Death to individualism

by liberal japonicus

I flagged a story about Air Canada's use of an AI chatbot and how they tried to avoid responsibility for the information it gave out and Janie pointed out in a comment that JB's John Cole mentioned it. Take it away, JC

In other news, the Air Canada Chatbot lawsuit thing is quite a hoot. For those who have not heard or read about it, apparently a guy’s mother died, so he was trying to buy tickets from Air Canada to go home. He didn’t have time to wait for an approval from them, so he went to their website, used the chatbox, and asked if in their bereavement policy it was allowed to buy the tickets ahead of time and request a refund later. Unbeknownst to the man, he was not actually talking to an employee at Air Canada, but an AI chatbot. And that AI chatbot had just created that policy out of the ether. Which just goes to show you that artificial intelligence has more common sense and empathy than the MA’s and corporate lawyers at a major corporation.

At any rate, the obvious happened. The guy said “Ok that seems reasonable” and went and purchased his tickets. Later on, he tried to get reimbursed, and living Air Canada employees told him to go get fucked that that was not their policy. So he sued, and the Air Canada lawyers actually argued in court that technically the AI chatbot on their website is not an Air Canada employee so technically we are not liable and don’t owe him a refund.

At which point, everyone on the tribunal said “get the fuck out of here with that bullshit” and found them liable:

Here's the Guardian link. His point that a chatbot actually has more empathy and common sense than the corporate lawyers is nice and a lot more cheerful than mine. I find the argument by Air Canada that the chatbot is responsible for it's own counsel links up with my feeling that the use of these tools is going to reflect different cultures, which I posted about earlier here. To return to that hobby horse, it's difficult for me to imagine an Asian company making that argument. This is not because Asian companies have some sort of greater integrity, it's just that they won't use 'just because we programmed that bot, it doesn't mean we are responsible'. (one point that I'd make is that you now need to screenshot any interactions with a chatbot, it's because this guy did that he was able to get the tribunal ruling in his favor)

Slowly, my Japanese colleagues are finding that their students are turning in assignments done with chatGPT and are lamenting. However, there is still no sense of crisis, no concern of what this does to their learning. I believe it is because they don't really have a concept of learning as a way to become more individualistic and as the Japanese system rushes to adopt bits and bobs from Western (basicaly US) education, they are just stuck to the outside.

I may have told this story in the comments, but we are being encouraged to create can-do statements for our classes. Now, Japanese university students who fail a class, say 1st year [whatever], are told that they have to take that class at the same time as they are 2nd year English [whatever] so I said gee, if we have can-do statements, that means the students need to first pass the 1st year course before moving on to the second year, what a great idea! To which my colleagues said no, why would that be the case? This is just one example, we have lots of things nailed here and there that are to adhere to some letter of the law, but not to the spirit.

What generally happens here is that everyone takes advantage of the system until it reaches the breaking point and then draconian rules are put in place. Enforcement is done because everyone is watching you, not because you feel a duty to do the right thing. If you can then get a certain percentage of the population to conform, you develop something like a herd immunity.

Anyway, that's what I was thinking when I read about Air Canada's chatbot 'going rogue'. If you had thoughts about this or anything else, knock yerselves out.

Comments

[CharlesWT]

Chatbots are about as bad now as they're ever going to be.

I was at first impressed with the free versions of Google's Bard/Gemini. They did several things better than the other use-for-free LLMs. But they have since been front-loaded with so many guide rails and nudges that they're often useless. They often refuse to answer seemingly innocuous questions. And to add insult to injury they will give you a moralistic lecture.

Similar to the Air Canada case, there have been instances of people talking car dealership chatbots into selling them a car for a dollar.

Illuminaries in the AI industry are predicting artificial general intelligence any day/week/month/year now. There are rumors and conspiracy theories that OpenAI has already developed an AGI and is keeping it under wraps. Some of the current frontier LLMs being talked about have very impressive capabilities.

In anticipation of the coming demand for computing capacity, Nvidia's market value has been boosted to about two trillion dollars. That's about the value of the entire Canadian economy.

[wj]

Have just spent some (definitely non-)quality time (loke an entire afternoon) with ChatGPT. Trying to create a Configuration Manual (how to) for a new software product. Gave it, as additional input, the directions being used internally by the developers. Results?

Well, the first couple of sections weren't too horrible. Needed some tweaks, but not bad. But then, the rest of the sections were utterly unlike anything real. And big chunks of the directions in the internal doc had just totally disappeared, and needed to be manually added. All this in a document that, exclusive of title page, copyright, and index, was maybe 7 pages long when I finished with it.

In summary, the AI gave a possibly useful skeleton. Possibly. But beyond that, I certainly wouldn't want my company's name on what it produced.

[CharlesWT]

Sometimes meta prompts will work. Such as "Write the prompt I would need to give you for you to write a configuration manual from this data."

[wj]

Thanks, Charles. I'll give that a try.

[nous]

Well, on the subject of letting some clever bit of tech do part of the work for you, I ordered an electric mountain bike this weekend. It cost as much as a 125cc enduro bike and still requires me to do a big portion of the work, but will help put a bigger dent in the local hills.

Quite the technological marvels, this latest generation of mid-power, light e-mtbs. Looking forward to hitting the trails with it when spring break arrives in a few weeks. It will be a far different experience than my first spring break with a mountain bike that didn’t even have a suspension fork. But then it’s also been a long time since I could put in an 80 mile ride without a second thought.

It remains to be seen how many miles I’ll be able to get out of the battery on the bike, but neither of us will be up to an 80 miler.

[wj]

nous, you might find this of interest.
https://www.washingtonpost.com/opinions/2024/02/22/artificial-intelligence-sal-khan/ (Paywalled. Sorry.)

It seems that someone has made notable progress at creating an AI which is actually useful when teaching, at least at the high school level. The Khan Academy may have put out stuff elsewhere that isn't paywalled, but I don't have time this morning to dig deeper.

[CharlesWT]

Here's a somewhat elaborate approach to roping an LLM into helping you engineer a prompt for a particular problem. I've used it with some success several times.

"I want you to become my Prompt Creator. Your goal is to help me craft the best possible prompt for my needs. The prompt will be used by you, ChatGPT. You will follow the following process: 1. Your first response will be to ask me what the prompt should be about. I will provide my answer, but we will need to improve it through continual iterations by going through the next steps. 2. Based on my input, you will generate 3 sections. a) Revised prompt (provide your rewritten prompt. It should be clear, concise, and easily understood by you), b) Suggestions (provide suggestions on what details to include in the prompt to improve it), and c) Questions (ask any relevant questions about what additional information is needed from me to improve the prompt). 3. We will continue this iterative process with me providing additional information to you and you updating the prompt in the Revised prompt section until it's complete.

Keep revising until you're happy with it, then simply copy and paste the revised prompt into a new chat."

[nous]

wj - The Khanate stuff looks interesting, and would certainly be better than most of what the LLM management teams get enthused over (as opposed to the things that the research teams worry over, and get overruled on). It's certainly useful for explaining how one goes about calculating a thing, and perhaps even some common metacognitive strategies for applying those calculative skills. On the qualitative side, it looks like it does a decent job of engaging in content delivery of the large lecture model, with a side order of Q&A.

My concerns would lie at the junction between the content delivery and the query response. I know from long experience that the most valuable bits of a student's actual education come from the back-and-forth questioning and listening between the people involved in the classroom community steered by a curious, experienced, and critically aware facilitator who is responsible for maintaining a proper degree of rigor without letting bias distort the conversation.

That last part is crucial. Without it, you end up with people who are technically competent specialists, but whose learning has been largely self-directed, and lacks sufficient corrective input for matters of judgment and perspective (like my brother, the professional engineer and serial conspiracy theorist).

I fear that the educational economies created by something like Gen AI will produce an educational ecology with too small a population of experts to sustain that facilitator/interlocutor role that ensures that the system remains strong enough to deal with novel situations and moments of punctuated equilibrium.

Which pretty much describes the future we are heading towards.

Don't want the intellectual shelves to look like the grocery stores during the early days of the COVID pandemic. Gotta ensure that critical intellectual capacity exceeds peak demand. I fear that AI is too efficient, in the economic sense, to leave us the reserves we need for emergencies.

[CharlesWT]

"At the recent World Government Summit in Dubai, Nvidia CEO Jensen Huang made a counterintuitive break with what he thinks is a long tradition of tech CEOs advising youngsters to learn how to code. Huang argued that, even at this early stage of the artificial intelligence (AI) revolution, programming is no longer a vital skill. With coding taken care of by AI, humans can instead focus on more valuable expertise like biology, education, manufacturing, or farming, reasoned the Nvidia head."
Jensen Huang says kids shouldn't learn to code — they should leave it up to AI: But this isn't the first time a tech exec has predicted the death of coding.

[nous]

Also, I try to keep in mind that what Khan Academy does most is tutoring, which overlaps with, but is not identical to, education. They are there to build scaffolding for the learning. It's like having a personal trainer, moreso than having a coach/mentor.

They often do some educating along side the tutoring, but the focus is on the scaffolding/support part.

---

I think Hoang is missing the importance of learning to code in shaping critical understanding of how to make AI do what we need it to do. It's like trying to learn higher level mathematical concepts without understanding how calculation methods can affect results. You may be able to farm out parts of the work, but you need to know enough of the process to make the right critical decision for how to get to the right end result.

[Tony P.]

Human: LLM1, I want you to become my Prompt Creator.
LLM1: What should the prompt be about?
Human: Arguing with Libertarians(TM).
LLM1: a)Try "banging your head against the wall". b)Include words like "guns", "abortion", "insurrection", "vaccination", and "responsibility". c)Ask "Whose Liberty(TM) do you advocate for?" or "How local -- federal, state, municipal, neighborhood, household -- do communal decisions have to be for Libertarians(TM) to accept them as legitimate?"
Human: Perfect. Thank you.
**********
Human: LLM2, please give me the True Libertarian(TM) answers to the above questions.
LLM2: Any True Libertarian(TM) advocates for the Liberty(TM) of every individual person. For Libertarian(TM) purposes, corporations are persons. All collective decision-making is an infringement on Liberty(TM). Voluntary associations may make only those collective decisions with which True Libertarians(TM) who are members of the association agree. For Libertarian(TM) purposes, corporations are not "associations". City councils, state legislatures, and the federal government are not voluntary associations of individual persons; they are collectivist cabals bent on suppressing individual Liberty(TM). True Libertarians(TM) are equally opposed to laws which forbid them to walk down the street armed and to laws which forbid them to walk down the street naked. True Libertarians(TM) are not obligated to defend anyone's Liberty(TM) but their own. In a leaky lifeboat, a True Libertarian(TM) will not bail if the other end of the boat is sinking. All True Libertarians(TM) are unanimous about these precepts, unlike sheeple who mindlessly hew to different precepts.
Human: Hooo, boy ...

--TP

[CharlesWT]

A libertarian might raise several objections to the characterizations presented:

1. Corporations as Persons: While libertarians generally advocate for individual liberty, the characterization that corporations are persons might not be universally accepted among libertarians. Some libertarians might argue that while corporations are made up of individuals, they are legal entities created by the state and should not necessarily be granted the same rights as individuals.

2. Collective Decision-making and Associations: Libertarians typically value voluntary associations and decisions made through voluntary means. However, the characterization that all collective decision-making is an infringement on liberty might be seen as overly simplistic. Libertarians might acknowledge that some collective decision-making, particularly when voluntary and consensual, can enhance individual liberty rather than detract from it.

3. Government and Individual Liberty: While libertarians generally advocate for limited government and individual liberty, the portrayal of all forms of government as inherently suppressive might be viewed as an oversimplification. Some libertarians might argue for the necessity of certain minimal government functions to protect individual rights and maintain social order.

4. Uniformity of Beliefs: The characterization that all "True Libertarians" hold identical beliefs might be objected to by libertarians themselves. Libertarians, like any political group, hold a diversity of opinions on various issues, and there is no single, universally accepted set of beliefs that all libertarians adhere to.

5. Selective Defense of Liberty: The notion that libertarians are only obligated to defend their own liberty might be challenged by some libertarians. While individual liberty is paramount in libertarian philosophy, many libertarians also recognize the importance of defending the rights and liberties of others as part of a broader commitment to the principles of liberty and justice.

6. Leaky Lifeboat Scenario: The depiction of a True Libertarian refusing to bail out a sinking end of a lifeboat might not align with the principles of compassion or voluntary cooperation that many libertarians value. While libertarians prioritize individual autonomy, they also recognize the importance of voluntary cooperation and mutual aid in certain situations.

In summary, while the characterizations presented may resonate with certain strands of libertarian thought, they do not necessarily capture the full spectrum of libertarian perspectives, which are diverse and multifaceted. Libertarians may object to oversimplified portrayals that fail to capture the complexities and nuances of their beliefs and principles.

—ChatGPT 3.5

[wj]

tutoring, which overlaps with, but is not identical to, education. They are there to build scaffolding for the learning. It's like having a personal trainer, moreso than having a coach/mentor.

And, as we have seen in sports, you have to have both for optimal results. I expect it will take us a while to get to the right balance. Which, as always, is hard on the cohort which is in the system during the transition. But I think there is some reason to be hopeful that we will end up with AI-based tutors to help with some parts of learning, while human teachers (as distinct from instructors**) help children learn how to think and how to evaluate information.

** The distinction I am trying for, and I may well be using the wrong terms, is that one group teaches facts -- how to find them, and how to mechanically manipulate them (e.g. arithmetic), etc. The other group teaches processes -- and how to evaluate the accuracy of sources, how to combine ideas about facts in to a coherent structure, etc. In short, how to learn.

[nous]

wj - your word choice is adequate to your purpose and clear enough to follow. I'm not going to worry about how insiders would describe those same concerns in a venue that consists mostly of competent outsiders.

But I think there is some reason to be hopeful that we will end up with AI-based tutors to help with some parts of learning, while human teachers (as distinct from instructors**) help children learn how to think and how to evaluate information.

Again, though, I am concerned about the economies of such things, and the knock-on effect of asymmetrical access. AI-based tutoring is being developed by capitalist corporate entities who will attempt to restrict and monetize that tutoring. The benefits - much like a lot of academic tutoring - will go to those who have the resources to make use of the service. And at a certain economy of scale, public institutions of learning will decide to offload integration of their own tutoring onto those private institutions.

When they do so, they will either decide to leave the tutoring laissez faire, or they will decide to fold that service into their budget and fees structure, and either raise tuition and fees, or cut other services. Either way, though, there will be a reduction in the number of people trained to be experts in these subjects (who only work as tutors because they are part of the pool of surplus educators unable to find another outlet for that side of their expertise).

Which again brings me around to that worry about having enough expertise to respond to the coming crises we will face due to climate change effects.

We need widespread expertise deeply represented in diverse communities across the globe at all economic levels. Too much inequality will act as a sort of stress riser if our systems are built with too little margin of safety.

[Tony P.]

CharlesWT,

That ChatGPT response is kinda heavy on "might" and "generally", but impressively articulate nonetheless. Care to share the prompt you gave it?

I have never used one of these LLM thingies, but when I get around to it I will probably ask:

1)Based on his comments on Obsidian Wings, how libertarian is CharlesWT?

2)Based on his comments on Obsidian Wings, how socialist is Tony P.?

Or maybe vice versa.

--TP

[CharlesWT]

AI-based tutoring might bolster this suggested approach to education in developing countries.

"Our new peer-reviewed research shows that two proven, affordable policies can make an amazing difference.

The first helps pupils learn more effectively. Almost universally, school classes put all nine-year-olds in one grade, 10-year-olds in another, etc. But many children in each of those classes are either far behind and ready to give up or far ahead and bored.

An effective way to address this problem is to use tablets to teach students one hour a day. With existing educational software, the tablet quickly assesses and starts teaching at exactly at the student’s level. For one hour a day, that student is taught at his or her right level, boosting learning. After just one year, testing shows, the student has learned what would normally have taken three full years.

The second proven strategy is “structured pedagogy” — helping teachers teach better. A trial in Kenya was so successful the approach was adopted for the whole country. With a full year of semi-structured teaching plans, coaching and encouraging text messages, the project helps teachers provide more engaging and useful instruction. Studies show this delivers learning that is equivalent to almost one extra year of schooling."
Two proven ways to help kids learn: Every dollar spent on these two educational initiatives worldwide would deliver $65 of social benefits

[CharlesWT]

Care to share the prompt you gave it?

"What objections could a libertarian make to these characterizations of libertarians?"

Google Gemini gave this response to the same prompt with some fingerwagging included.

Libertarian Objections to Characterizations

[Michael Cain]

I think Hoang is missing the importance of learning to code in shaping critical understanding of how to make AI do what we need it to do. It's like trying to learn higher level mathematical concepts without understanding how calculation methods can affect results.

I suspect that Hoang is saying that there's no sense in learning procedural coding. OTOH, declarative coding will be critical to "make the AI do what we need it to do." I write a one-page description of what the program is supposed to accomplish, and the AI generates the 50,000 lines of procedural code that specifies how a computer actually accomplishes it.

From time to time during my technical career I had to write functional requirements for a software module: a description of what it should do but not how it should do it. It's a black art, much like phrasing wishes to a genie so that you get what you want, not something that conforms to the letter of the wish but bites you in the backside. All of the use cases. All of the invariant conditions that have to be satisfied by the internal data structures at externally visible points.

This morning I finished the first version of the next illustration for the granddaughters' fairy tale. This afternoon I spent some time with one of the free online AI image generators trying to approximate it. I never did get very close. In my drawing, the dragon goes back to the old German and is mostly a giant snake. Put "dragon" into the prompt and the software instantly goes for the more modern version with four legs and wings and an incredible number of pointy protuberances. When I used "wyrm" instead it got closer. Never could get it to not put spines down the back, though.

[nous]

CharlesWT - neither of those suggestions that Lomborg proposes has anything to do with AI, and the first proposal is completely dead in the water for implementation in the circumstances covered in the second case. It would require putting tablets in Kenyan classrooms that currently cannot afford to put enough textbooks in those classrooms.

Even if you could put tablets in those classrooms, you couldn't actually get them delivering the instruction in Kiswahili, or get the tablet to read out the text in Kiswahili for the students (who are, after all, learning to read).

Might as well just suggest that everyone adopt the (fictional) Young Ladies Illustrated Primer from Neal Stephenson's The Diamond Age. About the same level of science fictionality in the two proposals.

Lomborg never fails to disappoint. Full of can-do thinking that can't.

[nous]

Michael Cain - my wife's students have tried to use AI to generate art for graphic novel projects. In every case the problem that they discovered was that it was almost impossible to create sequences of images with both the right content and with enough visual continuity between frames to look like it all belonged together sequentially.

(And, preemptively...I'm sure that there's something to be done to massage the prompt or iterate the processing and get closer to what was intended, but that sort of prompt scripting requires the sort of thick description that students also struggle with creating in their ordinary writing. And the thick description requires attention to detail that they also find challenging. Etc.)

[CharlesWT]

It would require putting tablets in Kenyan classrooms that currently cannot afford to put enough textbooks in those classrooms.

His argument is that this is one of the areas where first-world countries and philanthropists can spend their foreign aid money to the greatest effect. He may be wrong but this and other initiatives have been studied in detail by economists to determine the likely cost and benefits. He has a list of 12 initiatives with a price tag of about $36 billion. Which is less than a lot of things foreign aid gets spent on.

[CharlesWT]

In every case the problem that they discovered was that it was almost impossible to create sequences of images with both the right content and with enough visual continuity between frames to look like it all belonged together sequentially.

This problem is being worked on. One approach is to generate a seed from an image. When the seed is used with a prompt the image generated will be in the same style as the original image.

Plus context windows are getting larger. Some recent frontier AIs have context windows of a million to ten million tokens. Enough for them to keep an hour's worth of video in mind as they respond to prompts about it. A large context window should help in generating images with continuity.

[Pro Bono]

I'm raising both eyebrows at these suggestions of what AI can do.

The stuff I've seen is undergraduate-level internet fact-searching, rendered into somewhat clunky prose. If the AI is given a technical problem, it comes up with maths salad.

Has it secretly got much better?

[Deleted several duplicates of this comment. Typepad must be having a hiccup today. - ed.]

[Nigel]

..it was almost impossible to create sequences of images with both the right content and with enough visual continuity between frames to look like it all belonged together sequentially...

That problem appears to have been cracked as far as video is concerned.
https://www.vox.com/future-perfect/24080195/sora-openai-sam-altman-ai-generated-videos-disinformation-midjourney-dalle

[Nigel]

I filmed 3 separate bugs in an app and gave the videos to Gemini 1.5 Pro with my entire codebase.

It correctly identified & fixed each one.

AI is improving insanely fast.

[Nigel]

Link.
https://twitter.com/mckaywrigley/status/1762175776726663267

[GftNC]

I think this is an Open thread ("If you had thoughts about this or anything else"), so this piece was in today's Grauniad by Rhiannon Giddens. a) I think it's very good, and interesting, on the issue of appropriation of musical genres/traditions etc, and b) it reminded me of the first time I became aware of her, on Jools's Hootenanny, singing this, absolutely electrifyingly:

https://www.youtube.com/watch?v=xadnGzyYFq8

I also find her very beautiful, and, unusually, completely un-glammed up for someone in her profession. Bien dans sa peau.

[Michael Cain]

...my wife's students have tried to use AI to generate art for graphic novel projects. In every case the problem that they discovered was that it was almost impossible to create sequences of images with both the right content and with enough visual continuity between frames to look like it all belonged together sequentially.

During the Renaissance it was not uncommon for the master to draw sketches and the hierarchy of supervised apprentices do much of the grunt work while they learned the trade (and the master's style). I can envision a popular tool that is trained/can be trained to take a sketch and do a finished draft in a style it has learned. Perhaps allowing a human to fix it, and the software learning from that as well.

Perhaps I'm just terrified of details. I went back the other day and looked at some of the Tenniel illustrations from the Alice books. It's even worse when you consider how they made plates of those for the printing presses.

[JanieM]

GftNC --thanks for the Guardian reference.

The Waterboy video is "blocked in your [my] country" -- but here's another Rhiannon video -- go to 1:38 -- the last song and the encore are lovely. *She* is lovely, and i agree with you that some of it is her lack of artifice.

[Michael Cain]

It would require putting tablets in Kenyan classrooms that currently cannot afford to put enough textbooks in those classrooms.

It was about 25 years ago that I started whining to everyone I knew who worked at a tech company that did hardware research for "Mike's slab of plastic computer". The idea was for a roughly 9x12x0.5 inch slab of plastic, completely sealed. One of the big faces incorporated a display and high-resolution touch input. Wireless charging and networking. At the beginning of a term the students would download an e-copy of each of their textbooks. Thing also would function as a pad of "smart" paper. Needed to be cheap enough that after a semester of not needing paper books, the school would be ahead.

Granted I wasn't thinking about Kenya for it. I was just tired of watching fourth- and fifth-graders bent over on the walk from the bus stop to their houses because their backpacks weighed so much.

[JanieM]

More Rhiannon-related, also lots of cool images in this one:

https://www.youtube.com/watch?v=dWUa7aAIfLE

And one more, from after John Prine died early in the covid era:

https://www.youtube.com/watch?v=Qi-AHqQ48EM

[Hartmut]

At the moment a priority is to get the kids in Africa desks to sit at instead of on the floor.
Lawrence O'Donnell runs a fund (KIND = kids in need of desks) that finances that* and also scholarships for girls.
Tablets are rather low on the list of priorities, I presume

*the desks are produced in Africa and thus also create jobs

[GftNC]

Ha - I'm glad I'm not the only absent minded one around! I forgot to link the Grauniad piece:

https://theguardian.com/music/2024/feb/27/black-artistry-is-woven-into-the-fabric-of-country-music-it-belongs-to-everyone-beyonce-texas-hold-em-rhiannon-giddens

and Janie forgot to link the thing where we were supposed to go to 1.38!

So now I am about to follow the two links in Janie's 02.12, and (for anyone who would like it and can't get the Hootenanny one in the US) here is another one of her singing Waterboy (much more glammed up - in the hootenanny she was wearing no makeup, and her hair wasn't "done"):

https://www.youtube.com/watch?v=VbeekjZP-CI

Also on the same Hootenanny she sang a memorable St James Infirmary (one of my favourite songs) with Tom Jones, but I'm sorry if US people can't see it:

https://www.youtube.com/watch?v=rhVT0fx3rFs

[GftNC]

God, Janie, I loved those two links. The first one contained one of my favourite images:

https://www.teenvogue.com/story/baton-rouge-protest-photo-woman-dress-face-off-police

[nous]

I understand the attraction of the tablets-for-education plans. A lot of my early work in writing studies involved digital education initiatives, and I was an early adopter of several technologically based approaches to teaching. But during that time I've also seen program after program fail due to poor implementation or support, or crumble under the lack of institutional support and infrastructure required to keep the program running smoothly. And I've watched one hour of instruction become a half hour of instruction combined with a half hour of troubleshooting and tech support. Let me tell you, the latter is a huge source of burnout among teachers.

Anyone interested in the subject should take a few hours to look over the dozens of articles written about the LAUSD/Apple program that gave all students an iPad. No, don't start objecting to why that plan was dumber than what you would do, go and look at the various problems that arose from implementation and get an idea about what it looks like to actually try to implement a plan like this at the classroom level. Then go back and reexamine these other proposals in light of what we have learned to get a better idea of how feasible and effective they might be.

Everyone treats teachers as if they are the failure point that tech is meant to fix, rather than as the people who have been trying to use tech to fix education for thirty some years already. I've been a part of dozens of programs meant to integrate tech. I've got success stories and cautionary tales. It would be nice if someone were to listen to those rather than try to work out the sales pitch for the next thing they want to try to magic pill a difficult task with little actual understanding of how education actually gets done.

And before we do any of that, maybe make sure that the lights and HVAC work and that there's enough desks and classroom space for everyone, and that everyone in the room has been fed.

[JanieM]

Sorry about the missing link, and I played it and was reminded that she talks from about 1:38 to 1:42, when the 2nd to last song actually starts:

https://www.youtube.com/watch?v=Ignhso0iv9U&t=4s

(I plead babysitting. ;-) )

[Girl from the North Country]

She's just fantastic! I loved that (particularly the last song). Also, when she talks from 1.38.00, her concept of "enough" was so familiar from what you have often talked about. Thanks, Janie, for the generally uplifting interlude provided by all those links.

[JanieM]

During the Renaissance it was not uncommon for the master to draw sketches and the hierarchy of supervised apprentices do much of the grunt work while they learned the trade (and the master's style).

When I lived i Milwaukee in the early 80s I knew an artist who did wonderful drawings for children using a system like this -- she conceived the designs, art students helped color copies for sale. (I think they were done in colored pencil.)

I wonder how common that is now....

[JanieM]

A sample of her art. "We" (my family) have a bunch of them.

[Girl from the North Country]

Further to our recent discussions about words losing their true meanings, I am in the middle of a series called The Invisible Library, which is reasonable fun, and near the end of the fourth book (The Lost Plot) was delighted to come across a plea and a definition for the proper meaning of "disinterested". Three cheers for literate fantasy!

[Michael Cain]

The Invisible Library is fun.

[Michael Cain]

I wonder how common that is now....

A superficial web search suggests that there are so many different ways of producing paintings/drawings/whatever in quantity that I'm sure someone does that.

Instapaint is basically a clearing house to connect customers with artists. The artists will use a photo provided by the customer to render hand-done painting/colored pencils/charcoal/whatever reproduction. At what strikes me as ridiculous prices and schedules. Eg, a 24x36 inch oil painting in eight days for a bit over $200. Heck, Staples charges me $33 for that size poster-from-PDF out of their laser printer.

One of the most interesting were Chinese "studios" that produce thousands of copies of a single masterpiece. Hundreds of canvases hung in a warehouse. An artist with a big glob of white paint starts down the row, drawing one or a few strokes, then moves to the next canvas and draws the same strokes, etc. Another artist comes after them doing the same thing with one or a few strokes of blue. The article described the artists as artists: they understand exactly how to make a stroke so the brush marks look like the original.

Me, I'm a hack and I know it.

[hairshirthedonist]

The Henry Ford School of Art

[Donald]

https://www.politico.com/news/magazine/2024/02/28/government-buying-your-data-00143742

Most of this I sort of assumed without knowing details, but I didn’t know that our car movements can be tracked because of the tire pressure monitor.

And the last paragraph has a Handmaid’s Tale quality to it.

[JanieM]

And....

The TREAD (Transportation Recall Enhancement, Accountability and Documentation) Act was signed into United States Federal Law in the fall of 2000 and requires all vehicles manufactured on or after September 1, 2007 to be equipped with a tire pressure monitoring system.

In the United States, TPMS is mandatory for all vehicles produced after September 2007.

In the European Union, TPMS is mandatory for all new passenger vehicles released after November 1, 2012. From November 1, 2014, all new passenger vehicles sold in the European Union must be equipped with TPMS.

[Snarki, child of Loki]

Having had multiple cases where the car dealership didn't bother resetting the TPMS after changing tires (summer/winter changeover), I went and got the tool to do it myself.

And found that the range on those TPMS sensors is *tiny*. 10cm, maybe?

So while the modern car Sirius, Starlink, and cell-phone connections will certainly provide lots of tracking info, I have *serious* doubts that TPMS can do that.

[Michael Cain]

By accident more than intent -- I don't drive much so don't buy new tires very often -- all four of my TPMS sensors have died. The TPMS light on my dashboard is always on. Next September I'll buy new tires whether I really need them, and be sure to pay for new sensors.

[Tony P.]

How soon can we replace the SCOTUS with AI? Even artificial intelligence would be better than the stupidity of pretending that He, Trump's immunity claims are something that needs Nine Wise Souls to deliberate over it.

The Handmaid, the Frat Boy, the Receiver of Stolen Goods, The God-botherer, the Pervert, and the Caller of Balls and Strikes himself (or at least 4 of them) must think they could ask for briefs on the shape of the Earth next, and not be held in contempt by laymen. Lawyers might be another story. Chat-GPT could go either way, I suppose.

--TP

[Michael Cain]

I don't drive much...

Called to make a service appointment for my 2008 Fit at the Honda dealer today. The dealer's software takes my phone id and pulls up the appropriate service records automatically. The woman blurted out "Jesus!" Then apologized and told me she didn't think she'd ever seen a record for a Fit that old with so few miles.

I've thought seriously about replacing the Fit with an EV. All evidence suggests that I will probably have to give up driving due to age before the Fit wears out. So I now plan to keep it until I can replace it at reasonable cost with a used EV that can double as a battery backup for my residential electricity.

[hairshirthedonist]

My car has an indirect TPMS, meaning tire pressure is somehow detected by the ABS. (I just learned this after reading Janie’s comment.)

I was never sure if the light on the dash wasn’t simply a periodic reminder to check the tire pressure. You turn it off by pressing and holding a button in the glove box. Otherwise it stays on even if you ensure that your tire pressure is correct.

Maybe my indirect TPMS makes me Secret Squirrel.

[CharlesWT]

My car has an indirect TPMS, meaning tire pressure is somehow detected by the ABS.

Tires with different pressures have different rotation speeds.

[JanieM]

must think they could ask for briefs on the shape of the Earth next

I'm reserving this comparison for the self-appointed omniscient deities in the judiciary (and elsewhere) who know when a human life begins, and have decided that they're in charge of women's bodies, family planning, sex, and other trivial matters that need not concern the rest of us peons.

[Snarki, child of Loki]

The Supremely Deplorable Six (Lawless John, StripSearch, Token, Boof, Coathanger, and Squi) will engage in Olympic-level legal gymnastics to avoid the obvious conclusions they should reach instantly.

Worst Enid Blighton series, EVER.

[Hartmut]

Let's see, whether they can come up with something topping even Alabama's 'extrauterine children in a cryogenic nursery' [no, that is not from The Onion].

[hairshirthedonist]

During an interview on NPR with a supporter of the Alabama decision they played an audio clip of someone, not sure who, but someone worthy of an audio clip, claiming that we've learned so much in recent decades, that the scientific knowledge that the sperm and the egg combined to form an embryo didn't exist when Roe was decided, and that they probably would have decided differently with today's fancy, new sperm-and-egg knowledge. (I paraphrase.)

The zygote and blastocyst precede the embryo, but still, you don't get to the embryo without the sperm and the egg. So, okay, fine on that point. But, either way, I'm pretty sure this was all well established long before 1973. Somehow, no one said anything about this utterly bogus assertion.

What really bugs me is that it is only because of the people who have done the science and passed it on to humanity that these yahoos have any clue what's going on "when babies are made." Yet they can't recognize the difference between, one one hand, an undifferentiated collection of cells - no brain, no heart, no other organs, no nervous system, no feelings, no senses - that can be frozen for a decade and, on the other hand, ... a F**KING CHILD!!!

I don't believe for one second that these people really equate embryos in a laboratory with children, even if they think they do in some abstract way. Maybe it's a silly hypothetical, but imagine a lab with a few hundred frozen embryos on one side of an isolated town of radical "pro-life" evangelicals. The town has one small fire department. On the other side of town is an elementary school full of kids. The lab and the school both catch fire. Which one does everyone in town want the fire department to go to first?

Is anyone going to have a second's doubt about that choice?

[Snarki, child of Loki]

Alabama is now executing prisoners with Nitrogen, but still keeps their snowflake-babbies in liquid Nitrogen.

A land of contradictions.

[CharlesWT]

Alabama's purpose is to make Texas look civilized...

[Hartmut]

Of course the fire department will try to protect the church first and there will be discussions whether the school burning down is not actually a good thing. I have my suspicions that the fires are not accidental in the first place.

Technical question: is it an embryo before implantation to begin with? (I.e, is fertilized egg = embryo?)

[JanieM]

Hartmut -- this is google's top result, from the Cleveland Clinic (a very respectable institution) and often a useful website:

After conception, a fertilized egg travels through your fallopian tubes to your uterus. The fertilized egg (called an embryo) implants (attaches) into the wall of your uterus. This triggers the placenta to form. Your placenta begins producing and releasing human chorionic gonadotropin (hCG) into your blood and pee.

https://my.clevelandclinic.org/health/articles/11585-conception

[JanieM]

But also, in the twisted minds of the people who want to rule us all, a fertilized egg is a human being with far more rights than *any* living breathing female human being, and than many others of other kinds as well.

[Girl from the North Country]

Further to recent discussions on myself and yourself, I just heard the independent candidate who came second to the ghastly George Galloway in the Rochdale by-election saying "I am so honoured and grateful to all the people who voted for myself."

On the fertilised egg/Alabama situation, I'm sure I don't need to emphasise how much I agree with what's already been said. And hsh: that's a great thought experiment. I'd love to hear it put to those crazies with a news crew nearby to record their answer.

[Michael Cain]

Minor peeve today... Yesterday I got a letter from the Democratic Congressional Campaign Committee. In addition to asking for money -- what else? -- they included a couple of longish paragraphs about what they wanted to do with the money.

The first paragraph talked about funding local work, in particular at the precinct level. They went on about the importance of local precinct activities. But Colorado doesn't have precincts, and hasn't for years. We are a vote by mail state. Every registered voter gets a ballot by mail. If people insist on voting in person, or are one of the edge cases such registering on election day, there are one or more vote centers per county where they can vote. At any vote center in their county.

The second paragraph emphasized the need to fund winning Lauren Boebert's Third District seat and keeping Boebert out of the House. But weeks ago Boebert filed the paperwork to be a candidate for the open seat in the Fourth District. No one will know until the end of June who's running in the Third and whether or not Boebert is the candidate in the Fourth (so far, not looking good for her).

I have complained for years that the national Democratic organization takes the West for granted. They don't understand it, they don't want to have to pay attention to it, despite it becoming critical to their future. This letter, they're not even trying.

[JanieM]

Not to defend the DCCC too enthusiastically, since I wouldn't even have gotten so far as to read the letter, the mailing would have gone straight to to the trash .

In fairness, though, the Third District seat is still Boebert's, and she is still trying to get back to the House one way or another.And presumably the Ds would like to win the Third District seat AND beat her, wherever she runs (if at all).

I obviously have no idea whether the DCCC hacks were really so ignorant as not to know she's switching districts, but I wouldn't be surprised if they just wanted to use her name as (in effect) clickbait.

As to this: But Colorado doesn't have precincts, and hasn't for years. We are a vote by mail state.

Apparently the Colorday Democratic Party hasn't heard the news. Or rather, the fact that the state doesn't have precincts for voting purposes doesn't seem to have stopped the D party from continuing to use the name as the basis for local party organizing.

[wj]

Alabama's purpose is to make Texas look civilized...

Time was when Alabamans said "Well, at least we aren't [as bad as] Mississippi." But with Mississippi doing things like (finally!) expanding Medicaid, they may be reduced to "We aren't as bad as Tennessee."

Granted, they're trying hard. But Tennessee Republicans seem determined to win this race to the bottom.

[Girl from the North Country]

Oh God, this headline in the Guardian:

Joe Biden twice confused Gaza with Ukraine as he was announcing that the US would provide desperately needed aid to Palestinians.

I am on record as believing (and I still believe) that Biden in a coma would be a better president than Trump, but unfortunately (on this subject as on many others) not everyone agrees with me. This situation is getting hairier and hairier, and I cannot think of any solution.

[wj]

Well, there is one obvious solution. The Supreme Court could rule that misoprostol (an OTC abortion drug) is illegal. Followed by reversing, or at least taking a case for review next year to look at, Griswald (the case that established that contraception could not be made illegal). THAT would make the dust up over Dobbs look like a tempest in a teapot!

Definitely not wishing for such a thing. Not that my wishes have influence in this. But it would undoubtedly trash a huge number of Republican candidates, not only Trump and members of Congress, but also at the state level.

[JanieM]

Joe Biden twice confused Gaza with Ukraine

I really don't want to get sucked into this topic, so I won’t say anything beyond this comment.

But I would bet everything I have that Biden has never confused Gaza with Ukraine for a millisecond. He misspoke. He does that.

The article itself continues to pound away at the substitution of one word for another. It's like there was a news article announcing the end of the world and all I could obsess about was the typos. It’s a perfect example of how the press is gleefully and poisonously greasing the skids of our slide toward disaster.

[novakant]

I don't think the slide towards disaster can be blamed on the press (least of all the Guardian, lol). To be precise, it's not that parts of the press don't have a role in this, but it's secondary.

Instead, look at the poll numbers which are scary and ask yourself: what happened? Trump leads Biden by 5 POINTS!

https://www.nytimes.com/2024/03/02/us/politics/biden-trump-times-siena-poll.html

[wj]

Before your blood pressure gets too far off the charts, ask yourself a couple of questions. First, what is, according to those running the poll, the margin of error? If the margin of error were 1 percent (ha! Never happen) then yes, there would be a problem. On the other hand, if the margin of error is 5 percent, then as a matter of statistics, there is no diffetence between the two.

Second, over time, how much do the numbers bounce around? That is, how far off the trend line (even if the trend is flat) do numbers range? Note that this is different from the margin of error of a single poll.

Basically, you don't have a series of dots or a single line. You have a band which covers most, but still not all**, of the data "points" over time.

So, if the trend line is down, you may have a problem. If the trend line has been up, but turns down (which requires several reports to identify), you potentially have a problem.

And that all ignores the detail that, this far ahead of the election, polls are nothing resembling predictive. Get some polls in October, and they might say something relevant. This far out? No so much.

** I'll spare you all a long discussion of standard deviations. Suffice it to say that the "margin of error" doesn't mean the actual number is within that range to either side of the reported data. Rather, that the actual number has an xx% chance of being within that range. Typically, xx will be 68% (one standard deviation plus or minus) or 95% (two standard deviations).

[nous]

Another thing to consider about polls is that the pollsters acquire the lists of registered voters, or likely voters, or whatever, and then wardial their phone numbers or spam their email looking for people who will pick up or respond. I get several of these a week and I have never once responded. The only people who do respond, generally, are older (even than me) voters, and voters with a grievance and motivation.

If I were a Palestinian voter, I'd definitely consider responding just to register my outrage and send a message. But I'd also be well aware that sending a message in a poll was not the same thing as casting a vote for Trump or withholding an actual vote from Biden in November.

I know the byelection in Rochdale dials up the concern for something similar happening in the US, but I also think that choosing an MP is a smaller gesture than choosing a president, and I'm not sure that all of the people who are registering their grievance would follow through and flip when the result is a nationwide win for even worse, rather than just a single district win for an independent.

IOW, I think the polls are more volatile and fractious than will be the actual voting. At least I hope so.

[CharlesWT]

I'll spare you all a long discussion of standard deviations.

The two candidates are abnormal deviations.

[Tony P.]

Where do the poll respondents get their information, if not "the press"? Sorry, I meant "information".

And where does "the press" get the "information" it feeds to potential poll respondents? Could it be from a paranoid fear of not "covering" what "everybody is talking about"?

Could it be that "immigration" is as irrelevant to most Americans as "the top marginal tax rate" in what we laughingly call Real Life, but if "the press" harps on it enough it becomes a major factor in Americans' mental universe? Could it be that "the press" is determined, as a business, to stay in the good graces of the coming fascist regime by making sure it plays up Biden's alleged shortcomings at least as much as any mention of He. Trump's criminality, paranoia, and dementia?

"The press" is not what it used to be, of course. Americans get their "information" from "social media" at least as much as from TV or newspapers. And "social media" has not even the pretense of objectivity that "the press" used to simulate. The main function of "the press" these days is to lend its pretend-objectivity to "trends" and "memes".

I'm not saying the poll numbers are wrong. As I have often pointed out, it doesn't matter whether people get convinced by advertising, or lies, or sheer repetition. If you convince people that Biden is past it, they become, well, convinced. And when you are convinced of something, even real life has a hard time changing your mind.

There is a cohort of Americans for whom He, Trump is the Second Coming of Jesus. Literally. They have been told so by their "pastors" and not seen or heard "the press" say "are you nuts, or what?" Maybe that's because this cohort is too small to bother with. Or maybe there's a different reason.

There is a cohort of Americans who polsters label "double haters" -- they hate He, Trump and Joe Biden both. I'm not sure where they get their "information" from. They can't all be Libertarians(TM), but they surely don't overlap with Second Coming MAGAts.

Then of course there's the cohort that considers itself "independent". Maybe it includes the double-haters; I don't know. I do know these people will vote Republican given the slightest excuse to do so. It's people like that who the GOP superPAC formerly known as the Supreme Court caters to, by pretending that He, Trump's immunity claims require serious deliberation.

Oh well, a government of the people, by the corporations, for the rich is what most Americans seem to want. Serve'em right if they get it.

--TP

[wj]

Then of course there's the cohort that considers itself "independent". Maybe it includes the double-haters; I don't know. I do know these people will vote Republican given the slightest excuse to do so.

I must take issue with this. I know a fair number of "independents" who mostly vote for Democrats due to how crazy the GOP has gotten. Which is probably typical here, seeing as the number of independents is on the same order as Democrats in California. And yet, pretty much none of them vote for Republicans, which is why this is effectively a solid blue state.

[CharlesWT]

It's people like that who the GOP superPAC formerly known as the Supreme Court caters to, by pretending that He, Trump's immunity claims require serious deliberation.

I predict that a majority, maybe even 9-0, will rule against Trump.

[Tony P.]

wj,
Glad to hear about the independents in CA. It's "independents" in place like WI, MI, PA, GA, NC, AZ, that I worry about.

CharlesWT,
The ruling is irrelevant. The point is the delay. You may be right about how they'll rule, if for no other reason than that Biden gets whatever "presidential immunity" they hand out.

--TP

[JanieM]

The ruling is irrelevant. The point is the delay. You may be right about how they'll rule, if for no other reason than that Biden gets whatever "presidential immunity" they hand out.

Or else they'll delay until after the election, and decide accordingly. Kind of like McConnell's rules: one set if a D is president (Obama/Garland), another if it's an R (Clickbait/Coney Barrett).

[CharlesWT]

The ruling is irrelevant. The point is the delay.

They didn't have to take it up this term. They could have said we'll look into it next term.

[Hartmut]

They didn't have to take it up this term. They could have said we'll look into it next term.

But that would have been bad optics.
If they just delay, the blame can be shifted to the lower courts, if they don't finish before the election. If they took it up in their next term, the result would be irrelevant since they would decide in the negative, if a Dem wins and a GOPster would simply ignore any decision to his disadvantage.

If this was about disenfranchisement, it would be a completely different case, of course since then the point would be to NOT decide while it could have an effect (since the very purpose is to effectively disenfrachise while pretending not to).

[wj]

Glad to hear about the independents in CA. It's "independents" in place like WI, MI, PA, GA, NC, AZ, that I worry about.

To be concerned is one thing. (Although I'd say that it is, at minimum, not proven that California is unique in this regard.) But your original comment read like you considered it an absolute certainty that the "independents" would all vote R.

[novakant]

All these reasons to be cautious about polls are fair enough, but they also have an air of whistling in the dark - I only want to prevent a repeat of Hillary 2016 and the Democratic Party can be remarkably tone deaf.

One thing Biden could do to win over young voters, anti-war voters and people with some connection or other to the ME is to finally tell Netanyahu what's what and back it up by decisive action.

It's amazing how badly this crisis has been handled by the Biden administration (just like Afghanistan, but it's of course much worse) and how they think they can just carry on reagrdless, paying a bit of lip service here and there.

[novakant]

I wouldn't dismiss people like e.g. this

https://www.nytimes.com/2024/02/26/us/politics/israel-gaza-biden-michigan-democrats.html

as "double haters". There are actual, seemingly insurmountable policy differences here.

[novakant]

This is interesting:

What Biden Is Thinking About the 2024 Election

https://www.newyorker.com/podcast/the-new-yorker-radio-hour/what-biden-is-thinking-about-the-2024-election

[wonkie]

What Novakant said except I'm more worried about the parallel to Humphrey's defeat to Nixon with Gaza as the war.

[JanieM]

As a side note (because as GftNC observed above, this does seem to be an open thread), this made me think of Michael's occasional comments on the power grid, which is where I learned that the Eastern and Western grids even exist as such:

Texas is unique in the United States for largely having its own power grid with minimal connections to either the Eastern or Western grids, the other large U.S. systems.

From this article, which I ran across somewhere...maybe even here?

[wj]

What Novakant said except I'm more worried about the parallel to Humphrey's defeat to Nixon with Gaza as the war.

The difference, and it is a massive difference, is that no American soldiers are fighting and dying in Gaza.

[wj]

One thing Biden could do to win over young voters, anti-war voters and people with some connection or other to the ME is to finally tell Netanyahu what's what and back it up by decisive action.

I hope (no doubt because I'm a compulsive optimist) that what we are seeing, or rather not seeing, is Biden doing the necessary (and necessarily out of view) background work. So that, when he does act, it's something that will stick. (It is my perception that, no matter what we say or do, Bibi will stand firm. Because he sees the alternative to continuing as he is will land him in jail. And he's not wrong.)

"Decisive action" is indeed what is required. But to make a gesture, and then have it go nowhere and fall apart, would be IMHO worse than doing nothing. Both for the Palestinians and for his own electoral prospects. However frustrating the wait is for those who are outside, looking in thru a hpglass darkly.

[Hartmut]

Let's see how Bibi reacts to supplies airdropped into Gaza. Will the Israeli Air Force be ordered to prevent* that or the ground forces to either seize them or to shell the area to make them inaccessible?

*not yet time to simply shoot them down but forced turning back or forced landing seems not impossible to me. If this was the hot phase of the US election campaign already, all bets would be off but this is still too early.

[JanieM]

Instead, look at the poll numbers which are scary and ask yourself: what happened? Trump leads Biden by 5 POINTS!

Instead, look at the poll and roll your eyes.

Sample of the analysis, which the author quotes from some other analyst:

I do not believe Biden is tied with women nationally 46-46… Biden got 57% of women in 2020. You're telling me that, post-Dobbs, his support among that demo group will drop to 46? Not credible.

Then there's this:

The news in this poll was partway decent for Biden when it came to young voters age 18-29. He leads Trump by 13 points among them, 54 to 41 percent—but that’s still around half the spread that other major polls have on this age group. But when it comes to messaging on the youth vote, the NYT prefers to emphasize the negatives, and its own data seems at odds with itself.

For example, as former pollster and turned sometime polling industry critic Adam Carlson notes, the NYT/Siena poll of swing states conducted back in late October showed Trump actually leading in this age group in AZ and GA, while being tied with Biden in MI. On this contrary, surprising and incorrect result, Nate Cohn of the Times did a whole serious write up about what it could mean.

But when the new poll shows Biden actually leading nationally within this group by 14 points, Carlson observes, the NYT analysis completely ignored this.

Perhaps that’s because it is hard to square that earlier result of Trump leading with this very different one of Trump trailing without calling one or both into serious question. Such huge swings in such a short amount of time don’t suggest that the electorate is moving quickly so much as that the polling might be way off.

[GftNC]

I have to say, I agreed with novakant on the possibility that the Guardian was in any way to blame for overemphasising Biden's weaknesses.

However, luckily today:

https://www.theguardian.com/us-news/2024/mar/03/trump-confuses-obama-biden-virginia-rally

I was particularly pleased that

Richmond crowd reportedly went silent as 77-year-old mixed up the president and ex-president for third time in past six months

[Tony P.]

novakant: I wouldn't dismiss people like this as "double haters".

So, they only hate Biden and NOT Bibi's buddy He, Trump?

The Palestinians have had a raw deal for a long time, but I can't imagine how the Orange Jesus could possibly make it better for them. I think Bibi is counting on his doppelganger regaining the White House, and pees himself with joy at the thought that Americans of Arab heritage might make it possible.

--TP

[novakant]

Below is a gift link to the article wihtout the paywall.

She's under no illusions about Trump

“The other person is not going to be any better,” she said, refusing to say Mr. Trump’s name.

she has tried to change things from within the Democratic party for a long time but nothing worked:

“You want my vote? You cannot kill my people in my name. As simple as that”

https://www.nytimes.com/2024/02/26/us/politics/israel-gaza-biden-michigan-democrats.html?unlocked_article_code=1.Z00.ZZ6S.rGg92jdJoib3&smid=url-share

[nous]

novakant - Yes, there are people like that. Yes, Biden may lose her vote. Yes, Biden's support of Israel makes the situation in Gaza worse. All those things need to be considered when looking at possible election outcomes.

I'm just pointing out that the nature of these polls, and the reporting of these polls, stresses the more worrisome possibilities, and tends to underrepresent the number of people who are upset about the above, but who have resigned themselves to a lack of practical alternatives.

The way these things get presented is to break the population into segments and then to treat the individual segments as binary switches: Biden has turned off the Arab-American switches with his pandering to Israel.

But the elections are multilateral negotiations, and complex. Shoring up votes in one area will have an unpredictable effect on other areas, and the other side of the contest will do their best to exploit that shift by moving quickly to gather up the new group of disaffected voters. And when they do, the media will be right there to second guess the change in heart and mine the uncertainty for more clicks.

Furthermore, from a rhetorical vantage point, I know that both the DNC and the Progressive Caucus are going to be trying to play both ends against the middle in their messaging as well. They will be arguing one side in the press (to lend their side strength in the public narrative) while sending off alarming, hot button emails to their entire mailing list of people who have ever donated to one of their candidates, trying to convince their supporters that money must be sent right now to prevent an electoral apocalypse. They too need that worry to drive clicks and revenue, and they can't afford to let anyone feel complacent.

But I have only so many spoons.

Effectively, I have zero influence over any of this. Nothing I can do will alter this in the big picture. I have a tiny amount of influence over the conversation here, among my friends on social media, and in my classroom discussion. Come what may, my tiny course of action in this is predetermined, so I might as well just focus on the things that I can affect, and worry about those.

[Girl from the North Country]

Rather fascinating story in today's Times about Sefton Delmer's radio campaign targeting the Nazis in WW2:

What a Nazi-baiting radio genius can teach us about today’s infowars

British propagandist Sefton Delmer’s remarkable ability to tap into the psyche of wartime Germans has lessons for our world of personality cults and conspiracy theories

Summer 1941. Hitler and his allies rule Europe from the Atlantic to the Black Sea. In the West, Britain is isolated. In Germany, the streets of the Third Reich are covered in a forest of swastika flags, the red and black illuminated at night with the glow of torch-lit processions.

The Führer’s speeches were festive celebrations and every good German was expected to listen. Wardens wearing swastika armbands patrolled the streets, ensuring everyone was within hearing distance of a radio or loudspeaker.

Germany had the most radios per capita in Europe, helping Nazis fuse the Volk. But radio was also a vulnerability — twiddle a dial and you could get alternative content too, which was why tuning into the BBC became a crime punishable by hanging. It was more difficult, however, to smear internal dissent that appeared to come from a loyal German.

In July 1941, the SS noted that Germans were tuning into a new station, hosted by an angry Wehrmacht officer known simply as “Der Chef”, who mixed anti-regime rants with coded instructions, so the audience had the sense they were eavesdropping on secret soldierly chatter. Der Chef swore incessantly and used racial slurs, calling Churchill a “dirty, Jew-loving drunk” and complaining the Nazi party was full of secret Bolsheviks.

Der Chef knew the shops where Nazi functionaries bought their secret stashes of luxury pâtés and Moselle wines. He even knew which Nazi official’s wife had just bought a second flat in Vienna thanks to her corrupt connections.

Such details, all the listeners understood, could come only from insider leaks. This meant Der Chef must have some sort of cover from his higher-ups. There had always been tension between the Prussian military aristocracy and the Nazi party. Could there be some sort of rebellion brewing? And where was Der Chef broadcasting from? Occupied France? A barge on the Rhine?

Even the US embassy in Berlin was tuning in to these broadcasts, and its service attaché saw Der Chef as a sign that there was an increase in anti-Nazi sentiment in the army. The Americans shared their insights with the British.

This was one of the more awkward moments in British-American Second World War relations. David Bowes-Lyon, the Queen Consort’s brother, had to travel to the White House to explain to President Roosevelt that Der Chef was actually in Bedfordshire, and part of a British covert propaganda operation.

The show was being recorded in the billiards room of a red-brick house near the village of Aspley Guise. Der Chef was not a renegade Nazi. He was being played by an anti-Nazi German novelist of Jewish descent. At the back of the billiards room stood a vast, bearded man inventing and directing this whole performance: Sefton Delmer. And in the summer of 1941, Delmer’s empire of tricks was just getting going. He would rise to become the head of special operations for the Political Warfare Executive, running a string of covert radio stations in a multitude of languages.

Though now largely forgotten, Delmer was considered a genius by his peers, with the remarkable ability to tap into his audiences’ psyches. Today, as we struggle to reach audiences across the globe who dwell in a universe of cults of personality, conspiracy theories and “alternative facts”, Delmer’s wartime experience has much to teach us.

A talent for trickery
Delmer had scraped his way into Britain’s secret propaganda war. The son of Australian-British academics, he had grown up in Berlin during the First World War. After Oxford, he went back to Germany and became a reporter for the Daily Express. He spoke fluent German and had a talent for performing any role that would get him a story, even impersonating being the adjutant of the stormtrooper boss Ernst Röhm to penetrate a closed gathering of brownshirts.

Because of his German connections, however, Delmer repeatedly failed to get clearance for secret work during the first year of the war. In mid-1941, however, with alarm growing in Whitehall that Germany was winning the information war, Delmer was finally given clearance and asked if he could try something similarly devious against the Nazis. “No holds are barred,” he was instructed.

Der Chef was his first experiment, and Delmer used it to express the repressed rage of German soldiers. He tore into SS men who were avoiding the toughest fighting: “These men are wearing the German uniform, but they drag the German name in the mud. They should send them to the east. All the pigs of the Waffen SS should be sent there. There everything else but their penis would get stiff.”

The more prudish members of Churchill’s cabinet were appalled by the foul language and frequent pornography, and Der Chef was nearly shut down. Sir Stafford Cripps, the UK’s teetotal ambassador to Moscow, was appalled by the detailed description of a named German admiral who took home four or five sailors and with his own mistress had an orgy involving butter and helmets.

But Delmer survived the criticism: the programmes were successful, and even in this example you can see that the real aim was to show how butter that was rationed for ordinary people was available for elite orgies. Instead of being canned, Delmer’s taboo-breaking broadcasts caught the ear of a naval intelligence officer at the Admiralty, Ian Fleming, who was looking for a way to undermine morale in the German armed forces. With Fleming’s backing, Delmer expanded his operations.

Cover for subversion

In October 1943, the SS Obergruppenführer of Munich was furious. He wrote to his superiors about a new radio station that had “caused the greatest unrest and confusion among the population”. Goebbels agreed. “The so-called Soldatensender Calais,” he wrote in his diary on November 28, 1943, “gave us something to worry about.”

The Soldatensender, Delmer’s latest wheeze, sounded like a regular Nazi station and broke all the latest communiqués from Nazi leadership. The rest of the content, however, was precisely the sort of stuff the Nazis always tried to censor, including incredible detail about the lives of soldiers and sailors and the grim realities of disease, faulty weapons and losses.

Though patriotic in style, this content made the Nazi leadership look terrible. It encouraged civilians to lose faith in the leadership, soldiers to surrender and Nazi officials to become paranoid about who was leaking their secrets.

The real genius of Soldatensender was that, unlike Der Chef, anyone listening in was meant to understand it was actually a British station. Der Chef had broadcast on the fringe short wave, which was full of pirate stations. Soldatensender was on the respectable medium wave, which only state-backed stations had the power to access. Yet much of the content was too subversive to be an actual Nazi station.

Audiences were being given the “cover” they needed, if they were caught listening, to pretend it was an official German broadcaster. It gave psychological cover too: Delmer knew most people wanted to remain loyal Germans, even as they wanted a way to express their secret frustrations and be given a “patriotic excuse” to disobey orders.

The station was full of music that personified this camouflaged approach: it played the sort of American jazz the Nazis banned but soldiers loved — but without the English words. Delmer had his own big band full of cabaret stars exiled from Berlin, many of them Jewish.

Undermining dictators

Back in Bedfordshire, Delmer was overseeing a large and complex multinational operation, with editorial writers and speakers from Italy, Hungary and Romania. “Our canteen became a tower of Babel, as dark-eyed gypsy beauties from the Balkans flirted with my fair-haired German prisoners over powdered egg omelettes, spam fritters and other irresistible delicacies from the repertory of war-time cooks,” he recalled. His team gathered stories by intercepting letters and planting microphones in the cells of German prisoners of war, to pick up the latest gossip from the front, right down to which Calais prostitutes were spreading which diseases.

How effective was this elaborate operation? Delmer always avoided directly answering this question — his aim was generally to erode German loyalty. But we do know that when captured German soldiers were surveyed, nearly half (41 per cent) said they tuned into the broadcasts. The leaders of the Valkyrie rebellion against Hitler were also encouraged by the broadcasts, thinking they showed the British would back an army rebellion.

Unlike the Eastern Front, where Germans fought to the end, on the Western Front — where Delmer’s media was directed — the army surrendered in droves. There were many reasons for this, of course, including the lack of fear of Red Army atrocities that kept soldiers fighting in the east. But Delmer’s operation played a role too.

The essence of Delmer’s propaganda project was about giving people the excuse to do what they secretly desired. He believed preaching abstract ideas about the virtues of democracy was useless to people under the sway of a Hitler. Instead you needed to tap into the emotions and interests that people cared about even more than the cruel ecstasies offered by the Nazis. You needed to engage their resentment and outrage, break their fear, arouse individualism and desire to make decisions that benefited themselves and not their leaders.

There are lessons to be learnt today from Delmer’s insights and innovations. As in the 1930s, dictators in Moscow, Beijing or Tehran are far ahead of us in developing their information warfare capabilities. We need to compete, though not by peddling lies and disinformation as they do — and like Goebbels did.

In Delmer’s best work, the audience was in on the act — they knew the British were probably behind the Soldatensender. Delmer created a masquerade to express the repressed emotions and information the Nazis didn’t want you to know or feel. To smuggle truth into the realm of lies, you need to have a disguise. Delmer may never have preached lectures about democracy, but so many of his tricks were designed to pull you free.

How to Win an Information War by Peter Pomerantsev is published by Faber, £20

[Snarki, child of Loki]

THIS is why I'm hoping that there's British working to convince Putin that there's a secret Ukrainian anti-Russian HQ hidden somewhere in Moscow.

They're tricksy that way. STILL have most people convinced that eating carrots is good for eyesight, 80 years after it was used as cover for early radar!

[Girl from the North Country]

Hmm, if only. I'm afraid our glory days of such brilliance are far behind us. Although, in Iran, they still credit us with astonishing tricksiness. This was from a piece in Prospect in 2008, but I was told not all that long ago that there was still a substantial hangover of these kind of views in Tehran. Of course, I have no idea if they survived BoJo, although I suppose one explanation for the Bojo phenomenon might have been that it couldn't possibly be what it seemed to be:

Many Iranians ascribe to the British a combination of Machiavellian cunning, Metternichian realism and mystical omniscience. Far from being America's servant, Britain is often credited with duping the Americans and getting them to do its bidding.

[GftNC]

https://www.prospectmagazine.co.uk/essays/52666/blame-it-on-the-brits#:~:text=Many%20Iranians%20ascribe%20to%20the,them%20to%20do%20its%20bidding.

[Michael Cain]

Texas is unique in the United States for largely having its own power grid with minimal connections to either the Eastern or Western grids, the other large U.S. systems.

Texas's decision to be independent was based on (1) they had enough natural gas to pull it off, (2) their demand centers were isolated enough, and (3) they have always had an aversion to federal "interference" in Texas.

California could probably have done the same thing. But by 1970 they were putting a lot of pressure on themselves to clean up their air. So they built Path 65 to bring surplus hydro power from the BPA, and a very large coal-fired plant in Utah* plus Path 27. The TransWest Express HVDC line is under construction to carry large amounts of wind power from Wyoming to the edge of the Phoenix/Las Vegas/Southern California triangle. The SunZia transmission line will bring wind power from the downslope side of the Rockies in New Mexico to that same demand triangle.

I mention those western transmission projects because there are proposals to require ERCOT to build 4.5-6.0 GW of transmission connection to the Eastern Interconnect. What those proposals don't do is answer the question of where ERCOT will find 4.5-6.0 GW of idle generating capacity in the Eastern during events like Feb, 2021.

[novakant]

That's a great article, GftNC, thanks - the author knows what he's talking about, lol.

[novakant]

On a more serious note, this history and associated diplomatic tussles make life for Iranian expats unneccessarily hard, as if the situation wasn't difficult enough already.