Cognitive Architecture
Arthur T. Murray
AI has been solved with a cognitive architecture based on
making a genuine effort to understand the mind as a whole,
http://mentifex.virtualentity.com/theory5.html - AI Theory.
Tristan Miller
> http://freshmeat.net/projects/ai/#comment-30449 shows how
> AI has been solved with a cognitive architecture
http://freshmeat.net/projects/ai/#comment-29709 shows how the above claim
is utter nonsense.
--
_
_V.-o Tristan Miller [en,(fr,de,ia)] >< Space is limited
/ |`-' -=-=-=-=-=-=-=-=-=-=-=-=-=-=-= <> In a haiku, so it's hard
(7_\\ http://www.nothingisreal.com/ >< To finish what you
ian glendinning
I'm in no position, with no reason, to defend Mentifex (Arthur T
Murray), but I find the efforts (you have) taken to refute
scientifically something which is "only common sense" as an
interesting case of scientific fundamentalism in its own right.
Fascinating. Paraphrasing Susan Blackmore in "Waking from the Meme
Dream", logical positivism (scientific method) is itself a meme.
I'm genuinely not taking sides here and happy to believe your main
point that Mentifex has high nuisance and noise value for people going
about their serious scientific business. Apologies if I've
inadvertantly spread that particular virus, but the case provides
interest from a neutral perspective.
(BTW I have no scientific AI credentials either, so feel free to
discount my opinions.)
Ian Glendinning
www.psybertron.org
Tristan Miller
In article <f7b2e276.04041...@posting.google.com>, ian
glendinning wrote:
> I'm in no position, with no reason, to defend Mentifex (Arthur T
> Murray), but I find the efforts (you have) taken to refute
> scientifically something which is "only common sense" as an
> interesting case of scientific fundamentalism in its own right.
"Common sense" to those with some knowledge of AI or cognitive science,
perhaps, but not necessarily to the average programmer or hobbyist who
might visit these newsgroups. One need only examine Murray's copious
posting history to see the confusion he often causes. Murray's claim to
have solved AI is akin to the dozens of kooks out there who claim to have
invented perpetual motion machines. Some physics-challenged clerks have
sadly given credibility to these men by granting them patents. Such false
laurels are subsequently used by the inventors to dupe ignorant laymen
into investing (read: squandering) thousands of dollars on their nonsense.
Regards,
Tristan
Michael Mendelsohn
> In article <4081...@news.victoria.tc.ca>, Arthur T. Murray wrote:
> > http://freshmeat.net/projects/ai/#comment-30449 shows how
> > AI has been solved with a cognitive architecture
>
> http://freshmeat.net/projects/ai/#comment-29709 shows how the above claim
> is utter nonsense.
It doesn't show any such thing.
I think his claims are overstated; the system does not work yet because
key components such as "volition" and "emotion" have not been
implemented; the theoretical background is hard to understand because
ATM is not in line with the conventions of academic AI research. Thus,
his work is of limited interest; and the way he promotes it makes him
(seem?) a kook.
However, that is no grounds to pronounce it "utter nonsense".
Anyhow, a man that promotes an internet TLD .jam for impromtu music and
fruit preserves can't be all bad. :) I also liked his interview on
http://www.nanomagazine.com/i.php?id=01_10_24 .
OTOH, your FAQ on him and your in-group behaviour shows you seem to be
stalking the guy; your web site indicates you enjoy putting down the
underdog, and that you are generally not the most sensitive person.
You're not even polite enough to indicate you redirected the Follow-Up
to your post.
Michael
[Groups trimmed, F'up set to poster]
--
Feel the stare of my burning hamster and stop smoking!
Tristan Miller
In article <40839AA2...@msgid.michael.mendelsohn.de>, Michael
Mendelsohn wrote:
>> > AI has been solved with a cognitive architecture
>>
>> the above claim is utter nonsense.
>
> I think his claims are overstated; the system does not work yet because
> key components such as "volition" and "emotion" have not been
> implemented[...]
> However, that is no grounds to pronounce it "utter nonsense".
If neither in his original message, nor in the sites it linked to, nor in
the thirty years he has been adovacting Mentifex, no evidence has been
produced that his program "solves AI", then I fail to see how the claim
can be considered anything other than nonsense.
> OTOH, your FAQ on him and your in-group behaviour shows you seem to be
> stalking the guy;
I consider pointers to the FAQ a public service to newbies, in much the
same way that regulars of comp.lang.c and sci.math repeatedly post troll
alerts in response to the posts of E. Robert Tisdale and James Harris,
respectively. Harris in particular has a number of websites devoted to
debunking his so-called proofs of Fermat's Last Theorem. People in other
groups Murray haunts are posting links to the FAQ in an effort to deter
others from taking his posts at face value.
Anyway, it's not as though I actively seek out messages written by Murray.
If he happens to post his claptrap in a newsgroup I am reading, then I
will respond to it with a short pointer. I'm sure he has free reign to
spread his misinformation unchecked on Slashdot and other places I seldom
post to.
> your web site indicates you enjoy putting down the underdog
Some of the stories there may suggest that, though I believe those are
uniformly old and no longer representative of my current feelings. They
remain on the website due to popular request. Now that you bring it up,
though, I should probably add a notice to this effect.
> and that you are generally not the most sensitive person.
That may be; I've heard this accusation before.
> You're not even polite enough to indicate you redirected the Follow-Up
> to your post.
I wasn't aware that netiquette mandated this; I always viewed it as an
optional courtesy to the reader. Besides, surely this discussion has more
to do with (alleged) kookery than with computer science...? (Followups
set appropriately.)
Regards,
Tristan
Hank Oredson
in message news:40839AA2...@msgid.michael.mendelsohn.de...
> Tristan Miller wrote in comp.theory:
> > In article <4081...@news.victoria.tc.ca>, Arthur T. Murray wrote:
> > > http://freshmeat.net/projects/ai/#comment-30449 shows how
> > > AI has been solved with a cognitive architecture
> >
> > http://freshmeat.net/projects/ai/#comment-29709 shows how the above
claim
> > is utter nonsense.
>
> It doesn't show any such thing.
Oh?
> I think his claims are overstated; the system does not work yet because
"yet"?
> key components such as "volition" and "emotion" have not been
Many hundreds or perhaps thousands of other key components
are also missing. The components that do exist ... do not work.
> implemented; the theoretical background is hard to understand because
> ATM is not in line with the conventions of academic AI research. Thus,
> his work is of limited interest; and the way he promotes it makes him
> (seem?) a kook.
> However, that is no grounds to pronounce it "utter nonsense".
I'll have a go at it (again). It is utter nonesense.
> Anyhow, a man that promotes an internet TLD .jam for impromtu music and
> fruit preserves can't be all bad. :) I also liked his interview on
> http://www.nanomagazine.com/i.php?id=01_10_24 .
>
> OTOH, your FAQ on him and your in-group behaviour shows you seem to be
> stalking the guy; your web site indicates you enjoy putting down the
> underdog, and that you are generally not the most sensitive person.
> You're not even polite enough to indicate you redirected the Follow-Up
> to your post.
The amount of spam from him has at times almost killed
several very useful newsgroups. That spam has hardly
changed in the many years I've been avoiding it.
I'm retired. I worked in the field (AI) for awhile, long enough to recognize
nonesense when it is presented. Have seen better work from High School
kids who wrote D&D code. Think about that a bit ... D&D is just as "AI"
as what Mentifex presents, but better because it actually DOES something.
> Michael
> [Groups trimmed, F'up set to poster]
> --
> Feel the stare of my burning hamster and stop smoking!
Now I will go away. This is probably the first time I've
responded in a "Mentifex" thread for ten years or more ;-)
It is best not to feed the trolls, but sometimes it is fun.
--
... Hank
http://horedson.home.att.net
http://w0rli.home.att.net
Michael Mendelsohn
> "Michael Mendelsohn" <keine.Wer...@msgid.michael.mendelsohn.de> wrote
> in message news:40839AA2...@msgid.michael.mendelsohn.de...
> > Tristan Miller wrote in comp.theory:
> > > In article <4081...@news.victoria.tc.ca>, Arthur T. Murray wrote:
> > > > http://freshmeat.net/projects/ai/#comment-30449 shows how
> > > > AI has been solved with a cognitive architecture
> > >
> > > http://freshmeat.net/projects/ai/#comment-29709 shows how the above
> > > claim is utter nonsense.
> > It doesn't show any such thing.
> Oh?
Yes. It claims that his software doesn't work, but makes no statement
whether his theory would be able to "solve AI", were it implemented. The
ATM FAQ is a bit more broad on this, but that comment is definitely not
showing that to me.
> > I think his claims are overstated; the system does not work yet because
> "yet"?
OK, replace "yet" with "yet, if ever," ;)
> > key components such as "volition" and "emotion" have not been
>
> Many hundreds or perhaps thousands of other key components
> are also missing. The components that do exist ... do not work.
Which, in my eyes, is a testament to ATM's lack of programming prowess,
which covers up any merits or deficiciencies his theory might have.
> > implemented; the theoretical background is hard to understand because
> > ATM is not in line with the conventions of academic AI research. Thus,
> > his work is of limited interest; and the way he promotes it makes him
> > (seem?) a kook.
> > However, that is no grounds to pronounce it "utter nonsense".
>
> I'll have a go at it (again). It is utter nonesense.
Let me clarify: The claim "AI has been solved" is utter nonsense because
it carries no meaning. ATM's work contains some good (read: accepted
lore) bits, and as such, is not utter nonsense.
But feel free to pronounce all you want. :)
> The amount of spam from him has at times almost killed
> several very useful newsgroups.
Because people would not use killfiles?
> It is best not to feed the trolls, but sometimes it is fun.
Yes. :)
Michael