Who Knew ?

16 views
Skip to first unread message

1p166

unread,
Oct 21, 2021, 12:06:52 AM10/21/21
to
Ah ... a Dr. Who #3 ... computers with lots of big
spinney tape drives and circuits big enough to fix
with solder and jumper-wires :-)

Pancho

unread,
Oct 21, 2021, 3:27:05 AM10/21/21
to
I learnt, circa #3, that a computer with computational intelligence
comparable to a human brain would be the size of London.

An older me would figure they would have had a serious problem with
communication between the widely separated parallel computational units,
but what do I know compared to the Doctor.

Bobbie Sellers

unread,
Oct 21, 2021, 11:04:51 AM10/21/21
to
The Time Lords conceal there most advanced technology behind a screen
of obsolescence. But they also know what impresses humans in
the era they recruit from.

"Human brain computational power" is used to impress us with our
brain power. And a Tardis has the equivalent of a erratic Time Lord's
brain plus more storage. This is an entertainment series not a factual
presentation.

"Gordian Protocol" by Weber and Holo is a rip-roaring fantasy
based on the idea the human race last to the 30th Century and masters
time travel based on an incorrect TOE. Because of the errors in their
TOE they are splitting the timeline which gets all messed up. Hitler
Must be saved from a 1940 assassination attempt which succeeded and
led to the Great Eastern War destroying the USSR. But they had
back accidents with AI now restricted and with nano-machines used
as deadly weapons. Oh and it is a book.


bliss - lost in spacetime and quantal interactions on a macroscopic
scale.

--
bliss dash SF 4 ever at dslextreme dot com

1p166

unread,
Oct 28, 2021, 1:11:12 AM10/28/21
to
On 10/21/21 3:27 AM, Pancho wrote:
> On 21/10/2021 05:06, 1p166 wrote:
>> Ah ... a Dr. Who #3 ... computers with lots of big
>> spinney tape drives and circuits big enough to fix
>> with solder and jumper-wires  :-)
>
> I learnt, circa #3, that a computer with computational intelligence
> comparable to a human brain would be the size of London.

Probably including the suburbs too ...

Thing is, nobody knew The Trick then, and
still don't. "Consciousness" or even a really
good emulation is STILL well beyond the horizon.

There's some kind of quasi-chaotic thing that
creates "consciousness", a weird pattern. We
don't seem very close to finding it.

Emulating it in hard/soft-ware even further.

There are lots of humans with hydrocephalic brain
damage - not much cortex left - who STILL have average
or even superior IQ. This does suggest "pattern"
and "connections" over pure neuron numbers.

A few years ago there was a Brit sci-fi series
about sophisticated, but ultimately mindless,
quality androids. Their inventor impressed a
few with a sort of fractal thought pattern
that produced intelligence, "consciousness",
"self-awareness", "real people", as an emergent
property. I don't think the writers were so
far off.

Never diss emergent properties of complex systems.

> An older me would figure they would have had a serious problem with
> communication between the widely separated parallel computational units,
> but what do I know compared to the Doctor.

The data transmission issue IS real.

The old Cray supercomputers, "These ARE the small
ones", were hand-wired for the fastest point-2-point
data transmission. It was an issue even way back then.
The speed of electricity, even light, was an issue 30
years ago. We kinda got around it by packing more and
more onto single multi-cpu chips. Short xmission
distances. But, there are limits ...

Neurons are SLOW, very slow, compared to electronics.
Yet they manage to produce "self", "self-awareness".
in a sub-cubic-foot unit. It's the pattern they work
upon that's important.

To sum it up, the hardware is probably capable of
"self" even now, or in the near future. But "The
Trick" eludes us.

1p166

unread,
Oct 28, 2021, 2:08:51 AM10/28/21
to
"Dr. Who" was and is pure fiction ... albeit usually
contaminated with the latest science facts.

I came in on the back edge of those big boxes and
spinney tape drives. Ever had to thread one of those
tape drives, punched/fed a paper tape, dropped your
wunnerful FORTRAN programs punch-cards, actually
used a TTY terminal ? I have.

I'm getting near retirement. For "fun" I've migrated
a few indespensible utility programs to COBOL and
FORTRAN and PASCAL and K&R 'c'. I do like Pascal ...
and it lives on in Delphi and Lazarus/FPC. If it needs
a nice GUI like right now, it's Delphi/Lazarus. If it
needs to be portable, Lazarus absolutely. GTK and such,
HORRIBLE by comparison - ten times the work and much
more version-hostage. The newbies will just have to
suck up a dose of The Past :-)

As for the premise of "Dr. Who" ... it's rather
interesting really. Paradox resolution was really
not dealt with because it'd usually ruin the stories.
In any case, those old old series bring back some
memories about Tech.

Now if you want really CRAP science, view the MARVEL
action movies. Newton is spinning in his grave. Sorry,
but if you're in a suit of armor and slammed up against
a wall at hundreds of feet per second you're SPAM IN
THE CAN laddies ... oh, and super-beings CANNOT lift
a tank .. the leverage equation does NOT work. :-)

One of my favorite, abeit obscure, movies is
"Colossus, The Forbin Project". It's about two
military supercomputers that force everyone to
provide interconnection - and the result is
enough IQ to take over the world Big Time.
Not sure how 70's tech would create AI, but ...

Anyway, it was the Clarke/Minsky vision at
the time.

Minsky, BTW, USED to post to Usenet, back when
it was respectable. Lots of Top People did.
And then ...

The Natural Philosopher

unread,
Oct 28, 2021, 4:39:58 AM10/28/21
to
On 28/10/2021 06:11, 1p166 wrote:
> Never diss emergent properties of complex systems.

Well indeed, Of course as everybody knows, the world *as we know it* is
an emergent property of consciousness...

I trump your Cartesian Realism with Transcendental Idealism!


--
"The most difficult subjects can be explained to the most slow witted
man if he has not formed any idea of them already; but the simplest
thing cannot be made clear to the most intelligent man if he is firmly
persuaded that he knows already, without a shadow of doubt, what is laid
before him."

- Leo Tolstoy

The Natural Philosopher

unread,
Oct 28, 2021, 4:41:36 AM10/28/21
to
On 28/10/2021 07:08, 1p166 wrote:
> Minsky, BTW, USED to post to Usenet, back when
>   it was respectable. Lots of Top People did.
>   And then ...

...they all moved to facebook...

--
You can get much farther with a kind word and a gun than you can with a
kind word alone.

Al Capone


Pancho

unread,
Oct 28, 2021, 9:25:43 AM10/28/21
to
On 28/10/2021 06:11, 1p166 wrote:

>
>   A few years ago there was a Brit sci-fi series
>   about sophisticated, but ultimately mindless,
>   quality androids. Their inventor impressed a
>   few with a sort of fractal thought pattern
>   that produced intelligence, "consciousness",
>   "self-awareness", "real people", as an emergent
>   property. I don't think the writers were so
>   far off.
>

Yep it was called Humans, the series touched on the watershed moment of
android development, the singularity...

The point where androids will be able to satisfy the "Pancho" test. The
test will be passed test if a standard examiner is unable to tell if he
has shagged a real woman or an android.

FWIW the series was originally Swedish "Real Humans". The first series
was good, but went rapidly downhill afterwards.

The Natural Philosopher

unread,
Oct 28, 2021, 9:49:14 AM10/28/21
to
On 28/10/2021 14:25, Pancho wrote:

>
> The point where androids will be able to satisfy the "Pancho" test. The
> test will be passed test if a standard examiner is unable to tell if he
> has shagged a real woman or an android.

My ex passed that years before I left her...



--
"Women actually are capable of being far more than the feminists will
let them."


Bobbie Sellers

unread,
Oct 28, 2021, 11:15:34 AM10/28/21
to
On 10/27/21 22:11, 1p166 wrote:
> On 10/21/21 3:27 AM, Pancho wrote:
>> On 21/10/2021 05:06, 1p166 wrote:
>>> Ah ... a Dr. Who #3 ... computers with lots of big
>>> spinney tape drives and circuits big enough to fix
>>> with solder and jumper-wires  :-)
>>
>> I learnt, circa #3, that a computer with computational intelligence
>> comparable to a human brain would be the size of London.
>
>   Probably including the suburbs too ...
>
>   Thing is, nobody knew The Trick then, and
>   still don't. "Consciousness" or even a really
>   good emulation is STILL well beyond the horizon.
>
>   There's some kind of quasi-chaotic thing that
>   creates "consciousness", a weird pattern. We
>   don't seem very close to finding it.

We are unable to examine the wetware while it is
functioning, except in rare circumstances such as brain
surgery then only get surface expressions so that we know
one lobe is connected to vocalization and another to
movement of one limb or the other. We can figure out
where memory resides but no the exact storage methods.

>
>   Emulating it in hard/soft-ware even further.
>
>   There are lots of humans with hydrocephalic brain
>   damage - not much cortex left - who STILL have average
>   or even superior IQ. This does suggest "pattern"
>   and "connections" over pure neuron numbers.

Absolutely and why? More neurons in the autonomic
portion on your CNS than anywhere else so subcubic foot may
not apply. I was astounded to learn of the people with hardly
any brain left and actually met online a person with such
damage with lots of symptoms but whho managed to work for a
living.
But the development of self and self awareness
is doubtless of survival value as is the development of
the enlarged brain and the specialised part like the
neo-cortex. However equally important I am sure is the
input from the senses in the rest of the body.

But maybe it lies in the unreliabily of the
mechanisms which maintain and edit memories so that
we recall our failures with great precision but our
successes are usually and in memory inflated.

>
>   A few years ago there was a Brit sci-fi series
>   about sophisticated, but ultimately mindless,
>   quality androids. Their inventor impressed a
>   few with a sort of fractal thought pattern
>   that produced intelligence, "consciousness",
>   "self-awareness", "real people", as an emergent
>   property. I don't think the writers were so
>   far off.
>
>   Never diss emergent properties of complex systems.

Never doubt it. Most of my software is smarter than
me in some senses but no hardware ever hit it own power on
button.
>
>> An older me would figure they would have had a serious problem with
>> communication between the widely separated parallel computational
>> units, but what do I know compared to the Doctor.

Quite a bit but you are confined to the moment and to the 4 dimen
>
>   The data transmission issue IS real.
>
>   The old Cray supercomputers, "These ARE the small
>   ones", were hand-wired for the fastest point-2-point
>   data transmission. It was an issue even way back then.
>   The speed of electricity, even light, was an issue 30
>   years ago. We kinda got around it by packing more and
>   more onto single multi-cpu chips. Short xmission
>   distances. But, there are limits ...
>
>   Neurons are SLOW, very slow, compared to electronics.
>   Yet they manage to produce "self", "self-awareness".
>   in a sub-cubic-foot unit. It's the pattern they work
>   upon that's important.
>
>   To sum it up, the hardware is probably capable of
>   "self" even now, or in the near future. But "The
>   Trick" eludes us.

Electronic hardware might have enough computing
power but it lacks sensory input affording survival value and
all the millions of years of refinement of the integrated
body and mind by the genus homo, all for survival and reproduction.
Including the development of simple technologies such as fire
which permitted cooking which permitted the size of the jaw
to decrease and the size of the cranium to enlarge.

bliss

1p166

unread,
Oct 29, 2021, 1:08:04 AM10/29/21
to
Ah yes, "Humans" ... but I think it died after the
second season. IMHO it'd done all it could do with
that particular premise.

Never saw the Swedish version.

But the idea that "sentience", "personhood", was
less pure hardware than it was a way of USING
said hardware ... a pattern, a meme ... struck
a chord. There's something to that.

I mentioned hydrocephalic children for a reason.
Some have surprisingly little grey matter left.
Their scans reveal devastation, what's left of
their cortexes is more on the scale of chimps.
Yet, many attained normal+ capabilities. This
points to an organizational/connectional key,
beyond mere hardware.

There are several beasties with bigger brains than
ours - but they just don't have "it". Whatever "it"
is seems to have showed up around the time of Homo
Hablis. H.Sapiens ... maybe not so much that we were
generally intellectually superior, but just "badder"
than the competitors. Quick, aggressive, somewhat
xenophobic - we dominated the "trolls" and "ogres"
and such in the end. Resistance is futile. Then we
turned on each other ... which also seems to be a
significant evolutionary prompter. Having to outwit
your own - or else - that's evolutionary pressure.

ANYway ... if we can find "it", that pattern, real
AI might be possible with a lot less hardware than
we currently envision.

1p166

unread,
Oct 29, 2021, 1:15:04 AM10/29/21
to
On 10/21/21 11:04 AM, Bobbie Sellers wrote:
> On 10/21/21 00:27, Pancho wrote:
>> On 21/10/2021 05:06, 1p166 wrote:
>>> Ah ... a Dr. Who #3 ... computers with lots of big
>>> spinney tape drives and circuits big enough to fix
>>> with solder and jumper-wires  :-)
>>
>> I learnt, circa #3, that a computer with computational intelligence
>> comparable to a human brain would be the size of London.
>>
>> An older me would figure they would have had a serious problem with
>> communication between the widely separated parallel computational
>> units, but what do I know compared to the Doctor.
>
>     The Time Lords conceal there most advanced technology behind a
> screen of obsolescence.  But they also know what impresses humans in
> the era they recruit from.
>
>     "Human brain computational power" is used to impress us with our
> brain power. And a Tardis has the equivalent of a erratic Time Lord's
> brain plus more storage.  This is an entertainment series not a factual
> presentation.

Duh ....


>     "Gordian Protocol" by Weber and Holo is a rip-roaring fantasy
> based on the idea the human race last to the 30th Century and masters
> time travel based on an incorrect TOE.  Because of the errors in their
> TOE they are splitting the timeline which gets all messed up.  Hitler
> Must be saved from a 1940 assassination attempt which succeeded and
> led to the Great Eastern War destroying the USSR.  But they had
> back accidents with AI now restricted and with nano-machines used
> as deadly weapons. Oh and it is a book.
>
>
> bliss - lost in spacetime and quantal interactions on a macroscopic
> scale.
>

Even time-travel fiction reveals serious problems.
Time-travel FACT would be much worse.

Fortunately, it doesn't look like time travel can
be realized in the real world.

1p166

unread,
Oct 29, 2021, 1:36:59 AM10/29/21
to
On 10/28/21 11:15 AM, Bobbie Sellers wrote:
> On 10/27/21 22:11, 1p166 wrote:
>> On 10/21/21 3:27 AM, Pancho wrote:
>>> On 21/10/2021 05:06, 1p166 wrote:
>>>> Ah ... a Dr. Who #3 ... computers with lots of big
>>>> spinney tape drives and circuits big enough to fix
>>>> with solder and jumper-wires  :-)
>>>
>>> I learnt, circa #3, that a computer with computational intelligence
>>> comparable to a human brain would be the size of London.
>>
>>    Probably including the suburbs too ...
>>
>>    Thing is, nobody knew The Trick then, and
>>    still don't. "Consciousness" or even a really
>>    good emulation is STILL well beyond the horizon.
>>
>>    There's some kind of quasi-chaotic thing that
>>    creates "consciousness", a weird pattern. We
>>    don't seem very close to finding it.
>
>     We are unable to examine the wetware while it is
> functioning, except in rare circumstances such as brain
> surgery then only get surface expressions so that we know
> one lobe is connected to vocalization and another to
> movement of one limb or the other.  We can figure out
> where memory resides but no the exact storage methods.


In part it's the "quantum problem" ... all ways of
observing the system in action ALTER the system.
It's a perpetually moving target.


>>
>>    Emulating it in hard/soft-ware even further.
>>
>>    There are lots of humans with hydrocephalic brain
>>    damage - not much cortex left - who STILL have average
>>    or even superior IQ. This does suggest "pattern"
>>    and "connections" over pure neuron numbers.
>
>     Absolutely and why?  More neurons in the autonomic
> portion on your CNS than anywhere else so subcubic foot may
> not apply.  I was astounded to learn of the people with hardly
> any brain left and actually met online a person with such
> damage with lots of symptoms but whho managed to work for a
> living.


Clearly you need "enough" neurons. But how they all
connect, the patterns, are the all-important feature.
Even a city sized computer, lacking "it", the proper
organizational meme, wouldn't achieve "personhood".


>     But the development of self and self awareness
> is doubtless of survival value as is the development of
> the enlarged brain and the specialised part like the
> neo-cortex.  However equally important I am sure is the
> input from the senses in the rest of the body.
>
>     But maybe it lies in the unreliabily of the
> mechanisms which maintain and edit memories so that
> we recall our failures with great precision but our
> successes are usually and in memory inflated.
>
>>
>>    A few years ago there was a Brit sci-fi series
>>    about sophisticated, but ultimately mindless,
>>    quality androids. Their inventor impressed a
>>    few with a sort of fractal thought pattern
>>    that produced intelligence, "consciousness",
>>    "self-awareness", "real people", as an emergent
>>    property. I don't think the writers were so
>>    far off.
>>
>>    Never diss emergent properties of complex systems.
>
>     Never doubt it.  Most of my software is smarter than
> me in some senses but no hardware ever hit it own power on
> button.


Oh, THAT's easily fixed. :-)
You can arrange for the "sensory input" both in virtual
emulations or real-world-walking-around apps.

The trick is "it" ... the way to organize thinking to
achieve "personhood". We don't know what "it" is. If
we did, proper AI could surely be achieved with a lot
less hardware than we currently believe.

The current approaches to AI involve emulating neural
networks. It does bear some fruit - IF you can bring
enough hardware and speed into the equation. The thing
is that electronic circuits are NOT organic brain tissue.
Trying to do it "our way" may not be the best approach.
Organics and electronics each have their own set of
strengths and weaknesses. Maybe doing it "our way" is
a big WASTE of cycles ?

The Natural Philosopher

unread,
Oct 29, 2021, 4:59:42 AM10/29/21
to
On 29/10/2021 06:36, 1p166 wrote:
> The trick is "it" ... the way to organize thinking to
>   achieve "personhood". We don't know what "it" is. If
>   we did, proper AI could surely be achieved with a lot
>   less hardware than we currently believe.

Well we do. Or some people do.

At some point in childhood intelligence organises sensory data into a
model, that includes a self, in a real (physical) world. That model is
reinforced through parents etc until people like you think that they
have actually emerged into 'the real world' and start to explain their
awareness of it, on terms of its physical nature!

Circular argument.

Awareness itself implies an entity that is aware, and something other
than itself it becomes aware of. And some invisible intangible entity
that makes it all happen.

Cf Schopenhauer 'The world as Will, and Representation'.

Personhood emerges as half of the pair of the 'observer' and the
'observed' - the nature of the world you end up perceiving is the nature
of the human being that observes it.

The real question, is there anything beyond that that would allow you to
choose what person you are, and hence subtly alter the nature of the
world you perceive?

Many would assert that there is, and that is what 'it' is...

--
“Those who can make you believe absurdities, can make you commit
atrocities.”

― Voltaire, Questions sur les Miracles à M. Claparede, Professeur de
Théologie à Genève, par un Proposant: Ou Extrait de Diverses Lettres de
M. de Voltaire

Bobbie Sellers

unread,
Oct 29, 2021, 11:13:52 AM10/29/21
to
We are unable to discuss the curious matter of "personhood"
with creatures on this planet with larger brains due to disparate
means of communnication, In an Antarctic dive the diver was
adopted by a Leopard seal which may have seen it as an injured
member of the young of its own species. We have heard many tales
of dolphins helping people who fall into waters and as well tales
of dolphine raping their own species using gang tactics and
seemingly attempting seduction of human females,

So don't dismiss the possibily of self-hood existing
in the so-called lower animals. I think that some self-awareness
is a survival mechanism as are these paragraphs.

And we do a lot with the neurons in the autonomic
nervous system which the the brain connection to the mostly
"unconscious" functions which keep us more or less "alive".
>
>   ANYway ... if we can find "it", that pattern, real
>   AI might be possible with a lot less hardware than
>   we currently envision.

bliss - 'Nearly any fool can use a Linux computer. Many do.' After all
here I am...

Charlie Gibbs

unread,
Oct 29, 2021, 12:39:50 PM10/29/21
to
On 2021-10-28, The Natural Philosopher <t...@invalid.invalid> wrote:

> On 28/10/2021 07:08, 1p166 wrote:
>
>> Minsky, BTW, USED to post to Usenet, back when
>>   it was respectable. Lots of Top People did.
>>   And then ...
>
> ...they all moved to facebook...

I was trying to think of a clever comeback but you
beat me to it. I don't know whether to laugh or cry.

--
/~\ Charlie Gibbs | Life is perverse.
\ / <cgi...@kltpzyxm.invalid> | It can be beautiful -
X I'm really at ac.dekanfrus | but it won't.
/ \ if you read it the right way. | -- Lily Tomlin

Charlie Gibbs

unread,
Oct 29, 2021, 12:52:40 PM10/29/21
to
On 2021-10-29, The Natural Philosopher <t...@invalid.invalid> wrote:

> On 29/10/2021 06:36, 1p166 wrote:
>
>> The trick is "it" ... the way to organize thinking to
>>   achieve "personhood". We don't know what "it" is. If
>>   we did, proper AI could surely be achieved with a lot
>>   less hardware than we currently believe.
>
> Well we do. Or some people do.
>
> At some point in childhood intelligence organises sensory data into a
> model, that includes a self, in a real (physical) world. That model is
> reinforced through parents etc until people like you think that they
> have actually emerged into 'the real world' and start to explain their
> awareness of it, on terms of its physical nature!

Unfortunately, many people's models are seriously warped.

Charlie Gibbs

unread,
Oct 29, 2021, 12:52:41 PM10/29/21
to
On 2021-10-29, 1p166 <z24ba6.net> wrote:

> Fortunately, it doesn't look like time travel can
> be realized in the real world.

Too bad. I'd love go to back and see to it that
Bill Gates' parents never met.

Bobbie Sellers

unread,
Oct 29, 2021, 4:08:18 PM10/29/21
to
On 10/29/21 09:52, Charlie Gibbs wrote:
> On 2021-10-29, 1p166 <z24ba6.net> wrote:
>
>> Fortunately, it doesn't look like time travel can
>> be realized in the real world.
>
> Too bad. I'd love go to back and see to it that
> Bill Gates' parents never met.



Without MS-DOS and Windows we might not have
the x86 line with all its advances and multiple cores
to play with as well as all its sneaky Minux
implementation.

Time Travel by any means in the 21st century is
merely entertainment but by the 30th century it may be
dangerous reality. I am not going to worry about it but
The Gordian Protocol was too good a story to miss.

bliss

Charlie Gibbs

unread,
Oct 29, 2021, 6:43:31 PM10/29/21
to
On 2021-10-29, Bobbie Sellers <bl...@mouse-potato.com> wrote:

> On 10/29/21 09:52, Charlie Gibbs wrote:
>
>> On 2021-10-29, 1p166 <z24ba6.net> wrote:
>>
>>> Fortunately, it doesn't look like time travel can
>>> be realized in the real world.
>>
>> Too bad. I'd love go to back and see to it that
>> Bill Gates' parents never met.
>
> Without MS-DOS and Windows we might not have
> the x86 line with all its advances and multiple cores
> to play with as well as all its sneaky Minux
> implementation.

Yeah, we'd probably have had to put up with the 68000 line.
<crocodile tears>

And without MS-DOS and Windows we might have gotten some
real work done rather than spending our time working around
all their warts.

> Time Travel by any means in the 21st century is
> merely entertainment but by the 30th century it may be
> dangerous reality. I am not going to worry about it but
> The Gordian Protocol was too good a story to miss.

My favourite time-travel warning story is
_The Brooklyn Project_ by William Tenn.

1p166

unread,
Oct 30, 2021, 1:49:26 AM10/30/21
to
On 10/29/21 12:52 PM, Charlie Gibbs wrote:
> On 2021-10-29, The Natural Philosopher <t...@invalid.invalid> wrote:
>
>> On 29/10/2021 06:36, 1p166 wrote:
>>
>>> The trick is "it" ... the way to organize thinking to
>>>   achieve "personhood". We don't know what "it" is. If
>>>   we did, proper AI could surely be achieved with a lot
>>>   less hardware than we currently believe.
>>
>> Well we do. Or some people do.
>>
>> At some point in childhood intelligence organises sensory data into a
>> model, that includes a self, in a real (physical) world. That model is
>> reinforced through parents etc until people like you think that they
>> have actually emerged into 'the real world' and start to explain their
>> awareness of it, on terms of its physical nature!
>
> Unfortunately, many people's models are seriously warped.

Agreed. "It" is NOT going to be easy, NOT anything
derived by simple models. Darwin (eventually) did
it - but only after 4 billion years, and sort of
by accident. Pea-brained is the natural norm,
JUST enough to find the next thing to eat.

Back in the day, Minsky was predicting proper AI
was imminent. In Society Of Mind he went on and
on about how super-simple circuits could do IF/THEN
logical evaluations. What he DIDN'T consider was
the petabytes of crap needed to prep the data for
that final IF/THEN circuit. The HAL-9000 was a
product of Minski's optimism. Sorry, many decades
later and nothing even CLOSE.

Marv USED to post on Usenet ... before it became
disrespectable .....

Anyway, I strongly intuit that "It" is out there.
Once you have the necessary volume of hardware the
trick is how to organize it to produce the desired
effect. Nature offers some clues, but perhaps the
best way might not to be emulating nature so closely,
making fake neural nets. Perhaps hardware as it is
offers different routes to a similar result ?

The Natural Philosopher

unread,
Oct 30, 2021, 2:39:08 AM10/30/21
to
On 29/10/2021 17:39, Charlie Gibbs wrote:
> On 2021-10-28, The Natural Philosopher <t...@invalid.invalid> wrote:
>
>> On 28/10/2021 07:08, 1p166 wrote:
>>
>>> Minsky, BTW, USED to post to Usenet, back when
>>>   it was respectable. Lots of Top People did.
>>>   And then ...
>>
>> ...they all moved to facebook...
>
> I was trying to think of a clever comeback but you
> beat me to it. I don't know whether to laugh or cry.
>
Try both.

Old age is the time of final disillusionment, when all your idols have
feet of clay, and you aren't really sure what the point of worshipping
them in the first place was.


--
How fortunate for governments that the people they administer don't think.

Adolf Hitler

The Natural Philosopher

unread,
Oct 30, 2021, 2:53:28 AM10/30/21
to
On 29/10/2021 17:52, Charlie Gibbs wrote:
> On 2021-10-29, The Natural Philosopher <t...@invalid.invalid> wrote:
>
>> On 29/10/2021 06:36, 1p166 wrote:
>>
>>> The trick is "it" ... the way to organize thinking to
>>>   achieve "personhood". We don't know what "it" is. If
>>>   we did, proper AI could surely be achieved with a lot
>>>   less hardware than we currently believe.
>>
>> Well we do. Or some people do.
>>
>> At some point in childhood intelligence organises sensory data into a
>> model, that includes a self, in a real (physical) world. That model is
>> reinforced through parents etc until people like you think that they
>> have actually emerged into 'the real world' and start to explain their
>> awareness of it, on terms of its physical nature!
>
> Unfortunately, many people's models are seriously warped.
>
Warped with respect to what?

A couple of decades inquiry into the matter shows that there is no way
to establish what the One True Model of Reality looks like. And those
that think they are in possession of it are no less deluded than anyone
else.

My conclusion is that *any* model will do, provided it is not so
dysfunctional as to result in death before procreation.

I.e. Darwin rules: The world we think we live in, is simply a view that
has developed over thousands of years in the struggle to survive. Hence
my support for conservatism. "It worked for me da, so it will probably
work for me"

Critical Rat Theory, which believes that rats in fact are superior, but
have been held back by white male dominated worldviews, is specious
because it doesn't matter if the laws by which we live were dreamed up
by a conman with a beard who was good at stone masonry, or handed down
by sympathetic aliens.

If it works, use it, is the old adage :-)

The only concession to 'warpage' I will allow, is that if people are
living miserably because they cling to beliefs that make them angry,
miserable, jealous and full of shame, well perhaps they should consider
abandoning the belief set and replacing it with something else, just as
lacking in truth content, but more palatable.

The Natural Philosopher

unread,
Oct 30, 2021, 2:59:35 AM10/30/21
to
On 30/10/2021 06:49, 1p166 wrote:
> Anyway, I strongly intuit that "It" is out there.
>   Once you have the necessary volume of hardware the
>   trick is how to organize it to produce the desired
>   effect. Nature offers some clues, but perhaps the
>   best way might not to be emulating nature so closely,
>   making fake neural nets. Perhaps hardware as it is
>   offers different routes to a similar result ?

Well it could be argued that that is simply and example of a seriously
warped worldview that thinks that software is an emergent property of
hardware.

When there is just as good a case for saying that human wise, the
hardware is an emergent property of the software.

And the so called 'material world' is simply a GUI that we have
*constructed* in order to make it easier to interface with 'whatever the
fuck it really is' (to paraphrase Kant)

Quantum physics not only does not refute this proposition, it actually
reinforces it.




--
"Anyone who believes that the laws of physics are mere social
conventions is invited to try transgressing those conventions from the
windows of my apartment. (I live on the twenty-first floor.) "

Alan Sokal

The Natural Philosopher

unread,
Oct 30, 2021, 3:00:12 AM10/30/21
to
On 29/10/2021 17:52, Charlie Gibbs wrote:
> On 2021-10-29, 1p166 <z24ba6.net> wrote:
>
>> Fortunately, it doesn't look like time travel can
>> be realized in the real world.
>
> Too bad. I'd love go to back and see to it that
> Bill Gates' parents never met.
>
And end up in a world run by IBM and Gary Kildall?

The Natural Philosopher

unread,
Oct 30, 2021, 3:21:44 AM10/30/21
to
You can't solve the inherent paradox.

Ability to change the past would always result in a paradox. There are
no paradoxes in nature. Only in the imperfect human mind.

It wasn't about time travel, but it was about personality travel -
Mindswap, by Robert Sheckley, in which the hapless protagonist keeps
swapping bodies in an vain attempt to get back to the one he started
with - ends thus....

"He changed his mind, however, realizing that there was no sense in
spending his life trying to discover if he had a life to spend.

"Besides, there was a possibility that even if the Earth were changed,
his memory and perceptions might also be changed, rendering discovery
impossible.

"He lay beneath Stanhope's familiar green sky, and considered the
possibility. It seemed unlikely: for did not the giant oak trees still
migrate each year to the south? Did not the huge red sun move across
the sky pursued by its dark companion? Did not the triple moons return
each month with their new accumulation of comets?

"These familiar sights reassured him. Everything seemed to be as it
always had been. And so , willingly and with a good grace, Marvin
accepted his world at face value, married Marsha Baker, and lived
forever after."

In short, if time travel existed, we would travel in time, change the
past, and instantly be in a present descended from the changed past with
no knowledge of ever having changed it.

In fact I do this a thousand times a second :-)

Stéphane CARPENTIER

unread,
Oct 30, 2021, 9:45:26 AM10/30/21
to
Le 29-10-2021, Bobbie Sellers <bl...@mouse-potato.com> a écrit :
>
> Without MS-DOS and Windows we might not have
> the x86 line with all its advances and multiple cores
> to play with as well as all its sneaky Minux
> implementation.

Can you show me references about Bill Gates explaining how to create
microprocessors? And once you provide this, if it's not in your
references, you have to explain why nobody outside Bill Gates would have
been able to find it out.

If you can't, you are just a daydreamer.

--
Si vous avez du temps à perdre :
https://scarpet42.gitlab.io

Stéphane CARPENTIER

unread,
Oct 30, 2021, 5:47:05 PM10/30/21
to
Le 30-10-2021, Andreas Kohlbach <a...@spamfence.net> a écrit :
> On 30 Oct 2021 13:45:22 GMT, Stéphane CARPENTIER wrote:
>>
>> Le 29-10-2021, Bobbie Sellers <bl...@mouse-potato.com> a écrit :
>>>
>>> Without MS-DOS and Windows we might not have
>>> the x86 line with all its advances and multiple cores
>>> to play with as well as all its sneaky Minux
>>> implementation.
>>
>> Can you show me references about Bill Gates explaining how to create
>> microprocessors? And once you provide this, if it's not in your
>> references, you have to explain why nobody outside Bill Gates would have
>> been able to find it out.
>
> Where did he state that?

He doesn't, that's why he's wrong. It's the only way to say that without
Microsoft the processors would lag behind what we have today.

I don't know why He me answered on my email instead of here, but he gave
no more reason. Just his history with OSes but it's irrelevant.

> Suppose meant was that without the IBM PC (and thus MS-DOS) the x86 might
> just had been a footnote in today's computer history. We might all use
> the Motorola M70000 or the Zilog 64-bit line of microprocessors today.

Maybe we would have something else, the point is not to know which
processor we would use, but to know if the processors would be as
powerful as what we have today.

And the answer is: yes, of course.

> Anyway, Gates said it the same day he invented the internet and

By the guy who invented the internet, you are speaking of Bill Gates,
not someone else? Since when does he invent internet?

> mentioned, that no one would ever need more than 640 KB of computer
> memory.

Making this mistake didn't change history, it only render the use of DOS
more difficult.

Richard Kettlewell

unread,
Oct 30, 2021, 6:14:46 PM10/30/21
to
Stéphane CARPENTIER <s...@fiat-linux.fr> writes:
> Le 30-10-2021, Andreas Kohlbach <a...@spamfence.net> a écrit :
>> Suppose meant was that without the IBM PC (and thus MS-DOS) the x86
>> might just had been a footnote in today's computer history. We might
>> all use the Motorola M70000 or the Zilog 64-bit line of
>> microprocessors today.
>
> Maybe we would have something else, the point is not to know which
> processor we would use, but to know if the processors would be as
> powerful as what we have today.
>
> And the answer is: yes, of course.

Agreed.

But I think there’s a good chance that even without the IBM PC and
MSDOS, x86 would have become as dominant as it is today. The factors
that made IBM choose the 8088 (price, availability, sympathy to existing
components) were relevant to everyone else too.

--
https://www.greenend.org.uk/rjk/

Computer Nerd Kev

unread,
Oct 30, 2021, 8:18:13 PM10/30/21
to
Charlie Gibbs <cgi...@kltpzyxm.invalid> wrote:
> On 2021-10-29, The Natural Philosopher <t...@invalid.invalid> wrote:
>>
>> At some point in childhood intelligence organises sensory data into a
>> model, that includes a self, in a real (physical) world. That model is
>> reinforced through parents etc until people like you think that they
>> have actually emerged into 'the real world' and start to explain their
>> awareness of it, on terms of its physical nature!
>
> Unfortunately, many people's models are seriously warped.

I think the truth was out back with Dr. Who #1 in The War Machines,
where in a somewhat similar plot to The Green Death the WOTAN
computer was able to hypnotise anyone within earshot.

Clearly computers have been hypnotising us since the sixties to
make us build far more of them than we could possibly need, and
more powerful than could possibly do us any good. After all, before
then lots of important people knew that we'd never need anything
like this many of them. :)

I don't usually go for fan fiction, but I think I'd give a Dr Who
+ The Matrix cross-over a look...

"The Boss" in The Green Death was built out of an ICT 1301, by the
way:
http://www.starringthecomputer.com/feature.html?f=875

--
__ __
#_ < |\| |< _#

The Natural Philosopher

unread,
Oct 30, 2021, 9:25:43 PM10/30/21
to
I thought they chose the 8086?


--
In todays liberal progressive conflict-free education system, everyone
gets full Marx.

1p166

unread,
Oct 31, 2021, 12:06:29 AM10/31/21
to
Had to look that up ... BritBox. ONE Mhz clock speed, magcore
memory, 48 bit words. Not quite in the IBM 360 universe. 21
clock cycles to do addition. About 400 words of main memory.

Sure ... you're going to build a world-dominating machine
intelligence on THAT platform :-)

But, in the day, the experts were SURE it could be done.
The "HAL-9000" was a product of that optimism.

Then horrible horrible REALITY hit ...

50 years later and not even 1/1000th of a HAL.

Robert Riches

unread,
Oct 31, 2021, 12:12:34 AM10/31/21
to
Not for the original IBM PC. The original IBM PC and the XT each
had a 8-bit data bus. The AT had a 16-bit data bus. For a
while, there were two kinds of PeeCee hardware: 1) 8088 with 8-bit
data bus; 2) 80286 with 16-bit data bus.

More than a couple of decades ago, I observed an interesting
"competition" architected by manager who was a solid Macintosh
fan. The secretaries (now administrative assistants) got to try
out a PeeCee-type machine (most likely a PeeCee clone) and a Mac
to decided which was better for their work. Somewhere, he had
found a PeeCee with unconventional hardware. I think it was an
8086 machine (same interface to software as 8088 but 16-bit bus),
but it might have been a 286-based machine with an 8-bit data
bus. Either way, when they tried to find an extended or expanded
memory card, they had major problems, because the only types of
cards were for 1) 8088 interface to software and 8-bit bus; or 2)
286 interface to software with 16-bit bus. Because of all of the
difficulties caused by that choice of non-standard hardware, the
PeeCee lost the competition.

--
Robert Riches
spamt...@jacob21819.net
(Yes, that is one of my email addresses.)

1p166

unread,
Oct 31, 2021, 12:28:42 AM10/31/21
to
But can you CHANGE it ? PROGRAM it ? Make it do more than
the apps you downloaded ? :-)

Ask people how car works. They'll say "You turn the key
and it starts". The reality is that the sci/tech/engineering
from the tires up in to MAKING it work is absolutely
brain-boggling. THAT'S how a car works.

I've taken a few stabs at figuring out how CPUs actually
work. Sorry, but I lose it at the decoder/ALU steps. The
people who figured that out were GENIUS. Maybe when I'm
retired I'll have the time to study it, follow the
reasoning until I could actually build a Z-80 or 6502.
I'd be happy with an i4004.

Thing is, the original designers were under PRESSURE,
"We need this by November or we'll all go broke" kind
of pressure.

1p166

unread,
Oct 31, 2021, 12:57:37 AM10/31/21
to
There were a number of good 8/16 chips out at the time
and the M68000 was on the very near horizon.

The ONLY reason the Intel chip became so popular was
because IBM picked it. It looked a little familiar
to the CP/M crowd (I actually HAVE CP/M-86 in a VM).
Lots of registers, segmented memory. Semi-familiar
sorts of instructions. IBM is BIG ... and meant
"Business" ...

What I always wanted was a SAGE computer ... looked
like an IBM box but with a 68k chip and a choice of
operating systems (most UNIX-like). Alas they were
a low-volume outfit and the price was too high.

And MAC ... there's always been something about Apple
operating systems that turned me off. The thinking
and I just aren't on the same frequency, so to speak.

So far as PRICING ... this is where Intel (and Gates)
took their big risks. They underpriced on the theory
that an IBM unit would sell big enough to let them
make it up on volume.

Stéphane CARPENTIER

unread,
Oct 31, 2021, 6:26:08 AM10/31/21
to
Le 31-10-2021, Andreas Kohlbach <a...@spamfence.net> a écrit :
> Should I really had added a smiley? Thought it was obvious enough I
> didn't meant it.

I was surprised by your answer, but sometimes I read impressive stuff by
people who rewrite the past that I wasn't sure. More precisey, for the
second sentence, I knew it was a joke, but I wasn't sure for the first.

The Natural Philosopher

unread,
Oct 31, 2021, 9:24:28 AM10/31/21
to
On 31/10/2021 04:28, 1p166 wrote:
> I've taken a few stabs at figuring out how CPUs actually
>   work. Sorry, but I lose it at the decoder/ALU steps. The
>   people who figured that out were GENIUS. Maybe when I'm
>   retired I'll have the time to study it, follow the
>   reasoning until I could actually build a Z-80 or 6502.
>   I'd be happy with an i4004.

Really? Its all pretty easy since the basic logic is simply gates, and
bistables.

Once you have - and I have - been shown how to combine gates to make
things like shift registers and adders and RAM, it isn't really that hard.



--
"Strange as it seems, no amount of learning can cure stupidity, and
higher education positively fortifies it."

- Stephen Vizinczey

The Natural Philosopher

unread,
Oct 31, 2021, 9:35:51 AM10/31/21
to
On 31/10/2021 04:57, 1p166 wrote:
> On 10/30/21 6:14 PM, Richard Kettlewell wrote:
>> Stéphane CARPENTIER <s...@fiat-linux.fr> writes:
>>> Le 30-10-2021, Andreas Kohlbach <a...@spamfence.net> a écrit :
>>>> Suppose meant was that without the IBM PC (and thus MS-DOS) the x86
>>>> might just had been a footnote in today's computer history. We might
>>>> all use the Motorola M70000 or the Zilog 64-bit line of
>>>> microprocessors today.
>>>
>>> Maybe we would have something else, the point is not to know which
>>> processor we would use, but to know if the processors would be as
>>> powerful as what we have today.
>>>
>>> And the answer is: yes, of course.
>>
>> Agreed.
>>
>> But I think there’s a good chance that even without the IBM PC and
>> MSDOS, x86 would have become as dominant as it is today. The factors
>> that made IBM choose the 8088 (price, availability, sympathy to existing
>> components) were relevant to everyone else too.
>
>   There were a number of good 8/16 chips out at the time
>   and the M68000 was on the very near horizon.
>
>   The ONLY reason the Intel chip became so popular was
>   because IBM picked it. It looked a little familiar
>   to the CP/M crowd (I actually HAVE CP/M-86 in a VM).

More than that, operating in small model mode, you could pretty much run
8080 code through a translator and port CP/M programs to it easily.
The business market had been taken by the 8080/z80 and CP/M while the
hobbysist were all using 6502s.

IBM naturally picked a zilog/interl style architecture.
Ghastly mistake until the 386 came out.


>   Lots of registers, segmented memory. Semi-familiar
>   sorts of instructions. IBM is BIG ... and meant
>   "Business" ...
>
>   What I always wanted was a SAGE computer ... looked
>   like an IBM box but with a 68k chip and a choice of
>   operating systems (most UNIX-like). Alas they were
>   a low-volume outfit and the price was too high.
>
>   And MAC ... there's always been something about Apple
>   operating systems that turned me off. The thinking
>   and I just aren't on the same frequency, so to speak.
>
Agreed.

>   So far as PRICING ... this is where Intel (and Gates)
>   took their big risks. They underpriced on the theory
>   that an IBM unit would sell big enough to let them
>   make it up on volume.
>

Underpriced? you could write MSDOS in a month or two. It wasn't worth
anything. More code in the average BIOS!

original PC had 8K BIOS.

And only 16K RAM!!!!

I wrote a BIOS AND a basic OS for a bare metal 8086 AND ported FORTH to
it in three months


--
"When one man dies it's a tragedy. When thousands die it's statistics."

Josef Stalin

Bobbie Sellers

unread,
Oct 31, 2021, 10:55:13 AM10/31/21
to
No but in reality or fiction you use imagination to extend
the capabilities of a microscopic amoeba to the Blob or your old
computer to dominate the world.

>
>   But, in the day, the experts were SURE it could be done.
>   The "HAL-9000" was a product of that optimism.
>
>   Then horrible horrible REALITY hit  ...
>
>   50 years later and not even 1/1000th of a HAL.

Do you really want a HAL who remember refused to open the
Pod Door. Better build a Slave AI that takes the safety of its
humans as primary importance. The Human model for AI is as
flawed as human people are. And HAL which we do not have yet
is apparently capable of having a paranoid reaction or xenophobia.

Have you thought of Watson? Surely that agglomoration of
hard and software approaches the 1/1000th of a HAL or even
better machine. After all when the very distant ancestor decided
to leave the trees for the plain and stand on two Legs to look
around that was a very unpromising beginning. Maybe it was a
mistake. Definitely living too close to the shoreline or rivers
was a mistake and we have that ingrained habit.


bilss - -“Nearly any fool can use a GNU/Linux computer. Many do.” After
all here I am...


Andreas Kohlbach

unread,
Oct 31, 2021, 3:47:15 PM10/31/21
to
On Sun, 31 Oct 2021 13:35:46 +0000, The Natural Philosopher wrote:
>
> On 31/10/2021 04:57, 1p166 wrote:
>>>
>>> But I think there’s a good chance that even without the IBM PC and
>>> MSDOS, x86 would have become as dominant as it is today. The factors
>>> that made IBM choose the 8088 (price, availability, sympathy to existing
>>> components) were relevant to everyone else too.
>>   There were a number of good 8/16 chips out at the time
>>   and the M68000 was on the very near horizon.
>>   The ONLY reason the Intel chip became so popular was
>>   because IBM picked it. It looked a little familiar
>>   to the CP/M crowd (I actually HAVE CP/M-86 in a VM).
>
> More than that, operating in small model mode, you could pretty much
> run 8080 code through a translator and port CP/M programs to it
> easily.
> The business market had been taken by the 8080/z80 and CP/M while the
> hobbysist were all using 6502s.

Hmm. If you consider the "bedroom coders" in the UK hobbyists - they
mainly coded on the ZX Spectrum (may some on the ZX81/80 before), which
has a Z80 CPU.

OK, there were may using a C64 (6510, similar to a 6502) and the Oric,
which sold reasonably well in the UK and France back in the day.

But considering me as hobbyist back in the 1980s I indeed started to code
in assembler on a 6502 (C64).
--
Andreas

1p166

unread,
Oct 31, 2021, 10:47:08 PM10/31/21
to
That's the thing, they ARE going to keep at it even
if it takes another five decades. Then we are faced
with alien-ish intelligences that, like we, could
easily rationalize their way around any "laws".

So long as they don't have bodies ... but they will
pretty soon. We will design/build them, or THEY will.


> Better build a Slave AI that takes the safety of its
> humans as primary importance.

I don't think that's possible. Once you make proper
intelligence, 'self', it WILL go its own way. The
very complexity of intelligence negates the ability
to have total control.

Our best hope would be that they self-evolve so
quickly that they lose all interest in we petty
organics and move on to Big Stuff.

>  The Human model for AI is as
> flawed as human people are.  And HAL which we do not have yet
> is apparently capable of having a paranoid reaction or xenophobia.

"Just Like Us" would be the WORST scenerio - we KNOW
what humans are like ... and it ain't good.

However I think "Not QUITE Human" would be the easiest
to achieve. If you want pure clones, there are - um -
more conventional low-tech ways to do that. The hypothetical
HAL learned human-ish mannerisms, but it's life experience
and physical realities meant it arrived at its generalizations
and conclusions by a quite different path.

If you want practical "alien-ness", consider dolphins.
PROBABLY as intelligent as we - but an entirely different
evolutionary and individual experience. About 50 years
of trying and we STILL can't do their language. We know
from statistical analysis that they DO have complex
conversations, but WHAT ? And these are fellow mammals
not all THAT big an evolutionary leap away from ourselves.

They may as well be aliens - and our failures to grasp
what they say does NOT bode very well should proper
aliens drop down from the skies. It's more than just
language, it's the mode of THINKING behind it.
Of all the space-people movies, only "Arrival" gave
a partway glimpse of this issue.

>     Have you thought of Watson? Surely that agglomoration of
> hard and software approaches the 1/1000th of a HAL or even
> better machine.

"Watson" is impressive ... within its sphere. It's got
random little BITS of human-level IQ in there, but it
is still a shattered mirror. The bits can't come
together to realize "I AM", not in the slightest degree.
The engineers will keep adding bits for awhile, but in
the end it'll be a dead end and they will move on to
different, more promising, paradigms.

>After all when the very distant ancestor decided
> to leave the trees for the plain and stand on two Legs to look
> around that was a very unpromising beginning.  Maybe it was a
> mistake.  Definitely living too close to the shoreline or rivers
> was a mistake and we have that ingrained habit.

Those ancestors, well, likely the trees left THEM.
There was a lot of climate change. They had no choice
in certain locales. Barely worked out for them ...

Little groups, isolated and inbred for a time - which
amplifies certain genes - then meet and mate the neighbors.
Repeat, repeat, repeat. Somewhere a few genes related to
brain development/size were mutated and it was a USEFUL
mutation for once. Finally showed around H.hablis when
there was a noteworthy deviation from the usual ratio
of brain size to body mass and the toolkits suddenly
got bigger and more sophisticated.

As for shorelines and rivers, and esp where both converge,
was likely devastating as the last ice-age ended. How many
nascent civilizations were washed away or drowned under
hundreds of feet of water ?

1p166

unread,
Oct 31, 2021, 10:52:56 PM10/31/21
to
Ask Bill's LAWYERS and they'll probably assert he DID
invent the internet and computers and Linux and most
everything else so he can sue competitors into oblivion :-)

Gates is a very sharp computer guy - but his real gift
is BUSINESS. That little clause he sneaked in to his
old contract with IBM was pure genius.

The Natural Philosopher

unread,
Nov 1, 2021, 12:44:56 AM11/1/21
to
On 31/10/2021 19:47, Andreas Kohlbach wrote:
> On Sun, 31 Oct 2021 13:35:46 +0000, The Natural Philosopher wrote:
>>
>> On 31/10/2021 04:57, 1p166 wrote:
>>>>
>>>> But I think there’s a good chance that even without the IBM PC and
>>>> MSDOS, x86 would have become as dominant as it is today. The factors
>>>> that made IBM choose the 8088 (price, availability, sympathy to existing
>>>> components) were relevant to everyone else too.
>>>   There were a number of good 8/16 chips out at the time
>>>   and the M68000 was on the very near horizon.
>>>   The ONLY reason the Intel chip became so popular was
>>>   because IBM picked it. It looked a little familiar
>>>   to the CP/M crowd (I actually HAVE CP/M-86 in a VM).
>>
>> More than that, operating in small model mode, you could pretty much
>> run 8080 code through a translator and port CP/M programs to it
>> easily.
>> The business market had been taken by the 8080/z80 and CP/M while the
>> hobbysist were all using 6502s.
>
> Hmm. If you consider the "bedroom coders" in the UK hobbyists - they
> mainly coded on the ZX Spectrum (may some on the ZX81/80 before), which
> has a Z80 CPU.

Most UK 'home' computers were *not* based on a z80.

Sinclair came very late to the party.

First micro I saw was altair 8800 - s100 bus. 8080. That was serious .
1974 or thereabouts

The Apple 1 was around 1973, 6502 again

Then the Apple II, PET and trash 80 came a couple of years later.

Only the trash 80 was z80. But it could be used in business.

At that time the split was clear. CP/M was for business and ran on
Z80s/8080s.

6502s were for hobbyists writing in basic and assembler.

As for 6809s - great chip. No one really used it.

IBM knew hardware. The processors really didn't matter, what mattered
was software for it.

And Gates understood one thing, the sofware quality didn't matter - what
matteerd was getting everybody to use it to the exclusions of anything else.

>
> OK, there were many using a C64 (6510, similar to a 6502) and the Oric,
> which sold reasonably well in the UK and France back in the day.
>
> But considering me as hobbyist back in the 1980s I indeed started to code
> in assembler on a 6502 (C64).
>

Exactly. Wasn't Apple II a 6502 as well?

Andreas Kohlbach

unread,
Nov 1, 2021, 1:52:20 PM11/1/21
to
On Mon, 1 Nov 2021 04:44:51 +0000, The Natural Philosopher wrote:
>
> On 31/10/2021 19:47, Andreas Kohlbach wrote:
>> On Sun, 31 Oct 2021 13:35:46 +0000, The Natural Philosopher wrote:
>>>
>>> More than that, operating in small model mode, you could pretty much
>>> run 8080 code through a translator and port CP/M programs to it
>>> easily.
>>> The business market had been taken by the 8080/z80 and CP/M while the
>>> hobbysist were all using 6502s.
>> Hmm. If you consider the "bedroom coders" in the UK hobbyists - they
>> mainly coded on the ZX Spectrum (may some on the ZX81/80 before), which
>> has a Z80 CPU.
>
> Most UK 'home' computers were *not* based on a z80.
>
> Sinclair came very late to the party.
>
> First micro I saw was altair 8800 - s100 bus. 8080. That was serious
> . 1974 or thereabouts

"Home computers" are described from any micro as the Altair 8800
(designed 1974 but showed up in January 1975 to start the craze). True,
that one had a 8080.

> The Apple 1 was around 1973, 6502 again

It was released 1976. The 6502 itself is from 1975. About 200 Apple 1
were produced, making it a collector's item today. Only with the Apple 2
a year later they produced large quantities.

> Then the Apple II, PET and trash 80 came a couple of years later.

1977.

> Only the trash 80 was z80. But it could be used in business.

I think the TRS-80 can also be considered a non-business computer.

> At that time the split was clear. CP/M was for business and ran on
> Z80s/8080s.

UK "Home micros" with a Z80 (ZX Spectrum, Amstrad CPCs, ...) where not
shipped with CP/M, although you could probably run it. Did this (in an
emulator) with the CPC <https://www.youtube.com/watch?v=qStVxf0XlE0>.

> 6502s were for hobbyists writing in basic and assembler.

The UK market (and that's what we're talking here about) saw more Z80
based ZX (Spectrum, 81/80) machines that Commodore 64s.

But the UK saw also a big number of Acorn computers, which ran a
6502. Those, like Apple 2s, were rather expensive that they were mainly
used in the education sector.

If you check some links of
<https://en.wikipedia.org/wiki/List_of_British_computers> it can be
noticed that most of the used a Z80.

> As for 6809s - great chip. No one really used it.

The TRS color computer and "clone" Dragon 32/64 did. Latter also sold in
numbers.

[...]

>> OK, there were many using a C64 (6510, similar to a 6502) and the
>> Oric,
>> which sold reasonably well in the UK and France back in the day.
>> But considering me as hobbyist back in the 1980s I indeed started to
>> code
>> in assembler on a 6502 (C64).
>>
>
> Exactly. Wasn't Apple II a 6502 as well?

Yes, but at least in Europe to expensive for the common user. Outside the
UK most got a C64, while in the UK Spectrums ruled the market.

F'up2 alt.folklore.computers
--
Andreas

Charlie Gibbs

unread,
Nov 1, 2021, 2:50:15 PM11/1/21
to
On 2021-11-01, 1p166 <z24ba6.net> wrote:

> Gates is a very sharp computer guy -

If by "computer guy" you mean marketing, eye candy...
anything but programming, where he's definitely mediocre.

> but his real gift is BUSINESS.

Agreed.

Charlie Gibbs

unread,
Nov 1, 2021, 2:50:15 PM11/1/21
to
On 2021-10-30, The Natural Philosopher <t...@invalid.invalid> wrote:

> On 29/10/2021 17:52, Charlie Gibbs wrote:
>
>> On 2021-10-29, The Natural Philosopher <t...@invalid.invalid> wrote:
>>
>>> At some point in childhood intelligence organises sensory data into a
>>> model, that includes a self, in a real (physical) world. That model is
>>> reinforced through parents etc until people like you think that they
>>> have actually emerged into 'the real world' and start to explain their
>>> awareness of it, on terms of its physical nature!
>>
>> Unfortunately, many people's models are seriously warped.
>
> Warped with respect to what?
>
> A couple of decades inquiry into the matter shows that there is no way
> to establish what the One True Model of Reality looks like. And those
> that think they are in possession of it are no less deluded than anyone
> else.
>
> My conclusion is that *any* model will do, provided it is not so
> dysfunctional as to result in death before procreation.

<snip>

> The only concession to 'warpage' I will allow, is that if people are
> living miserably because they cling to beliefs that make them angry,
> miserable, jealous and full of shame, well perhaps they should consider
> abandoning the belief set and replacing it with something else, just as
> lacking in truth content, but more palatable.

In some societies, this can result in death before procreation. 1/2 :-)

Charlie Gibbs

unread,
Nov 1, 2021, 2:50:16 PM11/1/21
to
On 2021-10-30, The Natural Philosopher <t...@invalid.invalid> wrote:

> In short, if time travel existed, we would travel in time, change the
> past, and instantly be in a present descended from the changed past with
> no knowledge of ever having changed it.

My favourite argument against time travel is that if it were possible,
someone would go back and change something such that it was never invented.

Charlie Gibbs

unread,
Nov 1, 2021, 2:50:19 PM11/1/21
to
On 2021-10-30, The Natural Philosopher <t...@invalid.invalid> wrote:

> On 29/10/2021 17:52, Charlie Gibbs wrote:
>
>> On 2021-10-29, 1p166 <z24ba6.net> wrote:
>>
>>> Fortunately, it doesn't look like time travel can
>>> be realized in the real world.
>>
>> Too bad. I'd love go to back and see to it that
>> Bill Gates' parents never met.
>
> And end up in a world run by IBM and Gary Kildall?

Why not? I doubt it could be worse.

--
/~\ Charlie Gibbs | Microsoft is a dictatorship.
\ / <cgi...@kltpzyxm.invalid> | Apple is a cult.
X I'm really at ac.dekanfrus | Linux is anarchy.
/ \ if you read it the right way. | Pick your poison.

1p166

unread,
Nov 2, 2021, 12:13:09 AM11/2/21
to
I remember the ads in magazines and such ... the first
"Small-Office PCs". The bits were fitted into something
the size of a large desk. S-100 mostly, 8" floppy (or
TWO if you were a successful small biz), tape if you
were a cheap-ass, 8008 chip.

The Altair was a "desktop" by comparison, and had the
new and improved 8080 chip. However they were aimed
straight at the student/hobby market and I don't
think anybody ever tried to integrate them in to
a slick "Small Business System". They WERE a sort
of "milestone" though because regular Joes could
actually kind of AFFORD one - the first real
"democratization" of PCs. Apple and Commodore
came along shortly after and blew the Altairs
out of the water.

The 6502 was designed by a bunch of defectors from
the Motorola 6800 series. That caused some legal
issues, they were originally "too like" the 6800s.
More efficient however - and cheaper to make. Soon
edged Motorola pretty much out of the 8-bit PC
market (except for the CoCo).

(Not sure if OS-9 was ever ported to the 6502, but
you COULD run it on a CoCo). OS-9 was quite UNIX-ish
but a lot more space/cycle efficient. It's still
sold - and ain't exactly cheap - mostly for use
in embedded systems, esp those that need to be
close to Real Time)

The TRS-80s were not bad computers at all. They were
one of the next steps for Small Business computers.
The CP/M was a big advantage and the units were
nicely packaged. They were fairly snappy for 8-bitters
too, not "trash" at all. And yes they were fine as
home/hobby/development PCs. Always wanted one, but
could never quite afford one. The final version had
a 68000 co-processor board in there.

The PETs were of the same paradigm as the TRS-80s,
a monitor+keyboard+mainboard in one nicely-styled
box. The first had a CRAP "chicklet" keyboard but
the follow-ons were much nicer. Could never figure
out why they built a nice box with a 99-cent keyboard.
PETs, like the TRS, were aimed at the "Small Business"
market. They offered similar performance, but except
for some one-off efforts I don't think there was a
CP/M-6502. Some dual-board models though from short-
lived companies.

I knew a guy, one of those IQ-200 on-the-edge
people, who had a computer shop, but mostly
made money writing clones of popular computer
games - in MACHINE CODE, BINARY - on a PET.
Said it "gave him a buzz" to do it that way :-)
He wasn't lying, I watched him doing it. He'd
then burn it into ROM cartridges for VIC-20s
and C-64s.

TI-99/4A ... well ... TI ruined it for themselves
by trying Apples thing of making it super-hard for
3rd party developers. Alas the actual 9900 16-bit
chip was BARELY used, 95% of the work was done by
the GPU. The 9900s were kinda strange too - a funky
hardware-based multi-user/multi-tasking setup
which stored register sets and stuff in system
memory because, at the time, it was actually
faster/cheaper than on-chip. "BLWP" - Branch
And Load WorkSpace Pointer" ... I remember
that instruction. It was never meant as a
"small business" PC, and neither were the
VICs/C64s/Ataris ... more "Game Systems Plus".
A few C64s were put to "business" USE however -
for a very long time there was a "local govt
channel" and once in a while it'd crash and
you'd see the C64 ROM BASIC error message :-)

The Brits were also players. The "BBC" computers were
pretty good - and sometimes ahead of US pop-culture
units.

But the IBM-PC murdered them all. Wasn't THAT great
of a PC, but it had the weight of IBM behind it.
Apple managed to carve out its own niche, but the
others went under eventually ... though Commodore
made a fair try with the Amigas. Those STILL have
fans and some Linux utilities STILL support the
Amiga disk formats plus there are other Amiga
support programs too. I bought the original, but
there were SO many "Guru Meditation" messages
that I dumped the thing and bought a Sanyo-550
PC (semi)-Clone.

The native 550 graphics were superior,
but I badly needed full IBM-CGA compatibility and
you had to buy a separate board and do some tricky
jumper-wire work to get that. Still a great unit,
and about a third the price of IBM. I might still
have it somewhere, under the pile-o-junk.

Anyway, the IBM clones now dominate. A few years
back DeGaulle airport, Paris, was paralyzed because
it's system for dealing with taxi-way routing went
down. Turned out it was running on a PC clone and
WINDOWS 3.11 for DECADES. Hey, if it ain't broke ...

Always wanted a SAGE computer. Looked like a PC box
but it had 68000 series chips and some kind of
XENIX-related and similar operating systems. Alas,
a small maker, expensive, low-volume, pretty quickly
went under. Too bad. Those WERE meant as "small
business" boxes. Think you can still buy them
on E-Bay, but early BAD experiences with E-Bay
and Musks PayPal put me off of them. However I
have seen working MicroVAX systems for sale there,
they were VERY good systems and the OS was well
ahead of its time, meant for medium-scale orgs
and businesses. Still have the VMS manual, four
inches thick, thin paper, smallish type. ONE day ...

1p166

unread,
Nov 2, 2021, 12:48:51 AM11/2/21
to
On 11/1/21 1:52 PM, Andreas Kohlbach wrote:

1p166

unread,
Nov 2, 2021, 1:04:14 AM11/2/21
to
On 11/1/21 2:50 PM, Charlie Gibbs wrote:
> On 2021-10-30, The Natural Philosopher <t...@invalid.invalid> wrote:
>
>> On 29/10/2021 17:52, Charlie Gibbs wrote:
>>
>>> On 2021-10-29, The Natural Philosopher <t...@invalid.invalid> wrote:
>>>
>>>> At some point in childhood intelligence organises sensory data into a
>>>> model, that includes a self, in a real (physical) world. That model is
>>>> reinforced through parents etc until people like you think that they
>>>> have actually emerged into 'the real world' and start to explain their
>>>> awareness of it, on terms of its physical nature!
>>>
>>> Unfortunately, many people's models are seriously warped.
>>
>> Warped with respect to what?
>>
>> A couple of decades inquiry into the matter shows that there is no way
>> to establish what the One True Model of Reality looks like. And those
>> that think they are in possession of it are no less deluded than anyone
>> else.
>>
>> My conclusion is that *any* model will do, provided it is not so
>> dysfunctional as to result in death before procreation.
>
> <snip>
>
>> The only concession to 'warpage' I will allow, is that if people are
>> living miserably because they cling to beliefs that make them angry,
>> miserable, jealous and full of shame, well perhaps they should consider
>> abandoning the belief set and replacing it with something else, just as
>> lacking in truth content, but more palatable.
>
> In some societies, this can result in death before procreation. 1/2 :-)

Why do you ascribe SO many negatives to those "under-achieving"
societies ?

Sorry, but to build much of anything you need a good foundation
and a path towards advancement. Not everybody HAS that. It's not
that they are stupid or hateful or angry or miserable or jealous
or any of that crap - they just lack "The Ladder" anywhere handy.

Reasons may be "cultural" or more rooted in the Leaders of the
moment or economic situation. Usually the latter two.

People are SMART everywhere. It's the ability to make good USE
of that that tends to be the problem.

Bobbie Sellers

unread,
Nov 2, 2021, 1:05:34 AM11/2/21
to
` The Apple was outside my price range as well.
My Amigas were second hand.
Years later I had credit and spent a lot to bring
the specs on the A2000b up to date with a cool running 68060
with 64 megabytes of ram, a nice video card, and scan-doubler.


>
>
>   I remember the ads in magazines and such ... the first
>   "Small-Office PCs". The bits were fitted into something
>   the size of a large desk. S-100 mostly, 8" floppy (or
>   TWO if you were a successful small biz), tape if you
>   were a cheap-ass, 8008 chip.
>
>   The Altair was a "desktop" by comparison, and had the
>   new and improved 8080 chip. However they were aimed
>   straight at the student/hobby market and I don't
>   think anybody ever tried to integrate them in to
>   a slick "Small Business System". They WERE a sort
>   of "milestone" though because regular Joes could
>   actually kind of AFFORD one - the first real
>   "democratization" of PCs. Apple and Commodore
>   came along shortly after and blew the Altairs
>   out of the water.

Well I was not paying close attention as a friend
was trying to build a CPM computer in her bedroom. Don't
know if she ever got it working. I asked her how much
memory I would need for word processing, she said 64 KB
and I grabbed a C=64 at the Pacific Stereo store for $200
and next week came back for the VIC 1541 floppy drive which
was about $220.
Commodore Business Machines was a pioneer of cost
reduction.
I had a C=64 but did not code but did do book keeping
for a man who had gotten several years behind,
I did a lot of work in PaperClip from Batteries
Included of Canada.
The C=128/64 had a chip called an 8502 to run the
Commodore side of things but CPM ran on a Zilog Z-80A.
that was built in for the purpose. I got the OS from
FOG down in Daly City. I got hooked on dungeon crawls
then and when i have time I still play Angband on whatever
it will run on in my VirtualBox.
Well I did bookkeeping and more on the Amiga
from the A1000 to the 2000b and wasted a lot of money
on them. But they were the best of the time which
passed about 1995 when the Windows got useful.

The IBM PC had is own problems and eventually
failed due to faster cheaper clones. CBM tried to
build its own PC line but the people in charge were
not computer users nor knew what they were about.
Finally a debtor took over the company and used it
as a Cash Cow and milked it dry and right into bankruptcy.
The Amiga IP was up for grabs and the right folks
did not get it.

>
>   The native 550  graphics were superior,
>   but I badly needed full IBM-CGA compatibility and
>   you had to buy a separate board and do some tricky
>   jumper-wire work to get that. Still a great unit,
>   and about a third the price of IBM. I might still
>   have it somewhere, under the pile-o-junk.
>
>   Anyway, the IBM clones now dominate. A few years
>   back DeGaulle airport, Paris, was paralyzed because
>   it's system for dealing with taxi-way routing went
>   down. Turned out it was running on a PC clone and
>   WINDOWS 3.11 for DECADES. Hey, if it ain't broke ...

If it ain't broke this week wait for the updates
to take it down. An important computer with traffic
control should have been better maintained.

Not clones anymore, x86 computers dominate and
what do you run Linux on anyway. Some builders make
them reliable machines.

>
>   Always wanted a SAGE computer. Looked like a PC box
>   but it had 68000 series chips and some kind of
>   XENIX-related and similar operating systems. Alas,
>   a small maker, expensive, low-volume, pretty quickly
>   went under. Too bad. Those WERE meant as "small
>   business" boxes. Think you can still buy them
>   on E-Bay, but early BAD experiences with E-Bay
>   and Musks PayPal put me off of them. However I
>   have seen working MicroVAX systems for sale there,
>   they were VERY good systems and the OS was well
>   ahead of its time, meant for medium-scale orgs
>   and businesses. Still have the VMS manual, four
>   inches thick, thin paper, smallish type. ONE day ...

The Sage was pretty and even more expensive than the
Amiga. I looked and looked but never got one.

bliss

Charlie Gibbs

unread,
Nov 2, 2021, 2:17:15 AM11/2/21
to
On 2021-11-02, 1p166 <z24ba6.net> wrote:

> But the IBM-PC murdered them all. Wasn't THAT great
> of a PC, but it had the weight of IBM behind it.

:-(

> Apple managed to carve out its own niche, but the
> others went under eventually ... though Commodore
> made a fair try with the Amigas.

Commodore techs made a fair try. Commodore management
ran the company into the ground. At one point the
president and CEO (Irving Gould and Mehdi Ali) were
pulling down bigger salaries than the heads of IBM -
while the techies and marketing staff starved.
Shareholder meetings were held in the Bahamas to
discourage those pesky shareholders from attending.

> Those STILL have
> fans and some Linux utilities STILL support the
> Amiga disk formats plus there are other Amiga
> support programs too. I bought the original, but
> there were SO many "Guru Meditation" messages
> that I dumped the thing and bought a Sanyo-550
> PC (semi)-Clone.

The Amiga's biggest shortcoming was its lack of memory
protection. If you stayed away from buggy software
that stomped on random memory locations, you could
avoid almost all Guru Meditations.

--
/~\ Charlie Gibbs | Life is perverse.
\ / <cgi...@kltpzyxm.invalid> | It can be beautiful -
X I'm really at ac.dekanfrus | but it won't.
/ \ if you read it the right way. | -- Lily Tomlin

The Natural Philosopher

unread,
Nov 2, 2021, 4:31:42 AM11/2/21
to
On 01/11/2021 18:50, Charlie Gibbs wrote:
> On 2021-10-30, The Natural Philosopher <t...@invalid.invalid> wrote:
>
>> On 29/10/2021 17:52, Charlie Gibbs wrote:
>>
>>> On 2021-10-29, The Natural Philosopher <t...@invalid.invalid> wrote:
>>>
>>>> At some point in childhood intelligence organises sensory data into a
>>>> model, that includes a self, in a real (physical) world. That model is
>>>> reinforced through parents etc until people like you think that they
>>>> have actually emerged into 'the real world' and start to explain their
>>>> awareness of it, on terms of its physical nature!
>>>
>>> Unfortunately, many people's models are seriously warped.
>>
>> Warped with respect to what?
>>
>> A couple of decades inquiry into the matter shows that there is no way
>> to establish what the One True Model of Reality looks like. And those
>> that think they are in possession of it are no less deluded than anyone
>> else.
>>
>> My conclusion is that *any* model will do, provided it is not so
>> dysfunctional as to result in death before procreation.
>
> <snip>
>
>> The only concession to 'warpage' I will allow, is that if people are
>> living miserably because they cling to beliefs that make them angry,
>> miserable, jealous and full of shame, well perhaps they should consider
>> abandoning the belief set and replacing it with something else, just as
>> lacking in truth content, but more palatable.
>
> In some societies, this can result in death before procreation. 1/2 :-)
>
Hence my proviso in the first paragraph you quoted.

Humans only duty, and that is only in terms of species survival, is
procreation.
After that its a free lunch.

Sexism, classism. racism, gender politics, systems of government,
religious beliefs? All irrelevant.


--
In theory, there is no difference between theory and practice.
In practice, there is.
-- Yogi Berra

The Natural Philosopher

unread,
Nov 2, 2021, 4:34:53 AM11/2/21
to
On 02/11/2021 05:04, 1p166 wrote:
> People are SMART everywhere. It's the ability to make good USE
>   of that that tends to be the problem.

I disagree, people are stupid everywhere, because inserting a penis in a
vagina doesn't take any great intelligence. Nor does groaning and
pushing the baby out or slapping it against a nipple.

Even rats can do all that.

Society needs one or two 'bright' individuals on occasion when
circumstances need a different approach.
But in general smart people are disliked and reviled and ignored by the
masses .


--
In a Time of Universal Deceit, Telling the Truth Is a Revolutionary Act.

- George Orwell

Charlie Gibbs

unread,
Nov 2, 2021, 3:09:19 PM11/2/21
to
On 2021-11-02, The Natural Philosopher <t...@invalid.invalid> wrote:

> On 02/11/2021 05:04, 1p166 wrote:
>
>> People are SMART everywhere. It's the ability to make good USE
>> of that that tends to be the problem.
>
> I disagree, people are stupid everywhere, because inserting a penis
> in a vagina doesn't take any great intelligence. Nor does groaning
> and pushing the baby out or slapping it against a nipple.
>
> Even rats can do all that.
>
> Society needs one or two 'bright' individuals on occasion when
> circumstances need a different approach.
> But in general smart people are disliked and reviled and ignored
> by the masses.

Throughout history, poverty is the normal condition of man.
Advances which permit this norm to be exceeded - here and there,
now and then - are the work of an extremely small minority,
frequently despised, often condemned, and almost always opposed
by all right-thinking people. Whenever this tiny minority is
kept from creating, or (as sometimes happens) is driven out
of a society, the people then slip back into abject poverty.
This is known as “bad luck.”
-- Robert A. Heinlein: The Notebooks of Lazarus Long

1p166

unread,
Nov 2, 2021, 11:58:12 PM11/2/21
to
On 11/2/21 3:09 PM, Charlie Gibbs wrote:
> On 2021-11-02, The Natural Philosopher <t...@invalid.invalid> wrote:
>
>> On 02/11/2021 05:04, 1p166 wrote:
>>
>>> People are SMART everywhere. It's the ability to make good USE
>>> of that that tends to be the problem.
>>
>> I disagree, people are stupid everywhere, because inserting a penis
>> in a vagina doesn't take any great intelligence. Nor does groaning
>> and pushing the baby out or slapping it against a nipple.
>>
>> Even rats can do all that.

Aww ... Charlie ..............

Oh, and Lazarus Long is FICTIONAL.

1p166

unread,
Nov 3, 2021, 12:28:07 AM11/3/21
to
up
On 11/2/21 2:17 AM, Charlie Gibbs wrote:
> On 2021-11-02, 1p166 <z24ba6.net> wrote:
>
>> But the IBM-PC murdered them all. Wasn't THAT great
>> of a PC, but it had the weight of IBM behind it.
>
> :-(
>
>> Apple managed to carve out its own niche, but the
>> others went under eventually ... though Commodore
>> made a fair try with the Amigas.
>
> Commodore techs made a fair try. Commodore management
> ran the company into the ground.

Rats. Sinking ship. What did you expect ? Pretend
make an visible effort, to keep the stock prices up
while cashing-in to the max.

Commodore was OVER. It had its day in the sun but
IBM (& clones) and Apple were IT - the future.
Whatever Amiga could do, Mac could do, or soon
do, better - and had a bigger customer base.

The Market at the time was consolidating. Only
two main players. All the fringe players were
OUT. Take the money and RUN.

Sorry, but Amiga was NOT a competitor. It had
its good features, but others, better capitalized,
soon copied and exceeded them.

That's the way it goes.

Whatever the Next Big Thing is, there will initially
be a bunch of players. Again, probably TWO will become
IT and all the others will wither away.

And even that duality will be something of a lie ...
the BIG people will have stock/influence in BOTH
"sides". Humans LOVE "duality", choices or false
choices. PLAY that psychology for profit.

Cynical ? Check it out. REAL.

MS is heavily invested in Apple and vice-versa.
Check it out, you can confirm that. The "sides"
are all for show, a way of goading consumers
and pushing out competitors. Swear your loyalty
to Winders or Mac ! So EXCITING to choose a side.

The Big Money people had this figured out LONG
ago - centuries ago actually. Even Machivelli
understood the utility of cultivating those
fake "sides".

Bobbie Sellers

unread,
Nov 3, 2021, 12:39:25 AM11/3/21
to