@weekend_editor@mathstodon.xyz
@weekend_editor@mathstodon.xyz avatar

weekend_editor

@weekend_editor@mathstodon.xyz

Retired physicist, after a career in machine learning & stats mostly for cancer drug discovery. Now blogging about stats in the news.

Avatar: convergence basins in the complex plane of Newton's algorithm searching for the cube roots of unity. (After a NYT column by https://mathstodon.xyz/@stevenstrogatz, long ago.)

Header: Quote from GK Chesterton, London Daily News, 1905-Aug-16 on epistemic humility and the ability to say "I am wrong" as the foundation of idealism.

#statistics #physics #r

This profile is from a federated server and may be incomplete. Browse more on the original instance.

inthehands, to random
@inthehands@hachyderm.io avatar

There’s a lot to chew on in this short article (ht @ajsadauskas):
https://www.bbc.com/worklife/article/20240214-ai-recruiting-hiring-software-bias-discrimination

“An AI resume screener…trained on CVs of employees already at the firm” gave candidates extra marks if they listed male-associated sports, and downgraded female-associated sports.

Bias like this is enraging, but completely unsurprising to anybody who knows half a thing about how machine learning works. Which apparently doesn’t include a lot of execs and HR folks.

1/

weekend_editor,
@weekend_editor@mathstodon.xyz avatar

@inthehands @ajsadauskas

If your dataset contains biases, then ANYTHING you train on it will inherit those biases (absent specific corrective action).

Pretty much every textbook tells you that your models will sometimes train on non-obvious "details". This applies to AI. It applies to machine learning. It applies to statistics, even simple old regression and classification.

If AI/statistics practitioners know this, why do we have to keep re-learning this lesson the hard way?

Perhaps managements need a couple knocks to the side of the head to beat this fact into them?

pauldrye, to random
@pauldrye@spacey.space avatar

So this happened. Given the colour and the fact they were working with nitric acid, I'm feeling somewhat doubtful this was (relatively benign) nitrous oxide.

Source: https://www.bbc.com/news/articles/cd11xzjkvk0o

weekend_editor,
@weekend_editor@mathstodon.xyz avatar

@cstross @mmby @nyrath @pauldrye

It sure LOOKS like a cloud of NO2 -- at sufficiently high temperature, it has that red/orange color. No fun to breathe, as it makes nitric acid from water in lung tissue.

A fertilizer plant would plausibly be working with ammonia, which can produce NO2 by the Ostwald process (used for making nitric acid, and this apparently was a plant that did that):

4 NH3 + 7 O2 -> 4 NO2 + 6 H2O

Or a fertilizer plant might also have nitric acid around, and some knucklehead didn't understand to keep it away from copper fittings:

4 HNO3 + Cu -> Cu(NO3)2 + 2 NO2 + 2 H2O

Those are the ones that come to mind as bulk reactions likely to be in a fertilizer plant. There are others, but then you have to figure out why a fertilizer plant might be doing that.

But the spokesperson who said it was "nitrous oxide" (which is transparent) or the reporter who mis-heard it need to have a word with some industrial chemists.

johncarlosbaez, to random
@johncarlosbaez@mathstodon.xyz avatar

Is there a chance that the physicist Oliver Heaviside was really Wolverine?

image/jpeg

weekend_editor,
@weekend_editor@mathstodon.xyz avatar

@johncarlosbaez

Well, that might explain why he kept obsessively painting his fingernails in his later years.

Maybe corrosion-resistant paint for the adamantium?

gutenberg_org, to books
@gutenberg_org@mastodon.social avatar

English self-taught mathematician and physicist Oliver Heaviside was born #OTD in 1850.

He invented a new technique for solving differential equations, independently developed vector calculus, and rewrote Maxwell's equations in the form commonly used today. He significantly shaped the way Maxwell's equations are understood and applied in the decades following Maxwell's death. His practical experience in telegraphy provided a foundation for his later theoretical work.

#books #science #physics

Cover of Electromagnetic theory by Heaviside, Oliver, 1850-1925 Publication date 1922 Topics Electromagnetic theory, Vector analysis, Electric waves Publisher London : Benn

weekend_editor,
@weekend_editor@mathstodon.xyz avatar

@gutenberg_org

He was an unusual man, who had some unfortunate mental health problems in later life: signing his name with the "title" suffix "W.O.R.M.", painting his nails pink (more unusual then than it would be now), replacing his furniture with granite blocks, and so on. (See Wikipedia, below.)

On the other hand, I sorta like the idea that there's a lot more room for weirdness in the world than we normally allow for.

https://en.wikipedia.org/wiki/Oliver_Heaviside#Later_years_and_views:~:text=In%20later%20years,3%5D%3A%E2%80%8Axx

cstross, to random
@cstross@wandering.shop avatar

New blog entry: Harmless Fun! https://www.antipope.org/charlie/blog-static/2024/05/harmless-fun.html

weekend_editor,
@weekend_editor@mathstodon.xyz avatar

@cstross

There is a promising line of thought you might wish to consider:

Ayn Rand is a Reptiloid.
Reptiloids are dinosaurs.
Dinosaurs are birds.
Birds aren't real.

Therefore, Ayn Rand isn't real.

amoroso, to Lisp
@amoroso@fosstodon.org avatar

I didn't realize ACM makes available the full-ish archive of the LISP Pointers journal SIGPLAN published from the late 1980s to the mid 1990s. It contains most of the papers of most of the issues, an historical treasure of practical value.

https://dl.acm.org/loi/sigplan-lisppointers

weekend_editor,
@weekend_editor@mathstodon.xyz avatar

@amoroso

I was just thinking how much I missed this, when it folded in the 90s!

weekend_editor,
@weekend_editor@mathstodon.xyz avatar

@amoroso

I may have some in a box around here somewhere. You can have them if you want, though given they're online, I dunno why anyone would want them.

Gotta find 'em first, though.

seraph, to random
@seraph@wetdry.world avatar

this shit is so funny

Are you serious, X Corp? Ahoy there, welcome to roblotwitter! I assure you, there's nothing fishy going on here, so feel free to read on. Yeah, it's a "honeypot". Sorry about that. I'm not trying to apologize and get away with it, though. But when you clicked on this link, you probably thought you were looking at something like "roblox.com". Simple URL substitution can cause this kind of thing to happen, so I made this site. So let's shout it out. "Are you serious, X Corp?"

weekend_editor,
@weekend_editor@mathstodon.xyz avatar

@dpnash @seraph @pluralistic
@cstross

Or even earlier, AOL and its problem with the good people of Scunthorpe:

https://en.wikipedia.org/wiki/Scunthorpe_problem

johncarlosbaez, (edited ) to random
@johncarlosbaez@mathstodon.xyz avatar

@TruthSandwich got me interested in the vibrational modes of bells. They're not harmonics with frequencies 1, 2, 3, 4, ... times the lowest frequency: they're much more complicated! That's why bells sound clangy. This chart shows how they sometimes work.

The lowest frequency vibrations are called:

• the 'hum' (the lowest frequency)

• the 'prime' (with frequency roughly 2 times that of the hum)

• the 'tierce' (roughly 2.4 times the hum, so a minor third above the prime)

• the 'quint' (roughly 3 times the hum, so a major fifth above the prime)

• the 'nominal' (roughly 4 times the hum, so an octave above the prime)

and so on. If you think these names are illogical, join the club! One reason it's tricky is that the loudest vibration is not the lowest one: it's the 'prime'.

The numbers I just gave you should be taken with a big grain of salt. They really depend on the shape of the bell, and you'd have to be great at designing bells to make them come out as shown here. It's not like a violin string or flute, where the math is on your side.

This quote helps explain the chart:

"Modern theory separates the modes of vibration into those produced by the "soundbow" and those produced by the remaining bell "shell". The bell vibrates both radially and axially and the principal vibrational modes are shown in the diagram together with their classification using the scheme proposed by Perrin et al. This scheme consists of the mode of vibration (RIR - Ring Inextensional Radial, RA - Ring Axial, R=n - Shell driven), the number of meridians (where “m” is half the number of meridians) and the number of nodal circles (n)."

Starting to sound like orbitals in quantum mechanics!

(1/3)

weekend_editor,
@weekend_editor@mathstodon.xyz avatar

@johncarlosbaez

And to the great sadness of everyone, Bell's Theorem in QM appertains to no bells.

cstross, to random
@cstross@wandering.shop avatar

Today's writing: what if Doctor Phibes murdered and impersonated Max, owner of the KitKatKlub, in order to run an elaborate sting against Doctor Mabuse—film recap, starring Vincent Price, Diana Rigg, Charlotte Rampling, and Peter Cushing.

weekend_editor,
@weekend_editor@mathstodon.xyz avatar

@cstross

Vincent Price & Dianna Rigg?

You could just lead with that. ("Oh, and there's some plot, too.")

johncarlosbaez, to random
@johncarlosbaez@mathstodon.xyz avatar

The press likes to talk about problems, so it paints everything as a problem. But some problems are better than other problems.

From here: https://archive.is/lx6CC

weekend_editor,
@weekend_editor@mathstodon.xyz avatar

@johncarlosbaez

An excellent way to put it.

cstross, to random
@cstross@wandering.shop avatar

From America with cash: Right-wing groups want to end abortion in the UK

A right-wing political and media ecosystem pushing a US-style anti-abortion agenda is gaining traction in the UK

https://www.opendemocracy.net/en/us-anti-abortion-culture-war-uk-stella-creasy-amendement-/

Meanwhile:

YouGov survey finds nearly 90% of Britons support abortion

https://www.msichoices.org/latest/yougov-survey-finds-nearly-90-of-britons-support-abortion-in-the-uk/

weekend_editor,
@weekend_editor@mathstodon.xyz avatar

@cstross

Probably I have a parochial American viewpoint, but it continually surprises me how much fascism revenant and its variations are a world-wide phenomenon.

weekend_editor,
@weekend_editor@mathstodon.xyz avatar

@cstross

Agreed the roots are deep, and the resurgence fills a need felt by authoritarian followers (e.g., consider Bob Altemeyer's absolutely masterful poli sci book, The Authoritarians).

After all, Genghis Khan, Amir Timur (Tamburlaine the Great), arguably Julius Caesar and some of the Tokugawa shoguns... they all came from somewhere.

It bewilders me why it keeps coming BACK. Some decades ago, stories from a Chilean friend who grew up under Pinochet were deeply confusing. Now I find my neighbors confusing, for similar reasons.

You may be onto something with the antagonism of democracy and the combination of imperialism and unregulated capitalism. Concentrated money wants concentrated power.

Surely there's some history book I'm missing?

weekend_editor,
@weekend_editor@mathstodon.xyz avatar

@cstross

I thought something similar during the financial crisis of 2007-2008.

Republican attempts to cut spending, particularly the social programs they hate, would have been anti-stimulus.

Adults who understood the Depression were gone, so we attempted to repeat their mistakes.

johncarlosbaez, (edited ) to random
@johncarlosbaez@mathstodon.xyz avatar

What math should you know before studying the proof of Fermat's Last Theorem? That's what this video answers.

It's pretty funny. I don't want to completely give it away, but I know a fair amount of math, and a sinking feeling spread over me as the video went on... so I'm curious when that feeling will hit you.

Eventually the guy suggests that you should really learn French... so you can read Grothendieck's 12-volume Séminaire de Géométrie Algébrique, which is thousands of pages long. But for me, even that's not the worst part!

The good news is that this guy, Anthony Vasutaro, admits it's fine to plunge in and learn stuff as needed. Even better, he is really serious about this stuff. He has a huge series of videos explaining the proof of Fermat's Last Theorem in detail.

Personally I'm more interested in the 'big ideas' than the actual proof. And for the big ideas, I'm finding Nigel Boston's paper quite nice:

• The proof of Fermat's Last Theorem, https://people.math.wisc.edu/~nboston/869.pdf

Beware: this is only for mathematicians! There are lots of other introductions at varying levels of intensity. No matter how little math you know, there's one that you'll enjoy.

https://www.youtube.com/watch?v=MnAvpa1oUEk

weekend_editor,
@weekend_editor@mathstodon.xyz avatar

@johncarlosbaez

"... learn French... so you can read Grothendieck..."?!

I guess it could have been worse. He could have told people to read Russell & Whitehad so they'd know the foundations? :-)

cstross, to random
@cstross@wandering.shop avatar

I notice that the Odysseus commercial lander a US company just plonked on the Moon cost NASA a little over $100M.

By way of comparison, the Indian government's Chandrayaan-3 mission, the lander of which touched down in 2023, cost roughly $90M.

So private space can be more expensive than 100% government agency-run missions.

weekend_editor,
@weekend_editor@mathstodon.xyz avatar

@cstross

Just wait 'til the private equity and hedge fund bros pick up the scent of money in space.

gutenberg_org, to science
@gutenberg_org@mastodon.social avatar

Austrian physicist and philosopher Ludwig Boltzmann was born in 1844.

His greatest achievements were the development of statistical mechanics, & the statistical explanation of the second law of thermodynamics. In 1877 he provided the current definition of entropy, S = kB ln Ω, where Ω is the number of microstates whose energy equals the system's energy, interpreted as a measure of statistical disorder of a system. Max Planck named the constant kB the Boltzmann constant.

Title page to volume I and II of "Vorlesungen über Gastheorie" by Ludwig Boltzmann (1896-1898). Copy located in the Niels Bohr Library & Archives, American Institute of Physics, in College Park, Maryland.

weekend_editor,
@weekend_editor@mathstodon.xyz avatar

@gutenberg_org

An interesting fact about Boltzmann and (S = k_B \ln \Omega) is that this equation is the epitaph on his tombstone in Vienna (image below).

Also, because of the manner of his death, the following introduction to DL Goodstein's 1975 book States of Matter is... well... also interesting (at least):

"Ludwig Boltzmann, who spent much of his life studying statistical mechanics, died in 1906, by his own hand. Paul Ehrenfest, carrying on the work, died similarly in 1933. Now it is our turn to study statistical mechanics. Perhaps it will be wise to approach the subject cautiously."

Image is Chapter 1, section 1.1, page 1. At least you get warned up front.

DL Goodstein, States of Matter, Dover 1975 (republished 1985). ISBNs: 9780486649276, 048664927X.
https://www.google.com/books/edition/States_of_Matter/AVeVAwAAQBAJ?hl=en

The very first thing one sees in Goodstein's 1975

johncarlosbaez, to random
@johncarlosbaez@mathstodon.xyz avatar

Today I got some questions about logic, like:

what do provability, decidability, consistency, and completeness mean?
how do we work with equality in set theory?

I don't know articles that explain topics like this to non-mathematicians, clearly and crisply, without becoming overlong or heavy with notation. Do you?

I looked around and found this "Introduction to First-Order Logic":

https://builds.openlogicproject.org/content/first-order-logic/introduction/introduction.pdf

but the very first sentence is

"You are probably familiar with first-order logic from your first introduction to formal logic."

which is basically a way of saying "fuck you - if you don't know this stuff already I won't explain it to you".

As a student I liked Boolos and Jeffrey's book "Computability and Logic":

http://alcom.ee.ntu.edu.tw/system/privatezone/uploads/Logic/20090928151927_George_S._Boolos,_John_P._Burgess,_Richard_C.

but that's more like a course than what I'm thinking of here: a collection of essays that explain different topics in plain English.

I also liked Hofstadter's "Gödel, Escher, Bach", but that's a massive quirky elaborate tale, not a simple clear explanation.

Wikipedia articles are packed with information but they aren't self-contained, clearly written essays. Articles in the Stanford Encyclopedia of Philosophy are better in some ways, but they often "show off" by including more advanced material.

Sigh....

weekend_editor,
@weekend_editor@mathstodon.xyz avatar

@johncarlosbaez

I've sometimes used JN Crossley's What is Mathematical Logic? to good effect for students who needed a lightning introduction. Also, it's short (~90ish pages) and cheap (Dover edition, usually under ten bucks). (Disclaimer: I haven't been in that sort of classroom for decades, though.)

Just reading the first couple chapters is enough, since the rest is determined to get you to Godel's theorem which they usually don't need.

It's not for non-mathematicians especially, but for those who need to be able to read and manipulate first-order classical logic.

https://www.amazon.com/Mathematical-Logic-Dover-Books-Mathematics/dp/0486264041

johncarlosbaez, (edited ) to random
@johncarlosbaez@mathstodon.xyz avatar

How can we save the wonderful wildness of nature?

In 2017, Douglas Eger founded Intrinsic with the goal of creating "natural asset companies" (NACs). The idea is that a landowner works with investors to create a NAC that licenses the rights to the ecosystem services the land produces. If the company is listed on an exchange, a public offering of shares would provide the landowner with a revenue stream.

Sound impractical? Here's what happened. The Rockefeller Foundation kicked in about $1.7 million to fund the effort. In 2021, Intrinsic announced its plan to list NACs on the New York Stock Exchange. They filed an application with the Securities Exchange Commission (SEC).

Then the American Stewards of Liberty, a Texas group that campaigns against conservation measures, picked up on the plan. Through both grass-roots organizing and high-level lobbying, they argued that NACs were a Trojan horse for foreign governments and “global elites” to lock up large swaths of rural America, prevent mining in parks, etc. The SEC started to get tons of criticism.

A group of 25 Republican attorneys general called it illegal and part of a “radical climate agenda.” On Jan. 11 this year, the Republican chairman of the House Natural Resources Committee sent a letter demanding a slew of documents relating to the proposal. Less than a week later, the SEC gave up on allowing NACs.

This makes me think the idea actually might have been practical! Otherwise why work so hard to kill it?

The above is paraphrased from a NY Times article. You can read the whole thing for free here:

https://web.archive.org/web/20240218132637/https://www.nytimes.com/2024/02/18/business/economy/natural-assets.html

weekend_editor, (edited )
@weekend_editor@mathstodon.xyz avatar

@johncarlosbaez

I wonder if putting things like this on the market would somehow be exploited by hedge fund/private equity types in an undesirable way.

Perhaps I'm too cynical.

But I don't think so.

cstross, to random
@cstross@wandering.shop avatar

Infosys enjoyed a boom in UK government invoices in 2023

Tech services biz Infosys enjoyed a 49 percent increase in its invoices from the UK government for 2023, according to research figures.

(Infosys was founded by the Prime Minister's father-in-law and his wife is a major shareholder.)

https://www.theregister.com/2024/02/13/infosys_uk_government_contracts/

weekend_editor,
@weekend_editor@mathstodon.xyz avatar

@cstross

Am I the only one who thought it would be cool if Infosys got confused for Infocom, and all government services were implemented in Zork?

And ZIL was the only programming language allowed for government purposes?

cstross, to random
@cstross@wandering.shop avatar

MAD CONSPIRACY THEORY: JFK, as POTUS, had access to the FTL drive of the saucer that crashed at Roswell.

FTL implies time travel. So it's a time machine.

He saw his 2nd term would be dominated by the Vietnam War turning to shit, so he assassinated himself to trigger a grandfather paradox then bumped off Elvis and replaced him then lived happily ever after.

This is as plausible as the current Taylor Swift nonsense, amirite?

weekend_editor,
@weekend_editor@mathstodon.xyz avatar

@cstross

You need to involve RFK and QAnon to maximize the number of exploded heads.

johncarlosbaez, (edited ) to random
@johncarlosbaez@mathstodon.xyz avatar

My mom grew up in rural Michigan in the 1940s, when almost everyone was a churchgoer and people used lots of euphemisms when swearing. So instead of the exclamation "God!" they'd say "gosh!" Instead of "hell!" they'd say "heck!" And though the exclamation "holy Moses!" isn't taking the Lord's name in vain, even that sometimes got watered down to "holy moly!"

So I loved this joke:

Q: What happens to people who don't believe in gosh?
A: They go to heck.

But yesterday something else happened to remind me of all this. A computer science student whose native language is not English wrote that they were going to a "heckathon". That's a great name for a hellishly hectic programming event!

weekend_editor,
@weekend_editor@mathstodon.xyz avatar

@johncarlosbaez

And you're SURE they weren't attending a prolonged ceremony for Heka, the Egyptian deity of magic & medicine?

https://en.wikipedia.org/wiki/Heka_(god)

gutenberg_org, to books
@gutenberg_org@mastodon.social avatar

English author, poet, mathematician and photographer Charles Dodgson was born in 1832. Within the academic discipline of mathematics, Dodgson worked primarily in the fields of geometry, linear and matrix algebra, mathematical logic, and recreational mathematics, producing nearly a dozen books under his real name. To promote letter writing, Dodgson invented "The Wonderland Postage-Stamp Case" in 1889.

Books by Lewis Carroll at PG:
https://www.gutenberg.org/ebooks/author/7

Photo of Alice Liddell taken by Lewis Carroll (1858) Lewis Carroll - The Life and Letters of Lewis Carroll, p. 80 Alice Liddell as a beggar-maid (from the story of Cophetua). Supposed tear hole or ink-blot in photo digitally removed. This was first published in Carroll's biography by his nephew: Collingwood, Stuart Dodgson (1898) The Life and Letters of Lewis Carroll, London: T. Fisher Unwin, pp. p. 80
Title page of The Life and Letters of Lewis Carroll (Rev. C. L. Dodgson) by Collingwood, which is available at PG: https://www.gutenberg.org/ebooks/11483

weekend_editor,
@weekend_editor@mathstodon.xyz avatar

@gutenberg_org

Well, if we're going to talk about Dodgson on letter writing and the Wonderland stamp case, the we simply cannot omit:

C. Dodgson, "Eight or Nine Wise Words About Letter-Writing", 1890.

Dodgson being Dodgson, it also contains what he apparently thought to be relevant advice upon the situation when one's mother-in-law turns into a gyroscope. (Dodgson not being married, this must be regarded as rather theoretical, not practical, advice.)

https://classic-literature.co.uk/lewis-carroll-eight-or-nine-wise-words-about-letter-writing/

https://en.wikipedia.org/wiki/Eight_or_Nine_Wise_Words_about_Letter-Writing

cstross, to random
@cstross@wandering.shop avatar

I am currently sleeping 11-12 hours overnight … and needing a 2 hour afternoon nap as well. And that's WITH the SAD lamps!

weekend_editor,
@weekend_editor@mathstodon.xyz avatar

@cstross
@Meaningness

David Chapman has a couple web pages documenting his experiments with SAD lamps.

The conclusion seems to be that the effect increases with increasing lux, and the usual SAD lamps aren't nearly bright enough.

He wrote about using much brighter lamps, including some stadium LED lighting available on Amazon.

Here's a sample, just to pique curiosity:

https://meaningness.com/sad-light-lumens

lilithsaintcrow, to random
@lilithsaintcrow@raggedfeathers.com avatar

“‘Impossible’ to create AI tools like ChatGPT without copyrighted material, OpenAI says.”

Then fucking pay for it honestly or fucking perish, you fucking grifters.

https://www.theguardian.com/technology/2024/jan/08/ai-tools-chatgpt-copyrighted-material-openai

weekend_editor,
@weekend_editor@mathstodon.xyz avatar

@spacelizard @fpbhb @lilithsaintcrow @cstross

The profitability argument was previously used to defend child labor, and slavery before that.

I'm comfortable saying if your business depends on slavery, child labor, pollution, or theft of the work of others... then your business is bad and should not exist. You can learn a better business model.

  • All
  • Subscribed
  • Moderated
  • Favorites
  • •
  • Leos
  • slotface
  • ngwrru68w68
  • modclub
  • normalnudes
  • everett
  • rosin
  • cubers
  • magazineikmin
  • thenastyranch
  • DreamBathrooms
  • Youngstown
  • osvaldo12
  • InstantRegret
  • JUstTest
  • kavyap
  • mdbf
  • anitta
  • ethstaker
  • cisconetworking
  • Durango
  • tacticalgear
  • khanakhh
  • megavids
  • provamag3
  • tester
  • GTA5RPClips
  • lostlight
  • All magazines
    •