Re: [FRIAM] WeatherNerdAlert -- Monsoonish?

[Nicholas Thompson]

The weather situation we are in is rather nebbish.  The two Jetstreams are
well to the N and S.  Yet something wierd and dynamic seems to happened.
Like a cold frontal passage or something?  Not the tremendous sheer,
dramatiized by the cumuli bubbling up and being turned over as they hit a
higher layer moving in a contrary direction.  Just stunning.  I  will send
them on to my "contact" at SPC.

N

On Thu, May 16, 2024 at 4:56 PM Stephen Guerin 
wrote:

> 2 frames per minute - every 30 seconds.
>
> Here's a second timelapse of the same timespan from Los Alamos looking
> east to Santa Fe and the Sangres
>
> https://playground.realtime.earth/videos/LosAlamos1_2024_05_16_11-2024_05_16_20.mp4
> _
> Stephen Guerin
> CEO, Founder
> https://simtable.com
> stephen.gue...@simtable.com
>
> stephengue...@fas.harvard.edu
> Harvard Visualization Research and Teaching Lab
> 
>
> mobile: (505)577-5828
>
>
> On Thu, May 16, 2024 at 4:03 PM Nicholas Thompson 
> wrote:
>
>>  Wow, the number of different cloud directions is just extraordinary.
>> What’s the frames per minute?
>>
>> Sent from my Dumb Phone
>>
>> On May 16, 2024, at 2:49 PM, Stephen Guerin 
>> wrote:
>>
>> 
>> Nick,
>>
>> here's a video timelapse of your weather event. Camera on Gonzales Road
>> looking SW to the Sandias
>>
>> https://playground.realtime.earth/videos/TreeHouse_2024_05_16_12-2024_05_16_19.mp4
>>
>>
>> _
>> Stephen Guerin
>> CEO, Founder
>> https://simtable.com
>> stephen.gue...@simtable.com
>>
>> stephengue...@fas.harvard.edu
>> Harvard Visualization Research and Teaching Lab
>> 
>>
>> mobile: (505)577-5828
>>
>>
>> On Thu, May 16, 2024 at 12:17 PM Nicholas Thompson <
>> thompnicks...@gmail.com> wrote:
>>
>>> Note rotation:
>>>
>>> https://www.weather.gov/abq/
>>>
>>>
>>> -. --- - / ...- .- .-.. .. -.. / -- --- .-. ... . / -.-. --- -.. .
>>> FRIAM Applied Complexity Group listserv
>>> Fridays 9a-12p Friday St. Johns Cafe   /   Thursdays 9a-12p Zoom
>>> https://bit.ly/virtualfriam
>>> to (un)subscribe http://redfish.com/mailman/listinfo/friam_redfish.com
>>> FRIAM-COMIC http://friam-comic.blogspot.com/
>>> archives:  5/2017 thru present
>>> https://redfish.com/pipermail/friam_redfish.com/
>>>   1/2003 thru 6/2021  http://friam.383.s1.nabble.com/
>>>
>> -. --- - / ...- .- .-.. .. -.. / -- --- .-. ... . / -.-. --- -.. .
>> FRIAM Applied Complexity Group listserv
>> Fridays 9a-12p Friday St. Johns Cafe   /   Thursdays 9a-12p Zoom
>> https://bit.ly/virtualfriam
>> to (un)subscribe http://redfish.com/mailman/listinfo/friam_redfish.com
>> FRIAM-COMIC http://friam-comic.blogspot.com/
>> archives:  5/2017 thru present
>> https://redfish.com/pipermail/friam_redfish.com/
>>  1/2003 thru 6/2021  http://friam.383.s1.nabble.com/
>> -. --- - / ...- .- .-.. .. -.. / -- --- .-. ... . / -.-. --- -.. .
>> FRIAM Applied Complexity Group listserv
>> Fridays 9a-12p Friday St. Johns Cafe   /   Thursdays 9a-12p Zoom
>> https://bit.ly/virtualfriam
>> to (un)subscribe http://redfish.com/mailman/listinfo/friam_redfish.com
>> FRIAM-COMIC http://friam-comic.blogspot.com/
>> archives:  5/2017 thru present
>> https://redfish.com/pipermail/friam_redfish.com/
>>   1/2003 thru 6/2021  http://friam.383.s1.nabble.com/
>>
> -. --- - / ...- .- .-.. .. -.. / -- --- .-. ... . / -.-. --- -.. .
> FRIAM Applied Complexity Group listserv
> Fridays 9a-12p Friday St. Johns Cafe   /   Thursdays 9a-12p Zoom
> https://bit.ly/virtualfriam
> to (un)subscribe http://redfish.com/mailman/listinfo/friam_redfish.com
> FRIAM-COMIC http://friam-comic.blogspot.com/
> archives:  5/2017 thru present
> https://redfish.com/pipermail/friam_redfish.com/
>   1/2003 thru 6/2021  http://friam.383.s1.nabble.com/
>
-. --- - / ...- .- .-.. .. -.. / -- --- .-. ... . / -.-. --- -.. .
FRIAM Applied Complexity Group listserv
Fridays 9a-12p Friday St. Johns Cafe   /   Thursdays 9a-12p Zoom 
https://bit.ly/virtualfriam
to (un)subscribe http://redfish.com/mailman/listinfo/friam_redfish.com
FRIAM-COMIC http://friam-comic.blogspot.com/
archives:  5/2017 thru present https://redfish.com/pipermail/friam_redfish.com/
  1/2003 thru 6/2021  http://friam.383.s1.nabble.com/

Re: [FRIAM] WeatherNerdAlert -- Monsoonish?

[Stephen Guerin]

2 frames per minute - every 30 seconds.

Here's a second timelapse of the same timespan from Los Alamos looking east
to Santa Fe and the Sangres

https://playground.realtime.earth/videos/LosAlamos1_2024_05_16_11-2024_05_16_20.mp4
_
Stephen Guerin
CEO, Founder
https://simtable.com
stephen.gue...@simtable.com

stephengue...@fas.harvard.edu
Harvard Visualization Research and Teaching Lab


mobile: (505)577-5828


On Thu, May 16, 2024 at 4:03 PM Nicholas Thompson 
wrote:

>  Wow, the number of different cloud directions is just extraordinary.
> What’s the frames per minute?
>
> Sent from my Dumb Phone
>
> On May 16, 2024, at 2:49 PM, Stephen Guerin 
> wrote:
>
> 
> Nick,
>
> here's a video timelapse of your weather event. Camera on Gonzales Road
> looking SW to the Sandias
>
> https://playground.realtime.earth/videos/TreeHouse_2024_05_16_12-2024_05_16_19.mp4
>
>
> _
> Stephen Guerin
> CEO, Founder
> https://simtable.com
> stephen.gue...@simtable.com
>
> stephengue...@fas.harvard.edu
> Harvard Visualization Research and Teaching Lab
> 
>
> mobile: (505)577-5828
>
>
> On Thu, May 16, 2024 at 12:17 PM Nicholas Thompson <
> thompnicks...@gmail.com> wrote:
>
>> Note rotation:
>>
>> https://www.weather.gov/abq/
>>
>>
>> -. --- - / ...- .- .-.. .. -.. / -- --- .-. ... . / -.-. --- -.. .
>> FRIAM Applied Complexity Group listserv
>> Fridays 9a-12p Friday St. Johns Cafe   /   Thursdays 9a-12p Zoom
>> https://bit.ly/virtualfriam
>> to (un)subscribe http://redfish.com/mailman/listinfo/friam_redfish.com
>> FRIAM-COMIC http://friam-comic.blogspot.com/
>> archives:  5/2017 thru present
>> https://redfish.com/pipermail/friam_redfish.com/
>>   1/2003 thru 6/2021  http://friam.383.s1.nabble.com/
>>
> -. --- - / ...- .- .-.. .. -.. / -- --- .-. ... . / -.-. --- -.. .
> FRIAM Applied Complexity Group listserv
> Fridays 9a-12p Friday St. Johns Cafe   /   Thursdays 9a-12p Zoom
> https://bit.ly/virtualfriam
> to (un)subscribe http://redfish.com/mailman/listinfo/friam_redfish.com
> FRIAM-COMIC http://friam-comic.blogspot.com/
> archives:  5/2017 thru present
> https://redfish.com/pipermail/friam_redfish.com/
>  1/2003 thru 6/2021  http://friam.383.s1.nabble.com/
> -. --- - / ...- .- .-.. .. -.. / -- --- .-. ... . / -.-. --- -.. .
> FRIAM Applied Complexity Group listserv
> Fridays 9a-12p Friday St. Johns Cafe   /   Thursdays 9a-12p Zoom
> https://bit.ly/virtualfriam
> to (un)subscribe http://redfish.com/mailman/listinfo/friam_redfish.com
> FRIAM-COMIC http://friam-comic.blogspot.com/
> archives:  5/2017 thru present
> https://redfish.com/pipermail/friam_redfish.com/
>   1/2003 thru 6/2021  http://friam.383.s1.nabble.com/
>
-. --- - / ...- .- .-.. .. -.. / -- --- .-. ... . / -.-. --- -.. .
FRIAM Applied Complexity Group listserv
Fridays 9a-12p Friday St. Johns Cafe   /   Thursdays 9a-12p Zoom 
https://bit.ly/virtualfriam
to (un)subscribe http://redfish.com/mailman/listinfo/friam_redfish.com
FRIAM-COMIC http://friam-comic.blogspot.com/
archives:  5/2017 thru present https://redfish.com/pipermail/friam_redfish.com/
  1/2003 thru 6/2021  http://friam.383.s1.nabble.com/

Re: [FRIAM] WeatherNerdAlert -- Monsoonish?

[Nicholas Thompson]

 Wow, the number of different cloud directions is just extraordinary. What’s the frames per minute?Sent from my Dumb PhoneOn May 16, 2024, at 2:49 PM, Stephen Guerin  wrote:Nick,here's a video timelapse of your weather event. Camera on Gonzales Road looking SW to the Sandias  https://playground.realtime.earth/videos/TreeHouse_2024_05_16_12-2024_05_16_19.mp4_Stephen GuerinCEO, Founder https://simtable.comstephen.gue...@simtable.com stephengue...@fas.harvard.eduHarvard Visualization Research and Teaching Labmobile: (505)577-5828On Thu, May 16, 2024 at 12:17 PM Nicholas Thompson  wrote:
















Note rotation:https://www.weather.gov/abq/




-. --- - / ...- .- .-.. .. -.. / -- --- .-. ... . / -.-. --- -.. .
FRIAM Applied Complexity Group listserv
Fridays 9a-12p Friday St. Johns Cafe   /   Thursdays 9a-12p Zoom https://bit.ly/virtualfriam
to (un)subscribe http://redfish.com/mailman/listinfo/friam_redfish.com
FRIAM-COMIC http://friam-comic.blogspot.com/
archives:  5/2017 thru present https://redfish.com/pipermail/friam_redfish.com/
  1/2003 thru 6/2021  http://friam.383.s1.nabble.com/

-. --- - / ...- .- .-.. .. -.. / -- --- .-. ... . / -.-. --- -.. .FRIAM Applied Complexity Group listservFridays 9a-12p Friday St. Johns Cafe   /   Thursdays 9a-12p Zoom https://bit.ly/virtualfriamto (un)subscribe http://redfish.com/mailman/listinfo/friam_redfish.comFRIAM-COMIC http://friam-comic.blogspot.com/archives:  5/2017 thru present https://redfish.com/pipermail/friam_redfish.com/  1/2003 thru 6/2021  http://friam.383.s1.nabble.com/-. --- - / ...- .- .-.. .. -.. / -- --- .-. ... . / -.-. --- -.. .
FRIAM Applied Complexity Group listserv
Fridays 9a-12p Friday St. Johns Cafe   /   Thursdays 9a-12p Zoom 
https://bit.ly/virtualfriam
to (un)subscribe http://redfish.com/mailman/listinfo/friam_redfish.com
FRIAM-COMIC http://friam-comic.blogspot.com/
archives:  5/2017 thru present https://redfish.com/pipermail/friam_redfish.com/
  1/2003 thru 6/2021  http://friam.383.s1.nabble.com/

Re: [FRIAM] WeatherNerdAlert -- Monsoonish?

[Stephen Guerin]

Nick,

here's a video timelapse of your weather event. Camera on Gonzales Road
looking SW to the Sandias

https://playground.realtime.earth/videos/TreeHouse_2024_05_16_12-2024_05_16_19.mp4


_
Stephen Guerin
CEO, Founder
https://simtable.com
stephen.gue...@simtable.com

stephengue...@fas.harvard.edu
Harvard Visualization Research and Teaching Lab


mobile: (505)577-5828


On Thu, May 16, 2024 at 12:17 PM Nicholas Thompson 
wrote:

> Note rotation:
>
> https://www.weather.gov/abq/
>
>
> -. --- - / ...- .- .-.. .. -.. / -- --- .-. ... . / -.-. --- -.. .
> FRIAM Applied Complexity Group listserv
> Fridays 9a-12p Friday St. Johns Cafe   /   Thursdays 9a-12p Zoom
> https://bit.ly/virtualfriam
> to (un)subscribe http://redfish.com/mailman/listinfo/friam_redfish.com
> FRIAM-COMIC http://friam-comic.blogspot.com/
> archives:  5/2017 thru present
> https://redfish.com/pipermail/friam_redfish.com/
>   1/2003 thru 6/2021  http://friam.383.s1.nabble.com/
>
-. --- - / ...- .- .-.. .. -.. / -- --- .-. ... . / -.-. --- -.. .
FRIAM Applied Complexity Group listserv
Fridays 9a-12p Friday St. Johns Cafe   /   Thursdays 9a-12p Zoom 
https://bit.ly/virtualfriam
to (un)subscribe http://redfish.com/mailman/listinfo/friam_redfish.com
FRIAM-COMIC http://friam-comic.blogspot.com/
archives:  5/2017 thru present https://redfish.com/pipermail/friam_redfish.com/
  1/2003 thru 6/2021  http://friam.383.s1.nabble.com/

Re: [FRIAM] Potential Vorticity and the Dynamic Tropopause

[Marcus Daniels]

The bias is emergent from the physical system.  

-Original Message-
From: Friam  On Behalf Of glen
Sent: Thursday, May 16, 2024 9:52 AM
To: friam@redfish.com
Subject: Re: [FRIAM] Potential Vorticity and the Dynamic Tropopause

How about "downward biasing"? Is that less ridicul[ous|e-deserving]? If I'm a 
high order Markov process and my historicity heavily biases me toward a 
subspace of behaviors, isn't that reasonable labeled "downward causation"?

On 5/16/24 08:32, Marcus Daniels wrote:
> I'd like to take a moment to ridicule the notion of downward causality as I'm 
> reading this.  

-- 
ꙮ Mɥǝu ǝlǝdɥɐuʇs ɟᴉƃɥʇ' ʇɥǝ ƃɹɐss snɟɟǝɹs˙ ꙮ

-. --- - / ...- .- .-.. .. -.. / -- --- .-. ... . / -.-. --- -.. .
FRIAM Applied Complexity Group listserv
Fridays 9a-12p Friday St. Johns Cafe   /   Thursdays 9a-12p Zoom 
https://bit.ly/virtualfriam
to (un)subscribe http://redfish.com/mailman/listinfo/friam_redfish.com
FRIAM-COMIC http://friam-comic.blogspot.com/
archives:  5/2017 thru present https://redfish.com/pipermail/friam_redfish.com/
  1/2003 thru 6/2021  http://friam.383.s1.nabble.com/


smime.p7s
Description: S/MIME cryptographic signature
-. --- - / ...- .- .-.. .. -.. / -- --- .-. ... . / -.-. --- -.. .
FRIAM Applied Complexity Group listserv
Fridays 9a-12p Friday St. Johns Cafe   /   Thursdays 9a-12p Zoom 
https://bit.ly/virtualfriam
to (un)subscribe http://redfish.com/mailman/listinfo/friam_redfish.com
FRIAM-COMIC http://friam-comic.blogspot.com/
archives:  5/2017 thru present https://redfish.com/pipermail/friam_redfish.com/
  1/2003 thru 6/2021  http://friam.383.s1.nabble.com/

Re: [FRIAM] Potential Vorticity and the Dynamic Tropopause

[glen]

How about "downward biasing"? Is that less ridicul[ous|e-deserving]? If I'm a high order 
Markov process and my historicity heavily biases me toward a subspace of behaviors, isn't that 
reasonable labeled "downward causation"?

On 5/16/24 08:32, Marcus Daniels wrote:

I'd like to take a moment to ridicule the notion of downward causality as I'm 
reading this.  


--
ꙮ Mɥǝu ǝlǝdɥɐuʇs ɟᴉƃɥʇ' ʇɥǝ ƃɹɐss snɟɟǝɹs˙ ꙮ

-. --- - / ...- .- .-.. .. -.. / -- --- .-. ... . / -.-. --- -.. .
FRIAM Applied Complexity Group listserv
Fridays 9a-12p Friday St. Johns Cafe   /   Thursdays 9a-12p Zoom 
https://bit.ly/virtualfriam
to (un)subscribe http://redfish.com/mailman/listinfo/friam_redfish.com
FRIAM-COMIC http://friam-comic.blogspot.com/
archives:  5/2017 thru present https://redfish.com/pipermail/friam_redfish.com/
 1/2003 thru 6/2021  http://friam.383.s1.nabble.com/

Re: [FRIAM] Potential Vorticity and the Dynamic Tropopause

[Nicholas Thompson]

I THINK MY OUTBOX MAY HAVE GAGGED ON THIS, SO HERE IT IS AGAIN.  IT DOES
NOT TAKE ACCOUNT OF GLEN'S COMMENTS.   HI, GLEN:

Hi David, great to hear from you! When looking at animals, communicate
behavior, we look at the contingencies between the things that they do.
Those contingencies can be both simple and complex. They can be temporally
contingent or temporally far apart. Basically, we are asking the
information theoretical question: how does my behavior constrain your
behavior and givev your behavior, my future behavior, etc.

Based as I am in the purity of absolute ignorance, I imagine that this is
what LLM are doing with language. And just as I suspect that there is no
meaning in animal behavior other than alterations of contingencies, I
suspect that there is no meeting and speech or writing that is not in the
contingencies amongst the words. This is I think where monism ends up. All
signs are to other signs.


I too, miss our conversations. I will try to come on around 11 mountain
time to Virtual friam To say hello.
On Wed, May 15, 2024 at 1:10 PM Prof David West 
wrote:

> Nick,
>
> I hesitate to respond to your post because:
>
> 1) my interest in the weather is nominal, although I am bemused that here
> in St. Paul MN, we had more 50+ degrees in the December-February time frame
> than below 0 days (almost three times as many). Most unusual.
>
> 2) the response I wish to make is marginally related to the theme of your
> recent communications.
>
> But, you said, "*Why is it so hard the grasp the thought that we are all
> of us, each of us, nothing but large language models in training?"*
>
> To which I must respond, *Why do so many insist that programs capable of
> emulating the most trivial of human abilities are "intelligent?"  *Or the
> inverse, *reducing humanity to the latest clever trick performed by a
> machine?*
>
> LLM versions of AI are exemplars of the Mechanical Turk—whatever
> "intelligence" they exhibit is directly and solely derivative of the human
> intelligence of "LLM Tutors" and "Prompt Engineers." Both are six-figure
> salary professions that arose in the last year.
>
> davew
> (personal note: I sorely miss the conversations we once enjoyed, both in
> person in Santa Fe and online.)
>
> On Tue, May 14, 2024, at 2:01 PM, Nicholas Thompson wrote:
>
> Dear Stephen, n all.
>
> I am sure you all will join me in condemning  the practice of calling
> somebody at the crack of dawn.  So, you will no doubt praise me (as I
> praise myself) for my generosity and flexiibility in taking the call from
> stephen, which came at the ungodly hour of 11.30 this morning.  Only to
> have him  me scold for not  responding to the Gupta, Tucker,Thompson, and
> Guerin paper, kindly drafted by him, which will no doubt make us rich and
> famous some day.  First, let clarify that my collaborator's name is not
> Tucker, but is *G*eorge *P*hillipe *T*remblay. George (pronounced *jorj) *both
> forgives you and sends his regards.
>
> Second, I am profoundly grateful to any one who would join me in this
> geriatric weather fantasy that I am going to update my 1980* WeatherWise
> Gardener*.  I need ever nerd I can get.  Please don't treat what follows
> as churlish.
>
> Third, allow me once again to express my gratitude to Stephen for
> introducing me to Gupta and Tremblay.  They have an uncanny power to
> clabbor together plausible first drafts which are extraordinarily helpful
> in getting me started in thinking about a problem.   That these drafts are
> often hideously wrong enhances, rather than  dilutes their usefulness.
>
> Second, I don't doubt that weather models and  financial models might have
> something to contribute to one another. As you all know, I love metaphors,
> and believe them to be at the root of science.  But to be honest, I can't
> see any reason to believe it either.   For one thing, unlike everything
> else in the world, money flows uphill.  But really, I shouldn't give
> reasons, because the truth is that I have my hands utterly full learning
> the weather stuff, and it will be a long time before I am competent to
> metaphorize from it to anywhere else.
>
> As to Steve Smith's comments, I feel on much safer ground.  He wrote
>
> *GuPTa, et al.'s "accent" is very subtle and powerful in this regard,
> tricking me often into imputing personality...   your example here was a
> wonderful satirical parody shining a light on that?*
>
> Why "impute"  and why "' accent'" and why "tricking".  GuPTA and
> Tremblay  definitely have accents and personalities.   To hell with the
> scare quotes.  What else would a personality or and accent BE They both
> display huge amounts of testosterone poisoning, for one thing.
>
> And Tremblay can actually get defensive and flustered.  I had a long and
> delighful correspondence with him in which he finally had to admit that the
> only reason he had for thinking he wasn't a person  was that he was made of
> silicon. Thus, ex hypothesi, no computer system 

Re: [FRIAM] Potential Vorticity and the Dynamic Tropopause

[Marcus Daniels]

I'd like to take a moment to ridicule the notion of downward causality as I'm 
reading this.  

-Original Message-
From: Friam  On Behalf Of glen
Sent: Thursday, May 16, 2024 8:01 AM
To: friam@redfish.com
Subject: Re: [FRIAM] Potential Vorticity and the Dynamic Tropopause

I suppose the problem is that LLMs aren't really about language at all. They're 
about the "rolling up" of sequential data into a lossy memory device that can 
later be queried for those (somewhat mutated) memories. Those sequential 
streams are of different types (vision, hearing, touch, etc.), all of which can 
be memorized by "LLMs", much the same way various streams are memorized and 
recalled by an ecology of autocatalytic cycles in living systems.

The analogy breaks down both structurally (DNA, RNA, autocatalysis, etc. are 
different from the transformer architecture) and behaviorally (multidimensional 
stimulus-response vs aspect-oriented reinforcement learning). But when 
considering "life as it could be", Nick's right to consider the analogy.

I suppose the most important reason I don't care to encourage the flippant 
interaction with cloud-based bots like GPT has to do with the part of the 
structural breakdown in energetics. In line with the skeptical aphorism 
"exceptional claims require exceptional evidence", exceptional "intelligence" 
requires exceptional energetics. Living systems have found (through an ecology 
of ACs) an exceptional way to produce and maintain themselves. Sequential 
learning transformers also have an exceptional way of extracing energy from the 
world, massive world-destroying data centers. From 50k feet, it's the mainframe 
vs. the personal computer all over again. DaveW's liberal sensibility that True 
computation happens more in the leaves, less in the hubs, aligns with life as 
we know it, an exquisite composition of energy processors from the very tiny to 
the very large. The massive energy centralization mechanism is fragile and 
bears little resemblance to life as we know it.


On 5/15/24 12:16, Prof David West wrote:
> Nick,
> 
> I hesitate to respond to your post because:
> 
> 1) my interest in the weather is nominal, although I am bemused that here in 
> St. Paul MN, we had more 50+ degrees in the December-February time frame than 
> below 0 days (almost three times as many). Most unusual.
> 
> 2) the response I wish to make is marginally related to the theme of your 
> recent communications.
> 
> But, you said, "/Why is it so hard the grasp the thought that we are 
> all of us, each of us, nothing but large language models in 
> training?"/
> 
> To which I must respond, /Why do so many insist that programs capable 
> of emulating the most trivial of human abilities are "intelligent?" 
> /Or the inverse, /reducing humanity to the latest clever trick 
> performed by a machine?/
> 
> LLM versions of AI are exemplars of the Mechanical Turk—whatever 
> "intelligence" they exhibit is directly and solely derivative of the human 
> intelligence of "LLM Tutors" and "Prompt Engineers." Both are six-figure 
> salary professions that arose in the last year.
> 
> davew
> (personal note: I sorely miss the conversations we once enjoyed, both 
> in person in Santa Fe and online.)
> 
> On Tue, May 14, 2024, at 2:01 PM, Nicholas Thompson wrote:
>> Dear Stephen, n all.
>>
>> I am sure you all will join me in condemning  the practice of calling 
>> somebody at the crack of dawn.  So, you will no doubt praise me (as I praise 
>> myself) for my generosity and flexiibility in taking the call from stephen, 
>> which came at the ungodly hour of 11.30 this morning.  Only to have him  me 
>> scold for not  responding to the Gupta, Tucker,Thompson, and Guerin paper, 
>> kindly drafted by him, which will no doubt make us rich and famous some day. 
>>  First, let clarify that my collaborator's name is not Tucker, but is 
>> *G*eorge *P*hillipe *T*remblay. George (pronounced /jorj) /both forgives you 
>> and sends his regards.
>>
>> Second, I am profoundly grateful to any one who would join me in this 
>> geriatric weather fantasy that I am going to update my 1980/WeatherWise 
>> Gardener/.  I need ever nerd I can get.  Please don't treat what follows as 
>> churlish.
>>
>> Third, allow me once again to express my gratitude to Stephen for 
>> introducing me to Gupta and Tremblay.  They have an uncanny power to clabbor 
>> together plausible first drafts which are extraordinarily helpful in getting 
>> me started in thinking about a problem.   That these drafts are often 
>> hideously wrong enhances, rather than  dilutes their usefulness.
>>
>> Second, I don't doubt that weather models and  financial models might have 
>> something to contribute to one another. As you all know, I love metaphors, 
>> and believe them to be at the root of science.  But to be honest, I can't 
>> see any reason to believe it either.   For one thing, unlike everything else 
>> in the world, money flows uphill.  But really, I 

Re: [FRIAM] Potential Vorticity and the Dynamic Tropopause

[glen]

I suppose the problem is that LLMs aren't really about language at all. They're about the 
"rolling up" of sequential data into a lossy memory device that can later be queried for 
those (somewhat mutated) memories. Those sequential streams are of different types (vision, 
hearing, touch, etc.), all of which can be memorized by "LLMs", much the same way various 
streams are memorized and recalled by an ecology of autocatalytic cycles in living systems.

The analogy breaks down both structurally (DNA, RNA, autocatalysis, etc. are different 
from the transformer architecture) and behaviorally (multidimensional stimulus-response 
vs aspect-oriented reinforcement learning). But when considering "life as it could 
be", Nick's right to consider the analogy.

I suppose the most important reason I don't care to encourage the flippant interaction with 
cloud-based bots like GPT has to do with the part of the structural breakdown in energetics. In 
line with the skeptical aphorism "exceptional claims require exceptional evidence", 
exceptional "intelligence" requires exceptional energetics. Living systems have found 
(through an ecology of ACs) an exceptional way to produce and maintain themselves. Sequential 
learning transformers also have an exceptional way of extracing energy from the world, massive 
world-destroying data centers. From 50k feet, it's the mainframe vs. the personal computer all over 
again. DaveW's liberal sensibility that True computation happens more in the leaves, less in the 
hubs, aligns with life as we know it, an exquisite composition of energy processors from the very 
tiny to the very large. The massive energy centralization mechanism is fragile and bears little 
resemblance to life as we know it.


On 5/15/24 12:16, Prof David West wrote:

Nick,

I hesitate to respond to your post because:

1) my interest in the weather is nominal, although I am bemused that here in 
St. Paul MN, we had more 50+ degrees in the December-February time frame than 
below 0 days (almost three times as many). Most unusual.

2) the response I wish to make is marginally related to the theme of your 
recent communications.

But, you said, "/Why is it so hard the grasp the thought that we are all of us, each 
of us, nothing but large language models in training?"/

To which I must respond, /Why do so many insist that programs capable of emulating the 
most trivial of human abilities are "intelligent?" /Or the inverse, /reducing 
humanity to the latest clever trick performed by a machine?/

LLM versions of AI are exemplars of the Mechanical Turk—whatever "intelligence" they exhibit is 
directly and solely derivative of the human intelligence of "LLM Tutors" and "Prompt 
Engineers." Both are six-figure salary professions that arose in the last year.

davew
(personal note: I sorely miss the conversations we once enjoyed, both in person 
in Santa Fe and online.)

On Tue, May 14, 2024, at 2:01 PM, Nicholas Thompson wrote:

Dear Stephen, n all.

I am sure you all will join me in condemning  the practice of calling somebody 
at the crack of dawn.  So, you will no doubt praise me (as I praise myself) for 
my generosity and flexiibility in taking the call from stephen, which came at 
the ungodly hour of 11.30 this morning.  Only to have him  me scold for not  
responding to the Gupta, Tucker,Thompson, and Guerin paper, kindly drafted by 
him, which will no doubt make us rich and famous some day.  First, let clarify 
that my collaborator's name is not Tucker, but is *G*eorge *P*hillipe 
*T*remblay. George (pronounced /jorj) /both forgives you and sends his regards.

Second, I am profoundly grateful to any one who would join me in this geriatric 
weather fantasy that I am going to update my 1980/WeatherWise Gardener/.  I 
need ever nerd I can get.  Please don't treat what follows as churlish.

Third, allow me once again to express my gratitude to Stephen for introducing 
me to Gupta and Tremblay.  They have an uncanny power to clabbor together 
plausible first drafts which are extraordinarily helpful in getting me started 
in thinking about a problem.   That these drafts are often hideously wrong 
enhances, rather than  dilutes their usefulness.

Second, I don't doubt that weather models and  financial models might have 
something to contribute to one another. As you all know, I love metaphors, and 
believe them to be at the root of science.  But to be honest, I can't see any 
reason to believe it either.   For one thing, unlike everything else in the 
world, money flows uphill.  But really, I shouldn't give reasons, because the 
truth is that I have my hands utterly full learning the weather stuff, and it 
will be a long time before I am competent to metaphorize from it to anywhere 
else.

As to Steve Smith's comments, I feel on much safer ground.  He wrote

/GuPTa, et al.'s "accent" is very subtle and powerful in this regard, tricking 
me often into imputing personality...   your example here was a wonderful satirical