Injecting Facts into a Debate

by wj 

You may have seen, as I have, arguments back and forth concerning police vs mental health approaches to combatting violent crime in our cities.  Just for sheer novelty, here is a little something that may be of interest.  The results of a city (Richmond, CA) which actually tried it. 

Richmond, 2006: A man is shot in the face at the funeral of a teenager who was gunned down. The City Council is on the verge of declaring a state of emergency. Comparisons to Iraq are drawn by a state senator. The year sees 42 homicides and will go on to see 47 more in 2007.

Richmond, 2023: Eight homicides, the city’s lowest number since they started keeping record.

The city was home to one of the nation’s worst homicide rates. What happened?

Richmond implemented Advance Peace in 2007, a program that treats urban gun violence as a public health crisis. Its model consists of an intensive strategy called the Peacemaker Fellowship, an 18-month mentoring intervention for young people involved in gun violence.

Oakland’s Ceasefire program also focuses on high-risk individuals, but Advance Peace goes one step further to create pathways to healing from trauma.

The backbone of the Richmond program are the community-based violence intervention workers, people from the community with the same lived experience as the youth they work with. They mediate conflict in the neighborhood in real-time and — here is where the public health approach comes in — are trained in trauma-informed care. They don’t just stop youth from committing violent acts but actually help adolescents at the margins of society to understand their trauma — and to heal.

Just 10 years after implementation, Richmond saw its lowest number of cyclical and retaliatory firearm assaults and homicides in more than four decades. Some could argue that this is due to other anti-crime initiatives. However, no other city experienced a similar drop despite the existence of those same programs. One major difference is Advance Peace.

Sacramento, Fresno and Stockton have experimented with Advance Peace, but none has committed to this proven route to violence reduction.  Sacramento’s gun violence was reduced by 22% after its first 18 months in 2019 and by 39% in Del Paso Heights, an area known for prevalent gun violence. Sacramento also went two full years without suffering a single youth homicide.

Despite these outcomes, the city of Sacramento ceased funding Advance Peace in December 2021, thus experiencing a significant uptick in gun violence in 2022 and 2023.

Stockton gun violence was reduced by 21% over its first two years of implementing the Peacemaker Fellowship and by 47% in Council District 1. Stockton and Fresno still have Advance Peace operations, but their local governments do not give the program full funding support.

Let’s call on these cities to follow these strategies that work and institutionalize funding to Advance Peace. The funding is now independent in these cities, mostly from the state and one-time grants. One-time funding will not give the same results as ongoing consistent funding.

Richmond’s standout success lies in institutionalizing Advance Peace into the city budget like any other public safety program. Advance Peace is actually housed in a government agency called the Office of Neighborhood Safety — the nation’s first city office dedicated exclusively to gun violence prevention. Richmond is the model of what other cities could accomplish by putting sustainable funding into policy.

This does not mean replacing police but a complimentary expansion that also offsets costs. Police investigations, emergency services, court time and other government services lead each shooting to cost Stockton taxpayers $962,000 and each gun homicide $2.5 million. Investing in Advance Peace can disrupt these costs. For every dollar spent on the program, the public received $47-$123 in return in Stockton and $18-$41 in Sacramento.

Imagine the change if these cities gave real support. This would be a huge step in building the sustainable and community-based public safety infrastructure that we know works and that California cities need. 

I realize it can be considered poor form to work from data, rather than from claimed first principles.  But it does seem interesting. 

Open Thread as well, of course, since we are overdue.

Comments

[wj]

Apologies for failing to include the source. It was the East Bay Times:
https://www.eastbaytimes.com/2024/03/13/opinion-richmonds-standout-strategy-in-fight-to-reduce-gun-violence/

[JanieM]

This does not mean replacing police but a complimentary expansion that also offsets costs. Police investigations, emergency services, court time and other government services lead each shooting to cost Stockton taxpayers $962,000 and each gun homicide $2.5 million. Investing in Advance Peace can disrupt these costs.

I wonder how much the gun lobby is investing in sabotaging these programs......

[Hartmut]

I wonder how much the gun lobby is investing in sabotaging these programs......

My thoughts too. But then I had to cynically conclude that the total amount of money that needs investing there is by now rather low after the decades long installation of series of amplifiers that fire up at hair trigger sensitivity. So the money can be diverted to the creature comforts of the lobbyists without diminishing their effect.

[Hartmut]

Since this is an open thread: I somehow got this idea yesterday and would be interested, if any math freak around here could answer it:

The entry level punishment for mathematicians in hell is to solve the complete set of the satanic polynomial, i.e. to find the zeroes of all polynomials of the 666th degree with the coefficients 0 to 666 in all possible permutations (the second level will be the same with any coefficient as either + or -). As a preparation: Is there any permutation where all zeroes are integers? If not, is there any where all zeroes are rational? (Of course it's easy to find permutations where there is no zero in real numbers). For the actual math freaks around here: Is this actually a trivial question (at least for the integers)?

My guess is that there is none for the integers since the x° term would have to be divisible by all (666) zeroes but can itself be 666 at max.

[CharlesWT]

I somehow got this idea yesterday and would be interested, if any math freak around here could answer it

Since no math freaks have proffered an answer.

"Your intuition is correct. For a polynomial of degree 666 with integer coefficients, it is impossible for all the roots (zeros) to be integers.

This is a consequence of the Rational Root Theorem, which states that if a polynomial equation with integer coefficients has a rational root p/q (where p and q are coprime integers), then p must be a factor of the constant term, and q must be a factor of the leading coefficient.

[...]"
No Integer Zeros

[Hartmut]

Is that sufficient?
There could in theory be multiple identical roots. Since one coefficient has to be 0 (the polynomial has one more coefficient than its degree, so the count runs from 0 to 666) there has to be at least one pair of conjugated roots.
I still guess that there is no solution with integers only but I am not sure that the theorem alone is enough to proof it.
And even if this is the case, is a rational-roots-only solution also impossible?

I think that real-roots-only solutions exist in high numbers as well as complex-only solutions (since it is an even degree polynomial. In odd degree polynomials there must be at least one non-complex root since complex ones always occur in conjugated pairs, if all coefficents are real).

[CharlesWT]

"You make an excellent point. The Rational Root Theorem alone is not sufficient to definitively prove or disprove the existence of integer or rational root solutions for a polynomial of degree 666 with integer coefficients.

You are correct that the possibility of multiple identical roots (roots with multiplicity greater than 1) needs to be considered. The Rational Root Theorem only gives information about the possible values of distinct rational roots but does not preclude the existence of repeated roots.

[...]"
No Integer Zeros - 2

[bobbyp]

wj,
Thank you. There actually is such a thing as good public policy. Imagine that.

[wj]

bobbyp,
More to the point, there's such a thing as good public policy actually being implemented. Which is even more uncommon. :-)

[Hartmut]

Btw,
666!=2^621*3^330*5^165*7^109*11^65*13^54*17^41*19^36*23^30*29^22*31^21*37^18*41^16*43^15*
47^14*53^12*59^11*61^10*67^9*71^9*73^9*79^8*83^8*89^7*97^6*101^6*103^6*107^6*109^6*113^5*
127^5*131^5*137^4*139^4*149^4*151^4*157^4*163^4*167^3*173^3*179^3*181^3*191^3*193^3*197^3*
199^3*211^3*223^2*227^2*229^2*233^2*239^2*241^2*251^2*257^2*263^2*269^2*271^2*277^2*281^2*
283^2*293^2*307^2*311^2*313^2*317^2*331^2*337*347*349*353*359*367*373*379*383*389*397*401*
409*419*421*431*433*439*443*449*457*461*463*467*479*487*491*499*503*509*521*523*541*547*
557*563*569*571*577*587*593*599*601*607*613*617*619*631*641*643*647*653*659*661
The total number of satanic polynomials is 666*666! since there are 667 coefficients but one is 0 and cannot stand in front of x^666 without reducing the degree of the polynomial.

Still waiting for a proof that none of them has only integer roots or whether at least one has only rational roots.

[Hartmut]

Slight correction: The above are the pure positive satanic polynomials. If every coeffecient can be either + or - one has to multiply it with 2^666:

So, 666*666!*2^666 satanic polynomials in total.
Nice warming up problem for the eternally condemnend mathematician to solve them all.

[CharlesWT]

"You've provided an excellent analysis of the problem and the sheer magnitude of the number of cases to consider. Let me try to provide a more comprehensive proof regarding the existence (or non-existence) of integer or rational root solutions for polynomials of degree 666 with integer coefficients between 0 and 666.

Proof that no polynomial of degree 666 with integer coefficients between 0 and 666 can have all roots as integers:

Let P(x) be a polynomial of degree 666 with integer coefficients between 0 and 666.

If P(x) has all roots as integers, then by the Rational Root Theorem, all roots must be factors of the constant term (the coefficient of x^0).
[...]"
No Integer Zeros - 3

[Pro Bono]

If there's a permutation where all the roots are integers, almost all of the roots must be 1 or -1, since the product of the roots must be 666 or less.

I haven't worked out how to read the AI 'proof'. I expect it's nonsense.

[Hartmut]

666=2*3*3*37. That's rather few possible divisors indeed: +/-(1,2,3,6,9,18,37,74,111,222,333,666).
That's 24. If the x^0 term has the coefficient 0, then 0 is the 25th possible one. But that would only shift the same problem one step further up the line.
So it would probably be rational to look for rational solutions instead. ;-)

[CharlesWT]

You make an excellent observation. If there exists a permutation of a polynomial of degree 666 with integer coefficients between 0 and 666 that has all its roots as integers, then almost all of those roots must be either 1 or -1. Here's why:

Let P(x) be a polynomial of degree 666 with integer coefficients between 0 and 666, and let its roots be r1, r2, r3, ..., r666.

By the Fundamental Theorem of Algebra, we know that:
P(x) = a(x - r1)(x - r2)(x - r3)...(x - r666)

Where a is the leading coefficient of P(x).

Since the coefficients of P(x) are integers between 0 and 666, the leading coefficient a must be either 1 or -1.
[...]"
No Integer Zeros - 4

[CharlesWT]

"You're absolutely right! Your analysis of the divisors of 666 provides a strong argument against the possibility of having all 666 roots as integers for a polynomial of degree 666 with integer coefficients between 0 and 666.

Since 666 = 2 × 3 × 3 × 37, the only possible divisors of 666 are:

±1, ±2, ±3, ±6, ±9, ±18, ±37, ±74, ±111, ±222, ±333, ±666

This gives us a total of 24 possible integer divisors, and if we include 0 as a possible constant term, that's 25 possible values.
[...]"
No Integer Zeros - 5

[Pro Bono]

What's the point of this AI chaff?

[wj]

Hartmut incautiously raised a math question. And Charles ran amok.

[Donald]

I don’t know if you can read this without a Twitter account, but here is an Israeli academic’s argument that Israel is committing genocide. ( I am comfortable with the word now— I think the intent to destroy in part is pretty clearly met by the obstruction of food supplies.)

https://threadreaderapp.com/thread/1768844679439651111.html?utm_campaign=topunroll

And Biden is complicit. He has been arming them all along and so far refuses to stop.

[Girl from the North Country]

I am comfortable with the word now— I think the intent to destroy in part is pretty clearly met by the obstruction of food supplies.

Donald, I wouldn't say I'm exactly comfortable with it yet, but I'm definitely moving that way, and for the same reason.

[wonkie]

Yes, I think the intent is genocide. Netanyahu might not believe that he can literally kill all of the people of Gaza, but he seems, at least in my mind, to be killing as many as he can before world opinion finally stops him.

[Girl from the North Country]

wonkie, you've thought it was genocide from the beginning.

As for Netanyahu, I think he is a lot like Trump; I doubt he has any particularly murderous feelings towards the Palestinians, or Gazans, (this is not to excuse him, by the way - if I am right then his obliviousness to their humanity makes him some kind of psychopath), I think he is desperate to stay in power and perfectly happy to do whatever he has to to do so, and at the moment that means placating his abominable racist rightwing allies, who in my opinion DO have murderous feelings towards Palestinians and Gazans, Arabs in general, and no doubt many other groups who do not share their ideology.

[wj]

GftNC, I'm mostly on the same page. Except that I doubt Bibi has any particular reluctance about genocide. Perhaps not enough enthusiasm to push it on his own. But no interest, even absent his need to stay in power, in restraining it.

[Donald]

I think there was genocidal intent among some Israelis and Palestinians from the beginning ( and I mean in the leadership of Hamas and Israel) but intent by itself wouldn’t be enough for me to call it genocide. Presumably a lot of wars involve people on one or both sides with genocidal feelings. They don’t necessarily get to act on them for various reasons. But for me the famine crosses the line into the actual crime. I am somewhat hopeful that the Western governments will find this sufficiently disastrous on the PR front that they will force Israel to let more food in, so the famine is stopped. Maybe. Cynical as I am about Western foreign policy, it is hard to imagine they would just let this go on.

[CharlesWT]

This is the best obituary I've ever read.

Kenneth Pluhar Obituary

[Snarki, child of Loki]

Well, there's ONE obituary that many of us will be glad to see.

Maybe even show up at the funeral to make sure TFG is most sincerely dead.

I suspect they won't let in people carrying cattle-prods "to make sure", though.

[Pro Bono]

These's an element of the Jewish psyche which says "we will never again be helpless victims". Israel's response to the 7th October attacks should be understood in this light: there's no genocidal intent as such towards Palestinians, just a determination to extirpate Hamas with nothing else mattering.

I don't say this justifies what Israel is doing, but perhaps it goes some way to explaining it.

[wj]

there's no genocidal intent as such towards Palestinians, just a determination to extirpate Hamas with nothing else mattering.

I will accept the possibility that, for many (even most) Israelis, there is no genocidal intent. But I don't think you can say that for the settlers, or for their members in the government. They might be willing to settle for every single Palestinian leaving the area, never to return. But I'm guessing not. Extermination is exactly what they desire.

[nous]

There's a clear pattern of what trauma counselors would call moral injury that is complicating the response to 7th October on the Israeli side. It's not unusual to see survivors of moral injury commit atrocities in the wake of the injury.

Not at all an excuse or a justification, just an assessment of the situation.

Moral injury is, in many ways, a narrative injury. It's the world violating the script that the injured party believed governed that world and gave it a sense of order and rationale. They are mobilizing some of their more bloodthirsty religio-nationalist myths to try to repair that damage to their sense of the world.

[Ufficio]

Note that the satanic polynomial p(x) (with all nonnegative coefficients and c_0 != 0 for the sake of simplicity) can't have any nonnegative roots. If all roots were integers, we'd have a factorization p(x) = a(x+r_1)...(x+r_666) with a and all r_i positive integers. The x^2 term in that expansion would be the sum of C(666, 2) terms, each of the form bx^2, where b >= 1. Thus c_2 >= C(666, 2), which is a tad out of bounds. Hence no such factorization exists.

If c_0 = 0, apply a similar argument to the x^3 term.

[CharlesWT]

No Integer Zeros - 6

[JanieM]

@nous: or, as someone put it in a workshop I went to long ago, "hurt people hurt people." There was a time when I was naive enough to think "we" (humans) could learn to break that cycle, but I have lost that optimism, at least in relation to any scale that would matter in a situation like Palestine/Israel. So I go back to a suggestion made by nous and other people: do what you can....

*****

From the Wikipedia page on genocide:

Deliberately inflicting on the group conditions of life calculated to bring about its physical destruction Article II(c)

The third prohibited act is distinguished from the genocidal act of killing because the deaths are not immediate (or may not even come to pass), but rather create circumstances that do not support prolonged life.[10] Due to the longer period of time before the actual destruction would be achieved, the ICTR [International Criminal Tribunal for Rwanda] held that courts must consider the duration of time the conditions are imposed as an element of the act.[42] In the 19th century the United States federal government supported the extermination of bison, which Native Americans in the Great Plains relied on as a source of food. This was done for various reasons, primarily to pressure them onto reservations during times of conflict. Some genocide experts describe this as an example of genocide that involves removing the means of survival.[41]

The ICTR provided guidance into what constitutes a violation of the third act. In Akayesu, it identified "subjecting a group of people to a subsistence diet, systematic expulsion from homes and the reduction of essential medical services below minimum requirement"[43] as rising to genocide. In Kayishema and Ruzindana, it extended the list to include: "lack of proper housing, clothing, hygiene and medical care or excessive work or physical exertion" among the conditions.[42] It further noted that, in addition to deprivation of necessary resources, rape could also fit within this prohibited act.[citation needed] In August 2023, founding chief prosecutor of the International Criminal Court (ICC) Luis Moreno Ocampo published a report presenting evidence that Azerbaijan was committing genocide against the ethnic Armenians of Artsakh Nagorno-Karabakh under Article II(c) of the Genocide Convention by placing their historic land under a comprehensive blockade, cutting all access to food, medical supplies, electricity, gas, internet, and stopping all movement of people to and from Armenia.[44]

[wonkie]

I know that word "genocide" has a formal legal definition, but I don't know what that definition is. My opinion from the beginning has been that Netanyahu is intent on killing as many people as he can directly through military action or by indirect means through lack of access to the necessities of life. I also believe that, regardless of any changes in rhetoric, he will continue in that effort unless stopped by an outside force such as a refusal by the US to give him weapons. I think that the goal is to settle as much of Gaza as possible with settlers who are not Palestinian.

If not genocide, it is ethnic cleansing.

[Michael Cain]

Trump's lawyers have told the New York State court that he can't get the $454M bond, and lacks that much cash of his own. Reportedly, none of the 30 insurers that they approached would accept any of his real estate as collateral.

[nous]

I saw that the Trump legal team are arguing that the bond amount was "unconstitutionally high" (based on testimony from one of their own expert witnesses who the trial judge had noted was a close personal friend of the Trump family, and who had a financial interest in the outcome of the trial).

If I were the judge, and in the mood to throw some shade, I'd comment to the effect that the bond amount was based upon the Trump family's own valuations of their properties, and that the Trump team had argued that the financial institutions were all well satisfied with those valuations, so what is the problem with securing a bond for fair market value?

[Snarki, child of Loki]

While the US does provide Israel with "heavy weapons", it is historically perfectly possible to conduct a genocide/ethnic cleansing with small arms and machetes.

So it's not clear how much actual influence a cut-off of military supplies would be.

[Michael Cain]

I have read that the insurers that do the big bonds all (a) have a $100M cap on what they will put up and (b) have absolute rules against non-liquid assets like real estate. Add to that Trump's history of stretching lawsuits out for years and their capital's tied up for a long time. Add to that Trump's history of stiffing people, because the insurers don't want to have to go through court to get their money.

A bank may be happy to trust Trump with a $500M three-year loan based on a building that's not really worth that much. He's apparently very good about making interest payments. At the end of three years, the bank negotiates a new interest rate and rolls the loan over. When I worked in the cable TV business, the big guys all worked that way: there was no intent to pay off the debt, debt was something that you managed in order to keep the interest payments reasonable. The State of New York's not interested in interest and rolling over the judgement.

[hairshirthedonist]

"Donald Trump Tried to Get Bond From Company He Misled About Finances"

https://www.newsweek.com/donald-trump-tried-bond-company-misled-finances-1880583

Do I need to quote "Casablanca"?

[Snarki, child of Loki]

Casablanca?

More like "We won't get fooled again".

[hairshirthedonist]

I'm really extra super surprised that Rump did that. It was like being exposed to electricity.

[wj]

Trump's lawyers have told the New York State court that he can't get the $454M bond, and lacks that much cash of his own.

IANAL, but my understanding is that, if he can't, then

1. He cannot appeal the judgement. Because that requires posting 110% of the judgement (plus interest).
2. The court starts siezing his assets, selling off property, etc., to generate enough money to pay the judgement. Note that, even if he declares bankruptcy, that doesn't absolve him from paying the full amount. I think they can even garnish his Presidential pension to get the money.

If nothing else, the process should reveal just how much of a totally mortgaged house of cards he is.

[hairshirthedonist]

This is pretty cool:

Swirling Forces, Crushing Pressures Measured in the Proton

https://www.quantamagazine.org/swirling-forces-crushing-pressures-measured-in-the-proton-20240314/

Opening:

Physicists have begun to explore the proton as if it were a subatomic planet. Cutaway maps display newfound details of the particle’s interior. The proton’s core features pressures more intense than in any other known form of matter. Halfway to the surface, clashing vortices of force push against each other. And the “planet” as a whole is smaller than previous experiments had suggested.

[GftNC]

Whoah... as if there wasn't enough to worry about with the SCOTUS six etc:

https://www.theguardian.com/us-news/2024/mar/19/far-right-fraternal-order-sacr

These certainly are strange times.

The Guardian has previously reported on SACR’s close links to the Claremont Institute, an influential rightwing thinktank with fellows who have participated in attempts to overturn the 2020 election and promoted the idea that an authoritarian “Red Caesar” might redeem a US republic they see as decadent.

SACR’s origins appear to date to the latter half of 2020, with key milestones in the group’s development coming over the following 18 months.

https://www.claremont.org/

Oh boy.

[Girl from the North Country]

I've just been watching C4 News's correspondent in Israel. He said he and the C4 News team analysed 5 days of Hebrew newscasts on the biggest and most popular Israeli channels from last week, and that none of them showed any of the suffering in Gaza, at most one showed civilians running from a bombardment. Instead, the newscasts showed what is happening in Israel, mainly about the hostages. I don't know to what extent this is a result of government censorship, or of self-censorship. That the Israeli public is not seeing the starving children and similar horrors explains I guess why public opinion has not turned more against the war, despite the general contempt for Netanyahu. Yet another argument (as if we needed one) for the importance of independent journalism.

[Michael Cain]

If nothing else, the process should reveal just how much of a totally mortgaged house of cards he is.

I've regularly said that, given the time to unwind the >500 LLCs properly -- say, at least a decade -- the stack Trump winds up with is at least a billion. If he has to liquidate on a short schedule, he's bankrupt.

I was reading a piece the other day based on interviews with a number of people deeply involved in big real estate deals and developers in the NYC metro area. The bottom line was that given a decade to unwind things, all of them would end up with at least a billion in cash. Given only a year to unwind, all of them were bankrupt.

[nous]

Here are two articles that I found regarding Israel's use of information warfare in the Israel-Hamas War:

https://foreignpolicy.com/2024/03/11/israel-gaza-hamas-netanyahu-warfare-misinformation/

https://theintercept.com/2024/02/07/gaza-israel-netanyahu-propaganda-lies-palestinians/

I usually feel like I have to do some checking where The Intercept is concerned, so seeing the same basic facts being reported at Foreign Policy (Centrist/Fact Based Reporting), and both using links to document their claims does reinforce the troubling nature of the IDF lies.

As far as the SACR piece goes, I've been seeing a lot of evidence of this type of activity going on in RW circles for a while. It's the rejection of Liberalism (via Leo Strauss and others, depending on the group) that I've mentioned before, tied to a growing insistence on Natural Law approaches to governance that give these RW political philosophers the ability to sidestep any need for secularism or pluralism.

I don't know whether to take these yahoos seriously or not. As is so often the case, I think that often what starts off as irony and provocation slips into the realm of acceptable thinking pretty quickly where the alt-right is concerned.

There's a lot more in this than I have time for during finals week grading, so I will have to leave it for now.

[Donald]

“ usually feel like I have to do some checking where The Intercept is concerned, so seeing the same basic facts being reported at Foreign Policy (Centrist/Fact Based Reporting), “

That puzzles me. I feel like that to the extent possible, I have to check everybody. I place no extra value on centrist or mainstream publications that claim to be fact- based. On this particular subject I think the NYT, for example, is biased towards presenting its readers a world where Western atrocities are aberrations to be blamed on bad apples like Netanyahu, while our enemies do things out of pure evil. And this has an effect on their reporting. We, you know, are basically civilized.

To my mind the Gaza war reveals the deep hypocrisy of Western foreign policy in a way we haven’t seen before, not because there haven’t been horrifying Western supported wars, but because even with attempted Israeli censorship the internet has made it impossible to cover things up. I along with countless others were watching these grotesque videos out up by Israeli soldiers months before the NYT reported on it. I think in most previous horrible Western wars things had to be filtered through the MSM or alternatively you might read some lefty antiwar magazine with its own spin or some book written years later and in my case I would try to verify some of the leftish claims against mainstream sources where possible, or there was always the human rights organizations. But I didn’t give the mainstream sources special credence— it was a different pov, that’s all.

Now all you need is Twitter. You will see some almost live reporting of things the press may or may not get to eventually, or which it may downplay if it does report.

[nous]

That puzzles me. I feel like that to the extent possible, I have to check everybody. I place no extra value on centrist or mainstream publications that claim to be fact- based.

I do agree for the most part, but I also think that past history with a particular editorial stance shapes the degree to which one turns a critical eye on the reporting. When I say that I feel the need to check The Intercept, it's because I have in the past felt that their biases (laudable as some of those biases are) have gotten in the way of their analysis in a way that damages their ethos. With Foreign Policy, meanwhile, I find that I often disagree with their biases, but find that their journalistic ethics are strong and their presentation of the story is basically fair and faithful to the record. So when I see them reporting the same things as The Intercept, I feel like that helps to confirm what The Intercept is reporting, because they do not share an ideological bias.

On this particular subject I think the NYT, for example, is biased towards presenting its readers a world where Western atrocities are aberrations to be blamed on bad apples like Netanyahu, while our enemies do things out of pure evil. And this has an effect on their reporting. We, you know, are basically civilized.

Agree. This is why the NYT has been having a hard time of it lately, what with the tensions between the print and the digital sides over the Israel-Palestine reporting. There's a clear difference in fundamental worldview between the old and the new.

[Donald]

Not enough time to get into this regarding the Intercept— I think ther biases have interfered in ther reporting a few times, but in the area I am thinking of, it was a mainstream bias.

On the NYT we agree. Today they put out a piece which is a perfect illustration of what they often do—

https://www.nytimes.com/interactive/2024/03/20/world/middleeast/gaza-aid-delivery.html?unlocked_article_code=1.eE0.8v6_.HR35uKAE3MEg&smid=url-share

There is not a word here about Israeli officials stating they want to keep aid out or about Israeli protests blocking trucks or about polls showing Israelis opposing aid. I don’t think there is anything factually false in the piece, but it is an extremely deceptive piece in what it chooses to leave out and in how it reduces the aid problem to a series of technical issues and differing opinions between various people. If Israel were our enemy or even if they were some country which wasn’t an ally there is no way the Israeli statements would be left out. They would be the centerpiece of the article.

[wonkie]

"'ve just been watching C4 News's correspondent in Israel. He said he and the C4 News team analyzed 5 days of Hebrew newscasts on the biggest and most popular Israeli channels from last week, and that none of them showed any of the suffering in Gaza, at most one showed civilians running from a bombardment."

I understand that seeing is believing and not-seeing is not believing--but. BUT. A person who didn't believe in concentration camps due to not having seen one, but saw the smoke and the trains, engaged in willful not seeing.

Everyone in Israel knows that Gaza is surrounded by walls, is a small and very crowded area, that the population has been forced to abandon their means of sustaining life to avoid the bombs and has been bombed to the point where there couldn't possibly be any viable economy or infrastructure left, so I think there has to be some willful not seeing going on.

[hairshirthedonist]

It's not like you can't get coverage from Al Jazeera in Israel ... with very little effort ... if you're interested.

[Girl from the North Country]

so I think there has to be some willful not seeing going on.

Agreed.

[nous]

Information Warfare is designed to sustain that not-seeing. The basic idea there is to occupy and dominate the targets' attentions with messages that engage a threat response and provoke further engagement with other messages that reinforce the original one.

It takes a lot less effort to not-see when there are so many messages out there that are much more limbically rewarding to view that are within reach. If one is going to be upset, then one usually will prefer the flavor of upset that does not provoke moral crisis and attacks of conscience. This is especially true when in-group membership is threatened. Very few people want to risk their tribal membership over cognitive dissonance.

[wonkie]

Yes, I understand that willful not seeing is pretty much a universal human trait. Also, sometimes the not seeing is because of genuinely not seeing or it could be more a failure to make reasonable inferences from what is seen rather than an active turning of one's back.

[GftNC]

Meanwhile, Michael Cohen avoids punishment for using Google Bard to write (and give fake cases as precedent in) the motion that he filed before a federal court:

https://www.nytimes.com/2024/03/20/nyregion/michael-cohen-artificial-intelligence.html?unlocked_article_code=1.eU0.omHc.8g2dsFPlDQie&smid=url-share

The jury's out on Charles, however....

[wj]

Even Artificial "Intelligence" ought to have given him a leg up on the stumblebums TIFG keeps sending into court. But apparently his lawyer is from the same diploma mill.

[Pro Bono]

On an entirely unrelated subject, I'm curious about this recent xkcd strip.

I suppose that the suggestion is that (almost) all the vowels are pronounced as schwas. But to me, they would be (with some variation depending on how carefully one was speaking):

Whɒt's ʌp? Wɒs Dʌg gənnə cʌm? Dʌg lʌvs brʌnch.

Nəh əh, Dʌg's stʌck 'cɒs əf ə tʌnnəl ɒbstrʌctən. ə trʌck dʌmpd ə tʌn əf ʌniəns.

Where, if anywhere, are these mostly schwas?

[CharlesWT]

On an entirely unrelated subject, ...

"The field of phonetics delves into the problems of transcribing spoken languages into written symbols. Phoneticians study the ways in which specific combinations of consonants and vowels convey meaning. Many languages have vowel sounds which are represented by the schwa vowel symbol /ə/, as illustrated in this paragraph." —Anthropic Claude-3-Sonnet

[nous]

@Pro Bono - That XKCD pronunciation may be possible for someone with a strong California Vowel Shift (that's how I read it). A lot gets shifted forward and upward, and flattened out (environmental effects of sun, surf, and weed?).

I'd ask a friend who is a dialect coach, but he's out on sabbatical at the moment. Pretty sure he could turn this into an hour lecture, with demonstrations.

[Girl from the North Country]

Charles, since you quote the beginning of Pro Bono's question at the top of your (Anthropic Claude3's) post, I would be grateful if you would explain what exactly in PB's post led you to believe that he didn't already know and understand the information in your (Anthropic Claude3's) paragraph?

[Snarki, child of Loki]

Agree on the schwa-ization (is that a word? maybe now it is!) of californian dialect.

With the exception of the "u" sound in "Dude" that precedes/follows every sentence in surfer-speak, which gets even more emphasized.

As English-speakers migrated west across North America, all the accent differences got blended and homogenized. That's my story and I'm sticking to it, dude.

[JanieM]

Asked google for "schwa" and got their quick result from "Oxford Languages":

the unstressed central vowel (as in a mom e nt a go), represented by the symbol /ə/ in the International Phonetic Alphabet.

If the schwa is by definition unstressed, then the whole premise of the xkcd comic is mistaken and it falls flat. (Though that's apparently not Pro Bono's objection.)

*****

But if I were to accept the premise, then the xkcd pronunciation is very close to my dialect's pronunciation -- which doesn't require a California Vowel Shift, since it has its origins in northeastern Ohio 70+ years ago (where people have undergone some vowel shifts in the interim, for that matter), with fifty or so years in New England added on.

If I say "a cup," the two vowel sounds are almost indistinguishable. The "a" is unstressed, but my mouth is in the same position for both, or close enough.

*****

Afterthought: again from Oxford Languages, for the pronunciation of cup:

cup
/kəp/

So much for the schwa being unstressed by definition. ;-)

[Pro Bono]

I see that wiktionary offers distinct English and US audio recordings of 'cup'.

[nous]

Now is as good a time as any to pull out the HowCast accents series. Lots of good musings on oral posture and diphthongs and whatnot:

https://www.youtube.com/playlist?list=PLLALQuK1NDrjVi9sibfkhTN372_bd470N

(Andrea is also a very nice human being. She collaborates with my friend on a lot of speech and dialect stuff.)

[JanieM]

I see that wiktionary offers distinct English and US audio recordings of 'cup'.

Too bad they don't have a recording of someone from Ireland saying it. In my experience that would be much more distinguishable from *either* US or UK than either of the latter is from the other.

You can hear it quickly in passing in the first few seconds of this video:

https://www.youtube.com/watch?v=X1eC4WX_QI4

[Pro Bono]

They do mention the Northern English pronunciation /kʊp/, which is similar to the Irish.

[Michael Cain]

I come closer to the xkcd version if I'm in a casual setting. Stick me in front of a class or legislative committee and (involuntarily) I'm in the academic version of General American: over-enunciate a bit, vowels differentiated more, dropped consonants get quite rare. My wife used to call it "Mike, in lecture mode".

Even old speech recognition software was very good at getting "Mike, in lecture mode" correct. I assume that machine learning versions are at least very good for everyone these days.

[CharlesWT]

I would be grateful if you would explain what exactly in PB's post led you to believe that he didn't already know and understand the information in your (Anthropic Claude3's) paragraph?

I didn't make any assumptions about PB's knowledge or understanding of the subject of his comment.

The text is Claude's response to the prompt, "Write a paragraph about phonetics using words with no vowel sounds but the schwa vowel sound."

My comment was a backhanded way of pointing out that there is a dwindling number of subjects entirely unrelated to AI.

[Michael Cain]

The text is Claude's response to the prompt, "Write a paragraph about phonetics using words with no vowel sounds but the schwa vowel sound."

And of course, the second word in the response fails to satisfy the criterion. Not surprising, I think; adding a filter like only schwa vowel sounds is a rather unusual thing to put into the next-word decision. I wonder how complicated a prompt might be necessary to get something like that to be effective, even without limiting the general subject?

[CharlesWT]

And of course, the second word in the response fails to satisfy the criterion.

I defer. In grade school, the subject of phonics went completely over my head.

More from Claude.

"Th' fəld əv fənɛtɪks dɛlvz ɪntə th' prəblɛmz əv trənskrɪbɪŋ spəkən lɑːŋgwɪdʒɪz ɪntə rɪtən səmbəlz. Fənɛtɪʃənz stʌdi th' wɛz ɪn wɪtʃ spɛsɪfɪk kəmbɪnəʃənz əv kənsənənts ən vɐwlz kənvə mɪnɪŋ. Mənɪ lɑːŋgwɪdʒɪz hæv vɐwl səʊndz wɪtʃ ər rɛprzɛntɪd bɐ θə ʃwɑː vɐwl sɐmbəl /ə/, əz ɪlʌstrətɪd ɪn ðɪs pɛrəgrɑːf."

List the vowel sounds used in the text.

"The vowel sounds used in the text are:

/ə/ - the schwa vowel sound, as in "the", "consonants", "transcribing", etc.
/ɛ/ - as in "delves", "spellings"
/ɪ/ - as in "phonetics", "linguistics", "specific"
/ɑː/ - as in "languages", "transcribing"
/ʌ/ - as in "study", "illustrating"
/ɐ/ - as in "word", "convey"
/ɔː/ - as in "formations", "combinations"

No other vowel sounds besides the schwa /ə/ are used in this text."

Is it contradicting itself?

[nous]

CharlesWT - It's not contradicting itself because it's not actually making an argument. It's doing its best to imitate the form and content of an argument featuring the required ingredients.

It has no comprehension and is merely doing its best to script something that looks like it might score well with the trainers.

And as far as that goes, I'd be surprised if the training actually connects the graphemes it manipulates with the phonemes we are arguing over.

[JanieM]

It's not contradicting itself because it's not actually making an argument.

You don't think it's contradicting itself by listing the sounds used in the text (all 7 of them) and then saying there's only one used in the text? Are we living in two only partially overlapping universes?

*****

CharlesWT: My comment was a backhanded way of pointing out that there is a dwindling number of subjects entirely unrelated to AI.

I have no idea what this even means. What does it mean for a subject to be “related” to AI? As distinct, for example, from being a subject that AI can create semblances of English sentences about?

*****

Michael: adding a filter like only schwa vowel sounds is a rather unusual thing to put into the next-word decision.

And yet, I’ll bet the cartoonist who creates xkcd comics did it fairly easily. (If you will grant my provisional/reluctant agreement with the cartoonist that “a” and “cup” use the same vowel sound, even if they’re not both technically schwas -- which certainly seems to have been *his* premise.)

[JanieM]

PS -- nous, thanks for the link to the HowCast videos. Really cool!!!!!

I was once a sort of groupie for a Maine trio called Schooner Fare (whose music can be found on YouTube) -- Tom Rowe, who died young, used to do a between songs routine that involved imitating / demonstrating six or seven different *Maine* accents, never mind going outside the borders of the state. "This is Washington County, this is Oxford County, etc." And he could make them distinguishable. It was great fun.

[Hartmut]

Also note that it uses "transcribing" as demonstrating two different "a" sounds. ;-)

As for connections between graphemes and phonemes, one would probably need a talking AI to test that.

I better not make fun of any German dialects here, in particular not telling jokes about Saxon goose meat.

[nous]

You don't think it's contradicting itself by listing the sounds used in the text (all 7 of them) and then saying there's only one used in the text? Are we living in two only partially overlapping universes?

Oh, the information being presented is full of contradictions, but the LLM cannot contradict itself because it has no actual understanding and is therefore groundless in its output. It's like asking the LLM for a list of jammy red wines and then getting something that is either not a red, or not jammy in the list.

It's never actually seen or tasted wine, so it couldn't actually argue whether or not a given wine is jammy.

[Girl from the North Country]

I have no idea what this even means. What does it mean for a subject to be “related” to AI? As distinct, for example, from being a subject that AI can create semblances of English sentences about?

My thoughts exactly. And apart from the fact that when I say "a cup" the two sounds are completely different, and only the first is a schwa, I would also like to put on record that my late husband, North Country born and bred, pronounced "cup" very like the Irish example, and so did all the people thereabouts. And when I asked, once, in a supermarket there, whether they had a particular product fresh (as opposed to jarred), or if they only had it in summer, two separate women hesitantly said "sammer?" and didn't recognise the word.

[GftNC]

Interesting article on solidarity in today's NYT:

Policies can either foster solidarity and help repair the divides that separate us or deepen the fissures.

Today, the American welfare state too often does the latter. As sociologists including Suzanne Mettler and Matthew Desmond have detailed, lower-income people tend to be stigmatized for needing assistance, while more-affluent citizens reap a range of benefits that are comparatively invisible, mainly through tax credits and tax breaks. Both arrangements — the highly visible and stigmatized aid to the poor and the more invisible and socially acceptable aid to the affluent — serve to foster resentment and obscure how we are all dependent on the state in various ways.

Instead of treating citizens as passive and isolated recipients of services delivered from on high, a solidarity state would experiment with creative ways of fostering connection and participation at every opportunity for more Americans. What if we had basic guarantees that were universal rather than means-tested programs that distinguish between the deserving and undeserving, stigmatizing some and setting groups apart? What if, following the model of a widely admired program in Canada, the government aided groups of private citizens who want to sponsor and subsidize migrants and refugees? What if public schools, post offices, transit systems, parks, public utilities and jobs programs were explicitly designed to facilitate social connection and solidarity in addition to providing essential support and services?

https://www.nytimes.com/2024/03/21/opinion/democracy-solidarity-trump.html?unlocked_article_code=1.ek0.Yn7j.369XPkK7is9e&smid=url-share

[wj]

As I think on it, when I say "a", it is typically pronounced "uh" -- like the "u" in "cup". (I'm having no success with pasting IPA symbols. Including copying from Charles' post above. Sorry.)

UNLESS it is being used, in context, to mean "one and only one". In which case, it is pronounced like the "a" in "cape" or "space" or "face".

If I say "I'll have a cup" in the usual pronunciation, that accepts that there may be more to follow. But if I use the second pronunciation, that means no more after the first, so don't even bother to ask.

But then, my native Californian pronunciation dates from before the reputed vowel shift.

[Hartmut]

I am not Californian but I'd say I pronounce it the way that wj describes. I use the 'a' as in 'space' when I want to emphasize the 'a' and put it in opposition to 'the' (which in that context also gets a pronunciation shift towards 'thee') as in "this is *a* (possible) solution not *the* (=only) solution.".

[liberal japonicus]

Having lived and taught in Japan for more than 30 years, I have found that I don't use schwas but end up moving towards the dictionary pronunciations, which people over here refer to as 'english teacher-ese' We still use schwas, but not to the extent that they would be used in regular speech back home.

This might be of interest
https://www.scientificamerican.com/podcast/episode/linguists-hear-an-accent-begin/

[JanieM]

lj -- thanks for that article, very interesting. My private theory is that people want to talk like the cool kids talk....

And now I wonder whether the cool kids unconsciously alter their accents precisely to gain followers and affirmation for their coolness.

This is on my mind (both coolness and accents) because I have just finished watching the 4 seasons of Sex Education. A veritable feast of different accents!

[wonkie]

My nieces both speak in a way that seems to be a younger generation thing. It's not a matter of slang; it's intonation. I'm not around them or around others of their age and socio/economic demographic to explain the phenomenon very well, but I have seen it mocked on you tube or satirized in ads. An upward turn at the end of a sentence that makes everything a question. A drag on the first words of a sentence that makes all the following words seem slightly ironic or sardonic even if the subject matter is something as mundane as choice of scrambled or fried eggs for breakfast. An I'm-so-cool distancing.
It annoys me because it sounds pretentious, but I'm old.

[Michael Cain]

It's not a matter of slang; it's intonation.

My ten-year-old granddaughter and I (now 70) both use "cool". The intonations are radically different.

[Snarki, child of Loki]

wonkie, it's called "uptalk".
Yes, it has the implication of lack of confidence, but (guess) mostly to those not already kewl kidz.

Oldsters hand the youngsters a "tool" (language), and then are appalled when the youngsters use that tool for their own purposes, part the infinity.

[Girl from the North Country]

When that "uptalk" phenomenon started becoming prevalent here, some years ago, people used to say that it had originated in Australia. I don't know if that's true, but I am quite prepared to blame them for something else in addition to the origin of Rupert Murdoch.

[CharlesWT]

I first noticed uptalk about three decades ago at a temp job I had. Every instruction the supervisor gave sounded like a question.

[wonkie]

This is more in the nature of annoying verbal tics but have you all noticed how service people gush "Perfect!" at the slightest provocation? "Sign your name here---perfect!" "Debit or credit? Perfect!" "And now a signature here...perfect!" I'm probably a grumpy old lady, but I find it annoying to be congratulated for remembering my name.

[Michael Cain]

"Sign your name here---perfect!"

And they all changed to it seemingly overnight. I've finally got to the point where it no longer grates. What are they commenting on? That I hit the line they were pointing at? The quality of the signature on some scale?

[JanieM]

Okay, if we're doing old person pet peeves, mine has been around for decades, and it still annoys me. It's using "No problem" for "You're welcome."

Damned right it had better be no problem, it's your fncking job. As if, if they were in a different mood, my actually buying something in their store *would* be a problem, but they're magnanimously letting me do it just this once.

Grumpy-old-lady is my middle name. ;-)

[Michael Cain]

When did battleship gray become a preferred car color?

[JanieM]

Michael: funny you should mention that. I had car problems for a year and a half and it was time to replace my car anyhow, but when I first started looking in August of 2022, there was not a single car on the lot at the dealership I prefer to shop at that was on my list of possibles. (Inventory problems have been dire, I guess due at least in part to covid.)

Cut to a year and a half later, January 2024, and I was involved in a car accident, one factor in which was that the car I collided with was a very old, very small gray sedan that was almost invisible because it was the same color as the road, and my attention had been caught by the much larger white vehicle that was following it. (It was also a very glare-y post snowstorm day, but never mind all the complications.)

Thank goodness, it was a very slow speed collision and no one was hurt. But my car was totaled, so my hand was forced on the "get a new car" front. I talked with a couple of friends about it, and one of them brought up visibility issues. You can google it if you want; there has apparently been some research done on visibility. But I was clear on what color I wanted, the same color I *always* want (blue), and I was skeptical about the visibility research anyhow because it didn't make any mention of visibility in different landscapes. (It did mention time of day, and that dark colors are less visible at night....)

The most interesting theory I've heard about the popularity of gray (which probably came from someone at BJ, it's the only other blog I read regularly) is that car dealers stock a lot of neutral color cars (e.g. gray) because when people come in wanting a particular color, e.g. red, if the dealer doesn't have it the customer is more likely to settle for a neutral color than for some other strong color (e.g. blue or green).

Or -- maybe it's just a fad. Some "cool kid" chose gray and from there it spread....

I was adamant about *not* wanting gray, having just had a hard lesson in its (in)visibility.

[Snarki, child of Loki]

I heard a LONG time ago that white was a better car color, because it stays cooler in the summer.

Car engines put out lots of waste heat, so no problem heating in winter.

Visibility? Leave the lights on.

[CharlesWT]

(It did mention time of day, and that dark colors are less visible at night....)

This raises the question, "Why are so many emergency vehicles painted dark red?"

[Snarki, child of Loki]

IIRC, there were studies that showed that emergency vehicles should be bright YELLOW for improved visibility.

Some are! But it'll take a while for the obstructive traditionalists to vacate the stage.

[Hartmut]

Over here they use special reds reserved for firetrucks and emergency vehicles (and those have been regularly updated with extra contrast elements) so they stick out both day and night even when they do not use the siren and the blue signal lights.

[Michael Cain]

Car engines put out lots of waste heat...

I remember a talk I saw back in the 1990s where the speaker's thesis was that car design had reached the point that the limiting factor was heat management.

[Michael Cain]

Some "cool kid" chose gray and from there it spread....

The gray that really sticks out to my eye is the almost flat color, effectively no gloss at all. Probably a vinyl wrap of some sort. I can't imagine painting a car that color if you ever intend to sell or trade it.

[nous]

Gray is "safe" for resale, and it doesn't show dirt as much as either black or white will.

White is safer for visibility and has an aura of luxury about it, but it will show dirt and is harder to keep clean when it comes time to resell.

Any one particular color has a greater chance of being out-of-fashion when it comes time to resell than the grayscales, so gray, white, black, and silver continue to dominate.

When we got our Mini, my wife had wanted to get an orange/yellow one with black top and bonnet stripes, but the one that we got was a red one that had been sitting on the lot for a while. Our salesperson said that the orange color was mostly popular with younger men, and that the red would widen the resale appeal by at least double.

[Girl from the North Country]

What seems to be becoming popular here, at least on rather high end cars, is a completely matte powdery look, often (although not always) in khaki or olive so almost military looking. I wonder if that is what Michael Cain is talking about: certainly there is no iota of gloss. I wonder if it shows scratches as much as gloss finishes?

[Tony P.]

What strains MY aging eyes is the ridiculously "neutral" color scheme that many desktop "apps" have been adopting for some years. I mean, light gray slider buttons on off-white slider strips?! Practically invisible grid lines in Excel, fainter and fainter tones in Google Maps, WTF? Does nobody older than 30 design user interfaces any more?

And don't get me started on controls or menu bars that play hide-and-seek just to show off.

Bah, humbug!

--TP