because of Life.
That is not right. Wolfram's book was on Amazon's bestseller list before
the book was published. I dont think that has ever happened before for
any scientific book. You are trying to attribute that to recent media hype??
I dont know what you mean by trade literature. It certainly isn't cellular
automata papers becasue Wolfram's name is cited all the time, often
with a description of his classification scheme or mention of it.
I think I heard of Wolfram perhaps through Clifford Pickover who
writes great fractal and rec math books. The point about fractals
/chaos pushing cellular automata would be more immediate
for academics to publish papers, I agree. People who write home
fractal programs publish less also. I am sure this inspired quite a
few teenagers who went on to college to write papers also. The
upswing is largest later than 1984 which would tend to localize
Wolfram's impetus from 1983-1985.
But that is guesswork, fractals or Wolfram's paper and their
relative influence on the growth of the field translating to more papers.
Kent, you did not write anything to dissuade me from believing that
hobbyists did not generate papers. And indirectly, it takes more time.
My argument was about hobbyists/cheap home compters not making
many CA papers happen. I did not address fractals which do have merit.
> xanthian.
>
> [Oh, and the Game of Life newsletter was indeed hardcopy, Steve.]
There were thousands of people on this mailing list which was
independent of a general Apple development newsletter. I mean
was it specific to Life or covered other programming topics?
Well, I managed to absorb "silly" and "disjointed" without insulting you,
Stephen
Stephen Harris
"Tim Tyler" <t...@tt1.org> wrote in message news:GzpB0...@bath.ac.uk...
My point is that hackers and hobbyists provided few of the papers that
are measured by the Scientific Index for CA papers; they mainly come
from other academics. So the cheap home computers are going to take
a longer time to generate a showing of academic papers.
I do think chaos and fractals stirred academic interest as well as hobbyists
so could generate some more immediate papers from academics which
could spread to CA papers produced and measured. I also read about
modelling fluid dynamics from time to time(just as a topic). So fractals
which made a big splash (Barnsley, Pickover) seem like a lot stronger
argument to me than cheap home computers. I think those take longer
longer to be felt and percolate a swarm of CA papers.
That growth of the internet also corresponds to the increases of that
graph on page 878 and the dramatic increase of CA papers. The very
early computers were as cheap as modern computers but a lot harder
for most people to play with. The graph is flat for a long time in the
70's and starts moving up around 1984. I remember the Superbowl
Mac which was $2500.
> I imagine they would typically have been running Conway's Game of Life.
> Few of them would have even heard of Wolfram initially.
> --
I surely agree with you that there was academics who do write papers,
and hobbyists who seldom write papers, interested in Life. But this was
strong in the 70's as well which is another reason they would not have
heard of Wolfram. I guess that is what you mean by initially.
When I think of Life I think of a game. It does not conjure the thought
Oh, that is a type of CA. Now, if that is the same as for other people,
then it not going to promote much interest in writing papers about other
types of CAs. I am not sure that I read that Life was a CA in another
source than that Recurrent Universe (title?)(Poundstone) book.
Is that just me or has Conway been stereotyped to Life rather than CA?
Regards,
Stephen
Owen Rees
<GzpB0...@bath.ac.uk>:
>The BBC had "hi-res" graphics of 640x512 - but life in those modes would
>indeed have been too slow.
In about 1988, I implemented life on an Acorn Electron (basically a half
speed BBC micro) and got it up to a reasonable speed for a 255x255
universe. The speed was very dependent on the number of live cells, but
it ran the pentadecathlon at 92 gen/sec in the last version I measured.
Having publically (letter to the Guardian Jan 21 1988 - I still have the
cutting) claimed that it was the programming rather than the machines
that was the problem I felt I ought to prove, at least to myself, that I
was right. The goal I had set myself was to run the r-pentomino to
generation 1103, where it becomes a collection of simple patterns, in a
reasonable time, and was very happy when the Electron could do it in
less than two minutes (although some gliders had died at the edge of my
limited universe by then).
I later implemented an improved version of the algorithms on an Atari
STe, this time the goal was to run the breeder, which it did at 2
gen/sec at the stage when the first glider gun has been assembled.
While having all that fun, I was not interested in the general theory of
cellular automata (although I knew the term, and that Conway's "life"
was one). I think I may have been vaguely aware of Wolfram's name, but
having read Poundstone's book, would have been more likely to name Von
Neumann if asked who, apart from Conway, had done anything in the field.
--
Owen Rees - opinions expressed here are mine; for the full disclaimer
visit <http://www.users.waitrose.com/~owenrees/index.html#disclaimer>
for e-mail use "owenrees at waitrose.com" instead of the From address
Gordon D. Pusch
> "Kent Paul Dolan" <xant...@well.com> wrote in message
> news:a3eaa964.02072...@posting.google.com...
>> Tim Tyler ?t...@tt1.org? wrote:
>>
>> Steve Harris wrote:
>>
>> [Steve's post is too disjointed and silly to be worth answering
>> directly; told that the early-to-mid 1980's indeed contained a raft of
>> hobbiests doing CA on newly available personal computers with color
>> graphics displays, he still wants to argue a participant out of
>> believing his own life history.]
>
> Wolfram said that Scientific Index of CA papers increased dramatically
> about the time of his early paper around 1983.
Correlation does not imply causality.
Nor was Wolfram the first (let alone only!) person working on CAs
at that time.
Stephen Harris
"Gordon D. Pusch" <gdp...@NO.xnet.SPAM.com> wrote in message
news:gieldug...@pusch.xnet.com...
I agree. That is why I didn't think "cheap" home computers that hobbyists
used
to program Life was causal. It would be rare for a hacker to produce a paper
that would impact the Scientific Index for CAs. However Wolfram was an
academic
and did produce a paper I think in 1983 that is still cited. His paper has a
direct causal impact because it was read by other CA researcher who write
papers. Wolfram asserts his paper(s) was quite influential in reviving
interest
in cellular automata. There were other researchers as you say. Are you aware
of somebody who wrote an important paper at that time which would stimulate
interest in CA? Usually the number of citations is taken to reflect
influence.
Fractals appeared prominently around that time which though somewhat
indirect seems to be a reasonable co-cause of the surge of CA papers.
Tim Tyler suggested fractals and chaos as contributive factors. There had
been other CA papers/researchers before Wolfram but that was not a
factor in reviving interest as the graph was flat for like 12 years before.
This period of time covered when Life was actively investigated.
I think one needs to point to another seminal CA paper written by
a different researcher than Wolfram in order to substantiate other
researchers as motivating the resurgence of CA papers. That premise
is more circumstantial than Wolfram's claim, otherwise your suggestion
remains a maybe, just a possibility. Wolfram's name is in every review
of CA for that paper and classification scheme. How much did that
paper legitimicize CA as a science and spark interest for others I
suppose is a matter for speculation which I've devoted enough time to.
Evidently,
Stephen
Tim Tyler
: It would be rare for a hacker to produce a paper that would impact the
: Scientific Index for CAs. However Wolfram was an academic
: and did produce a paper I think in 1983 that is still cited. His paper has a
: direct causal impact because it was read by other CA researcher who write
: papers. Wolfram asserts his paper(s) was quite influential in reviving
: interest in cellular automata. There were other researchers as you
: say. Are you aware of somebody who wrote an important paper at that
: time which would stimulate interest in CA? Usually the number of
: citations is taken to reflect influence.
Berlekamp, Conway and Guy's 1982 epic proof that Life was
capable of supporting universal computation and self-reproduction
is what springs to mind.
Here are some oft-cited works in the field - prior to 1987 -
in order of publication.
von Neumann: Theory of Self Reproducing Automata:
81 citations (1966) http://citeseer.nj.nec.com/context/205825/0
Berlekamp, Conway and Guy - Winning Ways, vol 2
81 citations (1982) http://citeseer.nj.nec.com/context/49420/
Wolfram: Universality and complexity in cellular automata:
61 citations (1984) http://citeseer.nj.nec.com/context/59484/0
Wolfram (Editor): Theory and applications of cellular automata:
89 citations (1986) http://citeseer.nj.nec.com/context/59486/0
Tommaso Toffoli and Norman Margolus: Cellular Automata Machines
63 Citations (1987) http://citeseer.nj.nec.com/context/59481/0
I think I'll have to leave the matter there. I'm very happy
that Wolfram performed the work that he did - but the drawing of
graphs in an attempt to prove just how significant that work was
leaves me cold.
Kent Paul Dolan
and wrote:
> "Kent Paul Dolan" <xant...@well.com> wrote:
> Wolfram said that Scientific Index of CA papers increased dramatically
> about the time of his early paper around 1983.
And also, apparently, took credit to his paper for that increase, an
incredible display of hubris.
> If you or other hobbyists did not write papers then the increase
> of CA papers did not come from hobyyists. It came from people
> who do write scientific papers.
But we are looking for causes, not origins.
> I mean this is just repeating what you say below. (cheaper) home
> computers helped hobbyists write programs not write papers.
> The papers came from academics or scientists who did not
> need cheap home computers to run their programs, though this
> may have marginally helped them.
> I disputed this remark by Tim Tyler:
> :There *was* an explosion of interest in CAs in the early 1980s -
> : but the factor was rather obviously cheap home computers -
> :and not Wolfram.
Yep, and you _still_ miss Tim's point. Wolfram may have published a
paper, but Worfram's paper probably had next to nothing to do with the
effusion of papers about CAs. That was more likely due to all the
hobbiests doing CAs at home, some of whom then went to grad school and
wrote papers in an area they already knew and loved, not from reading
Wolfram's paper, but from enjoyable personal experience. Just because
_one_ hobbiest didn't become a paper writer doesn't imply that _no_
hobbiests became paper writers. You are indulging in a fallacy of
logic identified before the beginning of the common era: "after this,
therefore because of this" doesn't work. Given a choice between
"after one Wolfram paper, a deluge of CA papers", or "after thousands
of hobbiests get experience programming CAs at home, a deluge of CA
papers", the time ordering doesn't help, you need to use other methods
of reasoning.
> Wolfram wrote an important paper about classification which is
> cited all the time and which was adopted by many of the other
> CA researchers who do read papers and write them.
Bully for him, especially since stuff posted here continually calls
Wolfram's classification scheme into doubt. That still doesn't
promote others to write about CAs, even if, in writing about CAs, they
then go on to quote Wolfram's paper as for a while the only available
recent work: that _requirement_ for citation of recent work is how the
academic publishing game works, and again has nothing to do with
Wolfram's paper being causal for the outburst of CA efforts.
> It is these other researches who get tallied on the Scientific
> Index. Wolfram said that because the Scientific Index went up
> after his paper was read (by the people who do write papers)
> that this indicated his paper (and him by default were _influential_).
Which was incredibly arrogant of him, given the much more likely real
cause, of which he was surely fully aware, wouldn't you say?
[remainder of blather snipped unread; you really need to learn how to
say what you mean in 100 words or less]
xanthian.
Tim Tyler
: On Tue, 23 Jul 2002 12:07:15 GMT, Tim Tyler <t...@tt1.org> wrote:
:>The BBC had "hi-res" graphics of 640x512 - but life in those modes would
:>indeed have been too slow.
: In about 1988, I implemented life on an Acorn Electron (basically a half
: speed BBC micro) and got it up to a reasonable speed for a 255x255
: universe. The speed was very dependent on the number of live cells, but
: it ran the pentadecathlon at 92 gen/sec in the last version I measured.
: Having publically (letter to the Guardian Jan 21 1988 - I still have the
: cutting) claimed that it was the programming rather than the machines
: that was the problem I felt I ought to prove, at least to myself, that I
: was right. The goal I had set myself was to run the r-pentomino to
: generation 1103, where it becomes a collection of simple patterns, in a
: reasonable time, and was very happy when the Electron could do it in
: less than two minutes [...]
It sounds like no mean feat.
I had an Electron - which I used to write Electron Repton on.
The graphics performance was less than impressive - since the
main CPU did double time as the graphics chip.
I reckon a Life pattern at 92 gen/sec in a 2^16 universe
probably made you the Alan Hensel of the Acorn world ;-)
Dean Hickerson
> speed BBC micro) and got it up to a reasonable speed for a 255x255
> universe. The speed was very dependent on the number of live cells, but
> it ran the pentadecathlon at 92 gen/sec in the last version I measured.
> Having publically (letter to the Guardian Jan 21 1988 - I still have the
> cutting) claimed that it was the programming rather than the machines
> that was the problem I felt I ought to prove, at least to myself, that I
> was right. The goal I had set myself was to run the r-pentomino to
> generation 1103, where it becomes a collection of simple patterns, in a
> reasonable time, and was very happy when the Electron could do it in
> less than two minutes (although some gliders had died at the edge of my
> limited universe by then).
Around 1987 I wrote a Life program for an Apple IIe, which could run the
R pentomino in about 40 seconds, and the "acorn" (which finishes in
generation 5206) in about 8 minutes. A few years later I bought a speedup
chip which increased the processor speed from 1 megahertz to 3.5 megahertz,
reducing those times to about 12 seconds and 2.5 minutes.
The Apple had 2 memory areas that mapped to the display, using 7 bits
per byte to represent 7 pixels. The program worked directly with those
pages, so it displayed every generation. It kept track of which
regions had already become period 2, and didn't have to recompute them.
Unfortunately, the program was limited to the Apple's screen size
(192x280 if I remember right), so it couldn't run "rabbits".
For those of you who aren't familiar with these patterns, here they are:
#C R-pentomino
#C Runs for 1103 gens. Initial pop = 5. Final pop = 116.
x = 3, y = 3
b2o$2o$bo!
#C Acorn
#C Found by Charles Corderman, about 1971
#C Runs for 5206 gens. Initial pop = 7. Final pop = 633.
x = 7, y = 3
bo$3bo$2o2b3o!
#C Rabbits
#C Found by Andrew Trevorrow in 1986
#C Runs for 17331 gens. Initial pop = 9. Final pop = 1744.
#C Gens 4760-5165 include a rather rare 16-bit still-life called a "scorpion".
x = 7, y = 3
o3b3o$3o2bo$bo!
Incidentally, here's a new methuselah which I found recently:
#C Blom
#C Found by Dean Hickerson, 7/7/2002
#C Runs for 23314 gens. Initial pop = 13. Final pop = 2740.
x = 12, y = 5
o10bo$b4o6bo$2b2o7bo$10bo$8bobo!
If anyone knows of any small (say inside a 30x30 region) Life patterns that
take longer to finish, I'd like to hear about them.
Dean Hickerson
de...@math.ucdavis.edu
Stephen Harris
"Tim Tyler" <t...@tt1.org> wrote in message news:Gzr3...@bath.ac.uk...> : Are you aware of somebody who wrote an important paper at that
> : time which would stimulate interest in CA? Usually the number of
> : citations is taken to reflect influence.
>
> Berlekamp, Conway and Guy's 1982 epic proof that Life was
> capable of supporting universal computation and self-reproduction
> is what springs to mind.
>
> Here are some oft-cited works in the field - prior to 1987 -
> in order of publication.
>
>
> 81 citations (1982) http://citeseer.nj.nec.com/context/49420/
>
> Wolfram: Universality and complexity in cellular automata:
> 61 citations (1984) http://citeseer.nj.nec.com/context/59484/0
>
> that Wolfram performed the work that he did - but the drawing of
> graphs in an attempt to prove just how significant that work was
> leaves me cold.
> --
> __________
> |im |yler http://timtyler.org/ t...@tt1.org
I have seen that Conway paper mentioned also---and I
thought it was Wolfram's 1983 paper which was more cited.
So moving on... I have seen you define Penrose tilings as
pretty much the same thing as cellular autmata, I believe.
I wrote the following statement based on a post by Lyman Hurd,
which indicated my opinion that Penrose tiles could be transposed
or translated into Wang tiles which can simulate a UTM:
> On Sun, 21 Jul 2002, Stephen Harris wrote:
> > Penrose tiles have been proven to have Universal Turing Machine
> > equivalence, I think.
>
Chris Hillman replied:
> No! No! No! Certainly not!
>
SH: Chris explained maybe I was thinking of Wang tiles, which I was,
but I thought Penrose tiles could be translated into 16 or 24 Wang tiles.
Maybe this Wang tile capability of universal computability is not preserved?
I had read this post by Lyman Hurd (fairly expert) which said:
It was the finiteness of state space hypothesis that I had guessed at but
wanted to make certain about. I should mention a result of Kari, Culik,
and me (in Ergodic Thy & Dyn Sys) which states that a CA can satisfy:
1) The only temporally periodic point is a fixed point (null, zero,
quiscent, dead, an ex-parrot,...).
2) All finite periodic lattices evolve to this fixed point.
3) The rule has non-trivial dynamics.
To be specific about point 3, first not every configuration goes off to
zero (Kari has shown this fact to be undecidable for a general 1D rule.
For 2D and above this statement is also true the proof is much simpler.)
Furthermore the rule constructed has positive topological entropy and
(therefore) a non-trivial non-wandering set. Basically, within the given
framework one can make the rule as complex as one could want (for example
computationally universal). {SH: part of the original post}
^^^^^^^^^^^^^^^^^^^
We call such a beast an ``aperiodic'' CA, and it has a neat relationship
with Penrose Tiles (via a set of 16 Wang tiles appearing in Grunbaum &
Shephard's book).
SH: So I thought the aperiodic Penrose tiles were just another type
of aperiodic CA with the potential for a computationally universal rule,
which I think has to do with the sequence of assembly. Since Chris
Hillman is pretty expert I guess I dont know something basic?
I think I will post this under a new topic in case someone else looks.
Non Specific?
Stephen
Stephen Harris
"Kent Paul Dolan" <xant...@well.com> wrote in message
news:a3eaa964.02072...@posting.google.com...
> "Stephen Harris" <stephen....@worldnet.att.net> wrote, and wrote,
> and wrote:
> > "Kent Paul Dolan" <xant...@well.com> wrote:
>
> > Wolfram said that Scientific Index of CA papers increased dramatically
> > about the time of his early paper around 1983.
>
> And also, apparently, took credit to his paper for that increase, an
> incredible display of hubris.
>
There are arrogant statements it is unwise to make even if true.
> > If you or other hobbyists did not write papers then the increase
> > of CA papers did not come from hobyyists. It came from people
> > who do write scientific papers.
>
> But we are looking for causes, not origins.
>
And here I thought you could find the first cause at the origin.
Thanks for the clarification.
> > I mean this is just repeating what you say below. (cheaper) home
> > computers helped hobbyists write programs not write papers.
> > The papers came from academics or scientists who did not
> > need cheap home computers to run their programs, though this
> > may have marginally helped them.
>
> > I disputed this remark by Tim Tyler:
> > :There *was* an explosion of interest in CAs in the early 1980s -
> > : but the factor was rather obviously cheap home computers -
> > :and not Wolfram.
>
> Yep, and you _still_ miss Tim's point. Wolfram may have published a
> paper, but Worfram's paper probably had next to nothing to do with the
> effusion of papers about CAs. That was more likely due to all the
> hobbiests doing CAs at home, some of whom then went to grad school and
> wrote papers in an area they already knew and loved, not from reading
> Wolfram's paper, but from enjoyable personal experience. Just because
> _one_ hobbiest didn't become a paper writer doesn't imply that _no_
> hobbiests became paper writers. You are indulging in a fallacy of
I did not miss Tim's point. His point was due to "cheap home computers".
You are construing his point to mean: because there were cheap home
computers, teenagers experimented with Life and then went to college
and wrote CA papers a few years later. I used the word "rare" which
does not mean "_no_" one. My usage of the word rare is in agreement
with your description of the paper production of hobbyists within the
early 80's time frame, if you can remember that far back.
> logic identified before the beginning of the common era: "after this,
> therefore because of this" doesn't work. Given a choice between
> "after one Wolfram paper, a deluge of CA papers", or "after thousands
> of hobbiests get experience programming CAs at home, a deluge of CA
> papers", the time ordering doesn't help, you need to use other methods
> of reasoning.
>
Now, that seems like really good advice which I wish I had used earlier.
> > Wolfram wrote an important paper about classification which is
> > cited all the time and which was adopted by many of the other
> > CA researchers who do read papers and write them.
>
> Bully for him, especially since stuff posted here continually calls
> Wolfram's classification scheme into doubt. That still doesn't
> promote others to write about CAs, even if, in writing about CAs, they
> then go on to quote Wolfram's paper as for a while the only available
> recent work: that _requirement_ for citation of recent work is how the
> academic publishing game works, and again has nothing to do with
> Wolfram's paper being causal for the outburst of CA efforts.
>
I see you are capable of deep insights; have you had any stimulation
to augment your powers of perception? Speaking of causality, your
first sentence implies the present criticism of Wolfram's classification
scheme
occurring 10 or 15 years later means that his scheme was not important
at the time it was published even though that scheme is still widely used.
Is that a case of reversible cellular automata, your rebirth of history?
> > It is these other researches who get tallied on the Scientific
> > Index. Wolfram said that because the Scientific Index went up
> > after his paper was read (by the people who do write papers)
> > that this indicated his paper (and him by default were _influential_).
>
> Which was incredibly arrogant of him, given the much more likely real
> cause, of which he was surely fully aware, wouldn't you say?
^^^^^^
>
Would would that cause be: cheap home computers any teenager could
program and experience the joys of simulating Life, who then go onto
college and write CA papers; or perhaps wealthy hobbyists who give up
business and return to college which enables them to write CA papers?
> [remainder of blather snipped unread; you really need to learn how to
> say what you mean in 100 words or less]
>
A word to the wise is sufficient. If I followed your example I could say
what I
mean without any words at all since you have transcended conventional
wisdom.
--
Perspicuously perspicacious,
Providential Stephen Paul
Nicholas Geovanis
> "Tim Tyler" <t...@tt1.org> wrote in message news:Gzr3...@bath.ac.uk...
> >
> > : Are you aware of somebody who wrote an important paper at that
> > : time which would stimulate interest in CA? Usually the number of
> > : citations is taken to reflect influence.
> >
> > Berlekamp, Conway and Guy's 1982 epic proof that Life was
> > capable of supporting universal computation and self-reproduction
> > is what springs to mind.
>
> I have seen that Conway paper mentioned also---and I
> thought it was Wolfram's 1983 paper which was more cited.
>
> So moving on...
Translation:
Even though I previously insisted that Conway did nothing important with
CA in the 1980s (just played Life recreationally), and even though I also
insisted that Wolfram was much more cited than Conway or anyone else on
the subject of CA since 1980, and even though I have now been shown to be
incorrect on both counts.....I hereby ignore those three facts and switch
to another subject.
Not to change the subject :-) but since we're reminiscing:
I wrote my first Life program in 1976 for an HP 9730 (if I remember
the model number right), a desktop lab computer which ran a ROM Basic and
had 8K of RAM, a one-char-line LED display and a big toilet-paper-roll of
thermal paper in the top. When I got to college in 1978 I "ported" the
code to an HP-2000(?) timesharing system which supported about 35
teletypes. When I got my first mainframe system-janitor job in 1982, I
rewrote the program in PL/I with a 3270 display for the "graphics". It
was just then that I saw an Apple ][ implementation for the first time,
inspired by articles in Byte and Dr. Dobbs I believe.
Throughout this period, the Life enthusiasts in my little backwater
discussed and sometimes implemented 2D CA rulesets other than Life.
Gardner had briefly discussed them in his first Scientific American column
on Life in 1970. We also branched-out into 1D and took some stabs at 3D
shortly after 1982 or so. Moreover, excluding Burks' 1966 book, the first
academic work in broader CA which we heard about was Toffoli, not Wolfram.
Honor compels me to state that I haven't touched a mainframe in 12 years.
:-)
Oh incidentally, the very first "personal computer" which I saw (not
counting the HP machine during high-school) was a Radio Shack machine at a
university in 1978. So I'm skeptical of your earlier claim that academics
belatedly came-to-terms with the "personal computer revolution".
My recollection is that they were the first ones there.
> Stephen
* Nick Geovanis The very term 'icon' has been appropriated and
| IT Computing Svcs changed radically in our computer age, signifying
| Northwestern Univ an ultimately unreal, 'virtual' world.
| n-geo...@nwu.edu - Metropolitan Iakovos, Hierarch of Chicago
+------------------->
Stephen Harris
"Nicholas Geovanis" <nic...@merle.acns.nwu.edu> wrote in message
news:Pine.HPX.4.10.102072...@merle.acns.nwu.edu...
> On Wed, 24 Jul 2002, Stephen Harris wrote:
>
> > "Tim Tyler" <t...@tt1.org> wrote in message news:Gzr3...@bath.ac.uk...
> > >
> > > : Are you aware of somebody who wrote an important paper at that
> > > : time which would stimulate interest in CA? Usually the number of
> > > : citations is taken to reflect influence.
> > >
> > > Berlekamp, Conway and Guy's 1982 epic proof that Life was
> > > capable of supporting universal computation and self-reproduction
> > > is what springs to mind.
> >
> > I have seen that Conway paper mentioned also---and I
> > thought it was Wolfram's 1983 paper which was more cited.
> >
> > So moving on...
>
> Translation:
> Even though I previously insisted that Conway did nothing important with
> CA in the 1980s (just played Life recreationally), and even though I also
> insisted that Wolfram was much more cited than Conway or anyone else on
> the subject of CA since 1980, and even though I have now been shown to be
> incorrect on both counts.....I hereby ignore those three facts and switch
> to another subject.
>
It happened this way, Tim Tyler wrote:
> I think I'll have to leave the matter there.
You left this part out of my post. My response was intended to defer
to what I perceived as Tim Tyler's decision not to continue any more
argumentation. So I politely changed the subject to something I was
interested in as I had grown tired of people like you and Kent Paul Dolan.
I could have countered that with Conway invented Life around 1970
and it was very popular. Why should a paper written 12 years later
(and our focus was on papers as causal influences on the growth of CA)
that concerned a rather specific area of CA and a proof of Life's
universal computability ignite interest in the broader field? I think the
paper was important to Life enthusiasts but not to the field of CA.
Not hugely, as von Neumann had already proven certain CAs capable
of universal computation in the 40's or early 50's.
In case you get lost, these are distinct issues: how much Life contributed
to interest in CA vs. what is under discussion, how much a proof of
Life=universal computability did to stimulate the entire field of CA
which lead to the production of more CA papers. Did you know
Penrose found a solution to a two-hundred year old tiling problem in
about 1976. Two tiles to tile the plane aperiodically. This was also
published in Scientific American about 1977. Conway and Penrose
also published a similar proof. It occurs at the right time so as to be
a possible inspiration to CA. I dont think universal proofs are always
that motivational. The proof of rule 110 being universal didn't shake
any foundations. The discussion was turning into too much just opinion.
I do not defend Wolfram for his arrogance. But I see him as having
more intellectual integrity than most of his detractors. Translation:
Snipping Tim saying "I'll have to leave the matter there" as a
reasonable context to produce a decision to of "so moving on"
so you could fantasize your crafty little supercilious innuendo.
So I wasn't stipulating that Tim was right, but feeling he had a right
to his own opinion. Besides Tim wrote: "I'm very happy that Wolfram
performed the work that he did - but the drawing of graphs in an
attempt to prove just how significant that work was leaves me cold."
Which I did agree with and thought was a good way to let the topic drop.
Your statement summary was also factually incorrect about belittling Conway:
NG wrote:
"Even though I previously insisted that Conway did nothing important with
CA in the 1980s (just played Life recreationally)"
SH wrote:
: If you remove Conway, not all the other people added together would
: equal Wolfram's influence because the influence of the old names had
: disappeared... CA was a nearly dead field.
That seems to me to directly contradict your description of
how important I thought Conway was. I removed Conway
precisely because I did think he was an important possible
candidate. Actually the only one I could see.
NG wrote:
"and even though I also insisted that Wolfram was much more cited than
Conway or anyone else on the subject of CA since 1980, and even though
I have now been shown to be incorrect on both counts...
..I hereby ignore those three facts"
Berlekamp, Conway and Guy - Winning Ways, vol 2
81 citations (1982) http://citeseer.nj.nec.com/context/49420/
Wolfram: Universality and complexity in cellular automata:
61 citations (1984) http://citeseer.nj.nec.com/context/59484/0
Wolfram (Editor): Theory and applications of cellular automata:
89 citations (1986) http://citeseer.nj.nec.com/context/59486/0
That book by Conway has a lot to do with mathematical game theory
and analysis of different games. Besides not having cellular automata
in the title people could site this book when their papers were not
specific to cellular automata. You would expect a couple of books
which covered areas outside of cellular automata to attract more
attention because there are alternate reasons for citing the book.
So this example is not a very pure link between Conway and a
paper, actually book, that he wrote creating interest in the specific
area of CA. The title : Universality and complexity in cellular automata
the link to CA is not obscured by other possibilities. I didn't want
to nitpick with the objections becoming increasingly abstract/subjective.
I think the _book_ that most created interest in computers and
AI (related to Alife) is Hofstadter's Goedel Escher and Bach. '79-80
So that is why I changed the subject, Tim didn't want to continue
the discussion and I was also tired of it. Not because I couldn't
have countered. Your summary is also directly contradicted
by what I have actually written and quoted. It is factually wrong.
Your post makes it really hard to imagine why Wolfram keeps his
own counsel, why he doesn't value peer review. So I am interested
in other stuff now and if you want to believe that I cannot withstand
the pressure of your penetrating keen analysis, go right ahead.