Re: [FRIAM] new thermal tech

[Pieter Steenekamp]

I am somewhat puzzled by the discussion surrounding the efficiency of heat
pumps in relation to the laws of thermodynamics. If my understanding is
correct, the salient question being raised is whether it is possible for a
heat pump to transfer 3 to 4 kW of heat using only 1 kW of electricity,
which seems to contradict the laws of thermodynamics.

Allow me to clarify: a heat pump is not creating 3 to 4 kW of energy, but
rather transferring 3 to 4 kW of heat from a lower temperature to a higher
temperature using 1 kW of electrical energy. This is entirely within the
realm of thermodynamic principles, as I learned them in engineering school.
Heat pumps have been in use for some time, and it is not uncommon to find
off-the-shelf heat pumps capable of this task.

Pieter

On Fri, 6 Jan 2023 at 19:34, David Eric Smith  wrote:

> I assume “delivers” would have been a better word choice than “produces”.
>
> After we create better schools to teach our kids government and
> citizenship, we should move on and teach them to understand thermodynamics.
>
>
> > On Jan 6, 2023, at 12:27 PM, Gary Schiltz 
> wrote:
> >
> > Amusing indeed (the PV magazine article). The pump “reportedly
> > produces 3 kW to 4 kW of heat for every kilowatt of power it
> > consumes”. Say what?
> >
> > On Fri, Jan 6, 2023 at 8:28 AM Roger Critchlow  wrote:
> >>
> >> I was amused to see an announcement of a thermoacoustic heat pump  the
> other day:
> >>
> >>
> https://linkprotect.cudasvc.com/url?a=https%3a%2f%2fwww.pv-magazine.com%2f2023%2f01%2f02%2fresidential-thermo-acoustic-heat-pump-produces-water-up-to-80-c%2f&c=E,1,SIXaeD9pWPwQ9_a2yQcA2Mb6Oa19--caDioiCnI9tb9wHtj9F3T-fS0FbAdlNeoJjozEjvgRHvMxxDB874D65rIXmYYzllXMzQ-xbTUOen3OXZ0wb7n6ON_lX3U,&typo=1
> >>
> >> then an ionocaloric refrigerator announcement turns up this morning
> >>
> >>
> https://linkprotect.cudasvc.com/url?a=https%3a%2f%2fnewscenter.lbl.gov%2f2023%2f01%2f03%2fcool-new-method-of-refrigeration%2f&c=E,1,WtFGOsq-ZfdSeNDj-coqZXFEeJnYAea0pwlFL95XWPuzaTHMiuaDIsqM19iVS7Oaw1ElarAioZmixa4jBVRrZyrnp7sdhboRXZmwL4HYzjyNmyPqbA,,&typo=1
> >>
> >> It seems that you won't recognize your air conditioner in a few years.
> >>
> >> -- rec --
> >> -. --- - / ...- .- .-.. .. -.. / -- --- .-. ... . / -.-. --- -.. .
> >> FRIAM Applied Complexity Group listserv
> >> Fridays 9a-12p Friday St. Johns Cafe   /   Thursdays 9a-12p Zoom
> https://linkprotect.cudasvc.com/url?a=https%3a%2f%2fbit.ly%2fvirtualfriam&c=E,1,l2P4P2hAd9fOXrHFAqYE-soxRvcd-h7oG1MT-pwzYOcNLRhMJ0htXFfoau0plG4M8Sny3C4wi3OuWkOaPuGG7ig9z9x78f1OAPt8r97EV5kw&typo=1
> >> to (un)subscribe
> https://linkprotect.cudasvc.com/url?a=http%3a%2f%2fredfish.com%2fmailman%2flistinfo%2ffriam_redfish.com&c=E,1,Mv9je9W6-toCn2mSgebsBQ2oAvgPYBiznnu7oMzXwVVpVHCX5bHTuwkLHkMpS5xCLZNN1jOoUKDPMKGdl1DklBJnHT6qIi1-4MP3jz-xjyU06iWZj8gEGdY,&typo=1
> >> FRIAM-COMIC
> https://linkprotect.cudasvc.com/url?a=http%3a%2f%2ffriam-comic.blogspot.com%2f&c=E,1,LrVd5NHt3F5ZyNnYASIf5xseu84feQKjpBBwX-nIs29YXhMV8_76gf8cpBx-vaeUpFf-HGs2p2hq0x0mGoPN8kH0VhZGBCx3iClL1gICfwzbTw,,&typo=1
> >> archives:  5/2017 thru present
> https://linkprotect.cudasvc.com/url?a=https%3a%2f%2fredfish.com%2fpipermail%2ffriam_redfish.com%2f&c=E,1,uahCj4-Us9Y3YmIN-yVtgMzfbZ9PQRVMSGkO6IYV0N9KUU9ZFY1kQa3BLPFvEuXo3Aq1MJ2Cz_0O_ne5GEfvWgd-OkD5gkJxAvRbSAfX3t8NMsKUtaHB-cDpzOrn&typo=1
> >>  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://linkprotect.cudasvc.com/url?a=https%3a%2f%2fbit.ly%2fvirtualfriam&c=E,1,ISKO6LD8eQE8bCuHCQdZ0Gz68pToeZRW1lYvrOVc7g_fN_WR7boTcobKscWBk5yR1j-6b4R3aMqqzVOCJl2dqKUVnXNLqSONq-LfHDeMRXYahMWyKp3Af2c,&typo=1
> > to (un)subscribe
> https://linkprotect.cudasvc.com/url?a=http%3a%2f%2fredfish.com%2fmailman%2flistinfo%2ffriam_redfish.com&c=E,1,pC4zTC9TPXexDYqD_66fvJzl2DnWmxO_oNw6VBrLvm-RLF3uuPCVIIKR2OJQCiTV2oiJs8vos0DyebFkCE1wop0EL_KPEX9Ubs3iXHiL7U0CbzBDf23s&typo=1
> > FRIAM-COMIC
> https://linkprotect.cudasvc.com/url?a=http%3a%2f%2ffriam-comic.blogspot.com%2f&c=E,1,ZqqnEKAOuo-2XtVUgR8oNnQMs8sMJxA0yhR0UWNXecnKmVcuThV2HDMewZbrxblvfqZnqX-aDXbUG3c-hOx2euWFTs6SNgbC31UlEaVYkxcPI-I3NRM,&typo=1
> > archives:  5/2017 thru present
> https://linkprotect.cudasvc.com/url?a=https%3a%2f%2fredfish.com%2fpipermail%2ffriam_redfish.com%2f&c=E,1,YibNlqRVSsKKWCkGgCh8hBTgGfw3eLYOMmtcJqSQH0k5GE8ekETSArYhdys_xonOo_aLWqzXbvccM9FgCHeUXvK_w9dvfSzQRuPukdWmSf30uA,,&typo=1
> >  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_redf

Re: [FRIAM] more bullsh¡t

[Pieter Steenekamp]

In my current perspective, the integration of automation and artificial
intelligence *in the coming decade* will significantly impact employment
opportunities. However, I also believe that in this future, there will be
an abundance of material resources, allowing individuals to access the
products and services they desire within the limitations of physical laws.
It is impossible to predict the future with certainty, but these are my
current speculations based on my observations.

In terms of education, I advocate for a system that allows for self-guided
learning rather than mandatory teaching. By creating an environment that
supports learning and inviting guest instructors for specialized knowledge,
students can pursue their interests and passions. Even if some students
attempt to abuse the system, the benefits of fostering a love for learning
outweigh any potential harm.

Full disclosure: I haven't kissed the Blarney Stone and wrote a paragraph,
but it did not sound good. So I asked ChatGPT to rephrase and upon reading
I went yes this is exactly my point and it almost sounds as if I have
kissed the bloody stone. So the above is my exact message but rephrased
eloquently by ChatGPT.

On Fri, 6 Jan 2023 at 12:19, David Eric Smith  wrote:

> Your use of chatGTP, Pieter, is to my mind a very interesting thread.
>
> There was a columnist for the New York Times many years ago, named William
> Safire.  I don’t even remember now what he wrote about, but he was known,
> and was significant to me, for being an example of “a good writer”.
>
> Safire wrote something (column?  book?  article?) with the theme that, if
> one would write creatively, one should first do several years galley-rowing
> as an editor somewhere.  People who have an impulse to write “creatively”
> imagine all kinds of innovation in language that will be just dramatic and
> wonderful.  Editors, who have had to deal with those imaginations in the
> writings of others, know that most such ideas are awful and need to be
> beaten out of the writer if he or she is ever to become good.  So Safire’s
> thesis was that you really need to do this, for a number of years and a
> large number of other people’s pieces, to squeeze the nonsense out of you
> and develop a solid understanding of your language.  Kind of like, in books
> on French cooking, the author says “why did we have to spend the first year
> cooking mixed vegetables in mayonnaise over and over again; I don’t even
> like mixed vegetables in mayonnaise.  To which the answer, of course, is
> that one develops what the French term “method”: experience with uniform
> sizing of each ingredient, correct relative sizing across ingredients, time
> of introduction to heat, and on and on, so that one gets control and has
> everything cooked to the intended degree reliably.  Only then has one
> gained the tools to create.
>
> I have run Safire’s thesis by some writers I know to see what happened; my
> notable memories are the ones who hate it and think it is completely wrong.
>
> But to Glen’s point that we should think of literary AI the way we think
> of pocket calculators (another thing I was not allowed to have in school;
> my parents thought it would make me stupid).  ChatGTP can be sort of the
> William Safire level of basic method in language, not intending or intended
> to create anything, but somehow, as you say, to find a kind of solid and
> central expression for things.  One might even think of the appropriate
> training schedule for a tool meant to do just that, which could be a bit
> different from the ad hoc training that is probably first-gen of these
> tools.
>
> Eric
>
> On Jan 6, 2023, at 12:59 AM, Pieter Steenekamp 
> wrote:
>
> As a native of South Africa, I have personally witnessed the shortcomings
> of both our public primary and secondary education systems and the
> financial barriers that prevent many from accessing private schools. In
> response, I have dedicated the past year to establishing a private
> institution that is not only affordable, but also committed to providing a
> high-quality education. In reflecting on what constitutes a truly valuable
> education, I have come to the conclusion that the most essential outcome is
> not the acquisition of academic skills, but rather the development of
> strong relationships - both with oneself and with the outside world. While
> it is not possible to directly teach children how to cultivate such
> relationships, it is possible to create an environment in which they can
> learn and grow through unsupervised interactions with their peers.
>
> Full disclosure: I have not kissed  Blarney Stone and my ability to write
> (or speak for that matter) eloquently is just awful. I've written a
> paragraph and then I asked chatGPT, who have kissed the Blamey Stone, to
> rephrase it more eloquently. The above paragraph reflects exactly what I
> wish to say, but is just expressed so much better.
>
> On Fri, 6 Jan 2023 at 00:39, glen  wrote:
>
>>

Re: [FRIAM] Sorting Algorithm? AI? Identifying "types" within data

[Pieter Steenekamp]

I asked https://chat.openai.com/chat and here is the conversation:

*Pieter Steenekamp*
can you suggest a solution for the following problem "I'm hoping someone
here could help out. Let's imagine I had some data where each row was a
person's career. We could list major events every year.For example: 2004
they were highered, 2007 they get a promotion, 2010 they leave for a
different company, 2012 they come back at a higher level, 2015 get a
promotion, then no change until 2022.  Let's say I had data like this for
roughly 2 million people, and that there are around 10 different types of
changes that could happen during any time period (could be yearly,
quarterly, monthly, I can make it how I want).  I was hoping we could ask a
computer to tell us if there were "types of careers" that people had. We
could say "put all these careers into 4 buckets" or "7 buckets" based on
similarity. Then we could look at the piles the computer made and try to
make sense of them. One type might be "company man" for people who tend to
stay in place for 20 or more years, another type could be a "rotator", who
leaves and returns every 3 years or so. Etc. The point is, I want a
computer to make the piles for me, rather than trying to come up with
potential piles a priori. Are there methods for doing this? I know it's a
problem we've *talked* about a lot, but I don't know if there are
solutions. "
*https://chat.openai.com/chat *
There are several approaches that you could use to group the careers of
these 2 million people into clusters based on similarity. One approach is
to use unsupervised machine learning techniques, specifically clustering
algorithms, to automatically group the data into clusters. Some popular
clustering algorithms that you could use for this problem include k-means,
hierarchical clustering, and DBSCAN (Density-Based Spatial Clustering of
Applications with Noise). You could also use dimensionality reduction
techniques, such as PCA (Principal Component Analysis) or t-SNE
(t-Distributed Stochastic Neighbor Embedding), to reduce the dimensionality
of the data and then use a clustering algorithm to group the data into
clusters. Another approach is to use supervised machine learning
techniques, where you would need to manually label a subset of the data
with the cluster labels, and then use this labeled data to train a
classifier to predict the cluster labels for the remaining data. It's worth
noting that these approaches will require some feature engineering, where
you will need to extract relevant features from the data that can be used
to represent the careers of the people in the dataset. Once you have
identified and extracted these features, you can then use them as input to
the machine learning algorithms to group the data into clusters.
*Pieter Steenekamp*
would it then be possible to get  for example if there were "types of
careers" that people had?
*https://chat.openai.com/chat *
Yes, once you have grouped the careers of the 2 million people into
clusters using one of the approaches I mentioned above, you can then
analyze the characteristics of each cluster to identify common patterns and
trends within each cluster. This will allow you to identify "types of
careers" that people have, based on the similarity of their career events.
For example, if one of the clusters is characterized by people who have had
long tenures at a single company, you could label this cluster as "company
man" careers. Similarly, if another cluster is characterized by people who
frequently switch between companies, you could label this cluster as
"rotator" careers. You can then use this information to gain insights into
the types of careers that are represented in the dataset, and potentially
draw conclusions about the factors that may influence career paths and
success.
*Pieter Steenekamp*
can you propose python code to do sometingh like this please
*https://chat.openai.com/chat *
Sure, here is an example of how you could use the scikit-learn library in
Python to group the careers of the 2 million people into clusters using the
k-means algorithm:
Copy code
from sklearn.cluster import KMeans
# Load the data into a NumPy array
data = np.loadtxt('career_data.txt')
# Create the k-means model
model = KMeans(n_clusters=4)
# Fit the model to the data
model.fit(data)
# Predict the cluster labels for each data point
labels = model.predict(data)
# Print the cluster labels
print(labels)
This code will group the data into 4 clusters, as specified by the
n_clusters parameter. You can adjust this parameter to change the number of
clusters that the data is grouped into.
To extract features from the data that can be used as input to the k-means
model, you will need to preprocess the data and select relevant features.
This will likely involve some feature engineering, which will depend on the
specific characteristics of the data and the goals of your analysis.

[FRIAM] Sorting Algorithm? AI? Identifying "types" within data

[Eric Charles]

Greetings all,
I'm hoping someone here could help out. Let's imagine I had some data where
each row was a person's career. We could list major events every year.

For example: 2004 they were highered, 2007 they get a promotion, 2010 they
leave for a different company, 2012 they come back at a higher level, 2015
get a promotion, then no change until 2022.

Let's say I had data like this for roughly 2 million people, and that there
are around 10 different types of changes that could happen during any time
period (could be yearly, quarterly, monthly, I can make it how I want).

I was hoping we could ask a computer to tell us if there were "types of
careers" that people had. We could say "put all these careers into 4
buckets" or "7 buckets" based on similarity. Then we could look at the
piles the computer made and try to make sense of them.

One type might be "company man" for people who tend to stay in place for 20
or more years, another type could be a "rotator", who leaves and returns
every 3 years or so. Etc. The point is, I want a computer to make the piles
for me, rather than trying to come up with potential piles a priori.

Are there methods for doing this? I know it's a problem we've *talked*
about a lot, but I don't know if there are solutions.

Any help would be appreciated.

Best,
Eric


-. --- - / ...- .- .-.. .. -.. / -- --- .-. ... . / -.-. --- -.. .
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] new thermal tech

[David Eric Smith]

I assume “delivers” would have been a better word choice than “produces”.  

After we create better schools to teach our kids government and citizenship, we 
should move on and teach them to understand thermodynamics.


> On Jan 6, 2023, at 12:27 PM, Gary Schiltz  wrote:
> 
> Amusing indeed (the PV magazine article). The pump “reportedly
> produces 3 kW to 4 kW of heat for every kilowatt of power it
> consumes”. Say what?
> 
> On Fri, Jan 6, 2023 at 8:28 AM Roger Critchlow  wrote:
>> 
>> I was amused to see an announcement of a thermoacoustic heat pump  the other 
>> day:
>> 
>>  
>> https://linkprotect.cudasvc.com/url?a=https%3a%2f%2fwww.pv-magazine.com%2f2023%2f01%2f02%2fresidential-thermo-acoustic-heat-pump-produces-water-up-to-80-c%2f&c=E,1,SIXaeD9pWPwQ9_a2yQcA2Mb6Oa19--caDioiCnI9tb9wHtj9F3T-fS0FbAdlNeoJjozEjvgRHvMxxDB874D65rIXmYYzllXMzQ-xbTUOen3OXZ0wb7n6ON_lX3U,&typo=1
>> 
>> then an ionocaloric refrigerator announcement turns up this morning
>> 
>>  
>> https://linkprotect.cudasvc.com/url?a=https%3a%2f%2fnewscenter.lbl.gov%2f2023%2f01%2f03%2fcool-new-method-of-refrigeration%2f&c=E,1,WtFGOsq-ZfdSeNDj-coqZXFEeJnYAea0pwlFL95XWPuzaTHMiuaDIsqM19iVS7Oaw1ElarAioZmixa4jBVRrZyrnp7sdhboRXZmwL4HYzjyNmyPqbA,,&typo=1
>> 
>> It seems that you won't recognize your air conditioner in a few years.
>> 
>> -- rec --
>> -. --- - / ...- .- .-.. .. -.. / -- --- .-. ... . / -.-. --- -.. .
>> FRIAM Applied Complexity Group listserv
>> Fridays 9a-12p Friday St. Johns Cafe   /   Thursdays 9a-12p Zoom 
>> https://linkprotect.cudasvc.com/url?a=https%3a%2f%2fbit.ly%2fvirtualfriam&c=E,1,l2P4P2hAd9fOXrHFAqYE-soxRvcd-h7oG1MT-pwzYOcNLRhMJ0htXFfoau0plG4M8Sny3C4wi3OuWkOaPuGG7ig9z9x78f1OAPt8r97EV5kw&typo=1
>> to (un)subscribe 
>> https://linkprotect.cudasvc.com/url?a=http%3a%2f%2fredfish.com%2fmailman%2flistinfo%2ffriam_redfish.com&c=E,1,Mv9je9W6-toCn2mSgebsBQ2oAvgPYBiznnu7oMzXwVVpVHCX5bHTuwkLHkMpS5xCLZNN1jOoUKDPMKGdl1DklBJnHT6qIi1-4MP3jz-xjyU06iWZj8gEGdY,&typo=1
>> FRIAM-COMIC 
>> https://linkprotect.cudasvc.com/url?a=http%3a%2f%2ffriam-comic.blogspot.com%2f&c=E,1,LrVd5NHt3F5ZyNnYASIf5xseu84feQKjpBBwX-nIs29YXhMV8_76gf8cpBx-vaeUpFf-HGs2p2hq0x0mGoPN8kH0VhZGBCx3iClL1gICfwzbTw,,&typo=1
>> archives:  5/2017 thru present 
>> https://linkprotect.cudasvc.com/url?a=https%3a%2f%2fredfish.com%2fpipermail%2ffriam_redfish.com%2f&c=E,1,uahCj4-Us9Y3YmIN-yVtgMzfbZ9PQRVMSGkO6IYV0N9KUU9ZFY1kQa3BLPFvEuXo3Aq1MJ2Cz_0O_ne5GEfvWgd-OkD5gkJxAvRbSAfX3t8NMsKUtaHB-cDpzOrn&typo=1
>>  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://linkprotect.cudasvc.com/url?a=https%3a%2f%2fbit.ly%2fvirtualfriam&c=E,1,ISKO6LD8eQE8bCuHCQdZ0Gz68pToeZRW1lYvrOVc7g_fN_WR7boTcobKscWBk5yR1j-6b4R3aMqqzVOCJl2dqKUVnXNLqSONq-LfHDeMRXYahMWyKp3Af2c,&typo=1
> to (un)subscribe 
> https://linkprotect.cudasvc.com/url?a=http%3a%2f%2fredfish.com%2fmailman%2flistinfo%2ffriam_redfish.com&c=E,1,pC4zTC9TPXexDYqD_66fvJzl2DnWmxO_oNw6VBrLvm-RLF3uuPCVIIKR2OJQCiTV2oiJs8vos0DyebFkCE1wop0EL_KPEX9Ubs3iXHiL7U0CbzBDf23s&typo=1
> FRIAM-COMIC 
> https://linkprotect.cudasvc.com/url?a=http%3a%2f%2ffriam-comic.blogspot.com%2f&c=E,1,ZqqnEKAOuo-2XtVUgR8oNnQMs8sMJxA0yhR0UWNXecnKmVcuThV2HDMewZbrxblvfqZnqX-aDXbUG3c-hOx2euWFTs6SNgbC31UlEaVYkxcPI-I3NRM,&typo=1
> archives:  5/2017 thru present 
> https://linkprotect.cudasvc.com/url?a=https%3a%2f%2fredfish.com%2fpipermail%2ffriam_redfish.com%2f&c=E,1,YibNlqRVSsKKWCkGgCh8hBTgGfw3eLYOMmtcJqSQH0k5GE8ekETSArYhdys_xonOo_aLWqzXbvccM9FgCHeUXvK_w9dvfSzQRuPukdWmSf30uA,,&typo=1
>  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] new thermal tech

[Gary Schiltz]

Amusing indeed (the PV magazine article). The pump “reportedly
produces 3 kW to 4 kW of heat for every kilowatt of power it
consumes”. Say what?

On Fri, Jan 6, 2023 at 8:28 AM Roger Critchlow  wrote:
>
> I was amused to see an announcement of a thermoacoustic heat pump  the other 
> day:
>
>   
> https://www.pv-magazine.com/2023/01/02/residential-thermo-acoustic-heat-pump-produces-water-up-to-80-c/
>
> then an ionocaloric refrigerator announcement turns up this morning
>
>   https://newscenter.lbl.gov/2023/01/03/cool-new-method-of-refrigeration/
>
> It seems that you won't recognize your air conditioner in a few years.
>
> -- rec --
> -. --- - / ...- .- .-.. .. -.. / -- --- .-. ... . / -.-. --- -.. .
> 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] precision biology

[glen]

Very nice to see the Allen workflows plugged together this way. Thanks.

On 1/6/23 06:41, Roger Critchlow wrote:

https://www.nature.com/articles/s41586-022-05563-7 


Take your pluripotent human stem cell line, make 25 cell lines with different 
molecules tagged for fluorescent microscope imaging, run an automated pipeline 
to grow and gather images, align according to the apical basal axis, express 
the remainder of the spatial variation of the tagged features in a spherical 
harmonic basis.  Principal component analysis gets 70% of the variance in the 
first 8 components.

So starting with identical cells and growing in as identical conditions as a 
microsoft  fortune could buy, we get to populations of cells where at least 30% 
of the variation is uncorrelated noise?

I'm not sure whether I'm more impressed that the cells imposed that much order 
or more disappointed that the experimenters couldn't squeeze the population 
into a tighter cluster.



--
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/

[FRIAM] precision biology

[Roger Critchlow]

https://www.nature.com/articles/s41586-022-05563-7

Take your pluripotent human stem cell line, make 25 cell lines with
different molecules tagged for fluorescent microscope imaging, run an
automated pipeline to grow and gather images, align according to the apical
basal axis, express the remainder of the spatial variation of the tagged
features in a spherical harmonic basis.  Principal component analysis gets
70% of the variance in the first 8 components.

So starting with identical cells and growing in as identical conditions as
a microsoft  fortune could buy, we get to populations of cells where at
least 30% of the variation is uncorrelated noise?

I'm not sure whether I'm more impressed that the cells imposed that much
order or more disappointed that the experimenters couldn't squeeze the
population into a tighter cluster.

-- rec --
-. --- - / ...- .- .-.. .. -.. / -- --- .-. ... . / -.-. --- -.. .
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] new thermal tech

[Roger Critchlow]

I was amused to see an announcement of a thermoacoustic heat pump  the
other day:


https://www.pv-magazine.com/2023/01/02/residential-thermo-acoustic-heat-pump-produces-water-up-to-80-c/

then an ionocaloric refrigerator announcement turns up this morning

  https://newscenter.lbl.gov/2023/01/03/cool-new-method-of-refrigeration/

It seems that you won't recognize your air conditioner in a few years.

-- rec --
-. --- - / ...- .- .-.. .. -.. / -- --- .-. ... . / -.-. --- -.. .
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] more bullsh¡t

[David Eric Smith]

Your use of chatGTP, Pieter, is to my mind a very interesting thread.

There was a columnist for the New York Times many years ago, named William 
Safire.  I don’t even remember now what he wrote about, but he was known, and 
was significant to me, for being an example of “a good writer”.

Safire wrote something (column?  book?  article?) with the theme that, if one 
would write creatively, one should first do several years galley-rowing as an 
editor somewhere.  People who have an impulse to write “creatively” imagine all 
kinds of innovation in language that will be just dramatic and wonderful.  
Editors, who have had to deal with those imaginations in the writings of 
others, know that most such ideas are awful and need to be beaten out of the 
writer if he or she is ever to become good.  So Safire’s thesis was that you 
really need to do this, for a number of years and a large number of other 
people’s pieces, to squeeze the nonsense out of you and develop a solid 
understanding of your language.  Kind of like, in books on French cooking, the 
author says “why did we have to spend the first year cooking mixed vegetables 
in mayonnaise over and over again; I don’t even like mixed vegetables in 
mayonnaise.  To which the answer, of course, is that one develops what the 
French term “method”: experience with uniform sizing of each ingredient, 
correct relative sizing across ingredients, time of introduction to heat, and 
on and on, so that one gets control and has everything cooked to the intended 
degree reliably.  Only then has one gained the tools to create.

I have run Safire’s thesis by some writers I know to see what happened; my 
notable memories are the ones who hate it and think it is completely wrong.

But to Glen’s point that we should think of literary AI the way we think of 
pocket calculators (another thing I was not allowed to have in school; my 
parents thought it would make me stupid).  ChatGTP can be sort of the William 
Safire level of basic method in language, not intending or intended to create 
anything, but somehow, as you say, to find a kind of solid and central 
expression for things.  One might even think of the appropriate training 
schedule for a tool meant to do just that, which could be a bit different from 
the ad hoc training that is probably first-gen of these tools.

Eric

> On Jan 6, 2023, at 12:59 AM, Pieter Steenekamp  
> wrote:
> 
> As a native of South Africa, I have personally witnessed the shortcomings of 
> both our public primary and secondary education systems and the financial 
> barriers that prevent many from accessing private schools. In response, I 
> have dedicated the past year to establishing a private institution that is 
> not only affordable, but also committed to providing a high-quality 
> education. In reflecting on what constitutes a truly valuable education, I 
> have come to the conclusion that the most essential outcome is not the 
> acquisition of academic skills, but rather the development of strong 
> relationships - both with oneself and with the outside world. While it is not 
> possible to directly teach children how to cultivate such relationships, it 
> is possible to create an environment in which they can learn and grow through 
> unsupervised interactions with their peers.
> 
> Full disclosure: I have not kissed  Blarney Stone and my ability to write (or 
> speak for that matter) eloquently is just awful. I've written a paragraph and 
> then I asked chatGPT, who have kissed the Blamey Stone, to rephrase it more 
> eloquently. The above paragraph reflects exactly what I wish to say, but is 
> just expressed so much better.
> 
> On Fri, 6 Jan 2023 at 00:39, glen  > wrote:
> AI, Teaching, and "Our Willingness to Give Bullshit a Pass"
> https://dailynous.com/2023/01/05/ai-teaching-and-our-willingness-to-give-bullshit-a-pass/
>  
> 
> 
> The first time I heard this argument was from these guys:
> 
> https://www.audible.com/pd/Pill-Pod-104-AI-the-New-Crisis-of-Humanities-Education-Podcast/B0BPQ77Z8P
>  
> 
> 
> My phrasing of the idea being that tools like ChatGPT are analogous to 
> calculators, allowing the computer to do what it's good at and freeing humans 
> up to do what we're good at. Why require students to learn bullshit 
> rhetorical styling when we can teach them to think about the *substance* ... 
> a lesson many of us learned from Knuth's TeX a long time ago. The trick is 
> that tools like ChatGPT are built around the bullshit-generation use case. 
> What we need are tools built around the bullshit-detection use case.
> 
> With branch