⏚ Antoine Chambert-Loir

https://mathstodon.xyz/@antoinechambertloir

- Apprentice mathematician (Professor at Université Paris Cité),
interested in algebraic geometry, number theory and (more recently) proof formalization
- Apprentice musician (drums, tabla, cajon; previously Diderock; a rarely meeting jazz trio/quartet).
- Apprentice progressist.

In love with math, jazz, indian music and poetry. ⏚

Web: https://webusers.imj-prg.fr/~antoine.chambert-loir/index.xhtml
Blog: https://freedommathdance.blogspot.com
Orcid: https://orcid.org/0000-0001-8485-7711
BookWyrm: achambert-loir@bw.heraut.eu

Public Key

npub1j7kvjvr8hkvwzvs889v2ycwpvh0n7qns5ha84xzh2r36zrakk4wsc6a4dl

NIP-05 Address

antoinechambertloir_at_mathstodon.xyz@momostr.pink

Profile Code

nprofile1qqsf0txfxpnmmx8pxgrnjk9zv8qkthelqfc2t7n6npt4pcapp7mt2hgprpmhxue69uhhyetvv9ujumt0d4hhxarj9ecxjmntqywhwumn8ghj7mn0wd68yttsw43zuam9d3kx7unyv4ezumn9wspfgr54

Publishing to

relay.momostr.pink

nostr-pub.wellorder.net

yabu.me

Author Public Key

npub1j7kvjvr8hkvwzvs889v2ycwpvh0n7qns5ha84xzh2r36zrakk4wsc6a4dl

Show more details

Last Notes

2026-10-25 11:34:51 UTC

- reply

Well, Scott Armstrong warned us that there would be two other batches of theorems, some colleagues say we're irresponsible not to embrace the hype, etc.

2026-10-25 11:33:29 UTC

- reply

9000 signatures déjà !
https://petitions.assemblee-nationale.fr/initiatives/i-7415

2026-10-25 11:32:57 UTC

RE: https://mastodon.social/@lemonde/117501278148377955
Rappel qu'alors que les politiques nous présentaient Parcoursup comme censé régler tous les problèmes de l'accueil de l'enseignement supérieur, personne ne dit que supprimer Parcoursup serait la solution magique. La pétition en cours sur le site de l'Assemblée nationale formule quatre exigences, les trois premières,
- suppression de Parcoursup
- abrogation de la loi ORE
- abrogation de la loi Pénicaud
étant négatives, mais la quatrième,
- investissement massif dans l'Éducation nationale et l'enseignement supérieur
ne saurait être oubliée.

2026-10-25 10:33:40 UTC

- reply

I'd like to hear a rationale for the expectation that the results that didn't fall last month would fall after, say, a multiplication by 10 of the computing power.

2026-10-25 10:31:34 UTC

- reply

part of the trillion of dollars invested in AI could fund MathSciNet and ZbMath for several decades without much impact on AI.

2026-10-24 10:22:37 UTC

- reply

Du coup, il faudrait se dépécher d'installer reticulum, non ?

2026-10-22 06:24:39 UTC

- reply

bonnes vacances !

2026-10-16 21:29:14 UTC

- reply

Waouh! A possibly stupid question (sorry…) : in retrospect, would this work have been possible without the concept of univalence, or were you already leaning to it in some way?

2026-10-16 21:10:29 UTC

- reply

That's a lot of work! I am impressed!
Now, a possibly difficult or insane question: should you have, say, 4 or 5 hours to share this work to colleagues who are willing to hear about it, which unfortunately is even rare, what would you suggest?

2026-10-14 16:43:14 UTC

- reply

Mais j'ai répondu un peu à côté de ta question. Désolé. Apparemment, L'offensive est un collectif de groupements, et Onestlatech en est sinon adhérente, du moins proche, si je veux bien croire cet article qui les met côte à côte en les dézinguant joyeusement.
(NB. Je ne connais pas Nicolas Casaux.)
https://www.partage-le.com/2025/12/18/onestlatech-l-antifascisme-des-cadres-de-la-megamachine-par-nicolas-casaux/

2026-10-14 16:38:50 UTC

- reply

dans le fil d'hier matin, il y avait eu cette question de la zapatisation…

2026-10-14 16:34:28 UTC

- reply

oh non, on s'est déjà engueulés avec eux avant-hier matin (ou c'était hier ?)…

2026-10-10 18:59:00 UTC

- reply

17. India #JohnColtrane100
While this tune, “India”, definitely makes us leave latin jazz, there are obvious common points with yesterday's “Olé”. This is Coltrane experiencing different musical scales and different traditions; moreover, we have two basses, Reggie Workman and Jimmy Garrison, we also have Eric Dolphy on bass clarinet, Elvin Jones on drums, McCoy Tyner on piano, as well as Garvin Bushell on either oboe, and Ahmed Abdul-Malik on oud.
This tune was recorded live at Village Vanguard in November 1961. While one version was released in 1963, for the album *Impressions*, the one I chose comes from *The other Village Vanguard tapes* that were published in 1977, ten years after Coltrane's death. Indeed, the starting notes clearly show the will of the musicians to recreate the sound of hindustani music, using bass and oud to suggest the sound of a sitar and of a tanpura.
https://www.youtube.com/watch?v=5yIYo0qgBfU

2026-10-09 19:23:36 UTC

- reply

17. Olé #JohnColtrane100
This tune is not really latin, but its Spanish influence is definitely a step aside from the standard harmonies of jazz. It is in fact inspired from a traditional Spanish song, El Vito. The fast 3/4 tempo avoids the waltz feel by playing long assertive rhythmic figures by McCoy Tyner on piano and Elvin Jones on drums: two bars, most often, but even four bars for the bass introduction.
John Coltrane plays soprano saxophone, Freddie Hubbard, trumpet, and Eric Dolphy plays flute under the pseudonym of George Lane, because he had a contract for a different disc company.
Can you spot that there are two bass players here ? — Art Davis and Reggie Workman, one of them choruses with a bow around 9:00, while the other one goes on plucking. And then they change, the chorus is done with plucked chords while the other one plays the melody on the bow. And then both improvise together!
https://www.youtube.com/watch?v=qUSMON8eO9Y

2026-10-08 21:03:45 UTC

- reply

16. El Toro Valiente #JohnColtrane100
We continue our exploration of latin jazz in Coltrane's music with this tune recorded with Art Blakey's big band.
It is not so easy to track the saxophone player, maybe he's the one who plays the short chorus at the end — am I even able to distinguish a tenor saxophone from an alto one ? — but this short energetic piece is worth listening to.
https://www.youtube.com/watch?v=iZW9smNv1j8

2026-10-07 21:20:49 UTC

Statement of the Association for Human Mathematics about the release of hundreds of proofs by OpenAI
https://www.ahmath.org/statements

2026-10-07 16:05:41 UTC

- reply

15. Bahia #JohnColtrane100
Coltrane's music has sometimes be released much later than the recording date. Today's piece, “Bahia”, was recorded in 1958 but gave its title to an album that appeared in 1965, at a time where Coltrane had published much more personal music with his classic quartet. To the point that jazz critics of that time from *Record Mirror* wrote that “Talk of Coltrane being ‘anti-jazz’ must be dispelled by this little lot.”
This is a composition by Ary Barroso, and the rhythmic flow is sustained by a short motive on the piano and bass and an omnipresent cascara on the ride cymbal that is announced right away.
Red Garland is on piano, Paul Chambers on bass, and Art Taylor on drums. The album cover shows Coltrane on soprano saxophone but he plays tenor saxophone on that album.
https://www.youtube.com/watch?v=0yZq_VVB89E

2026-10-06 18:46:25 UTC

- reply

14. Eclypso #JohnColtrane100
I continue my investigation of latin jazz influences on John Coltrane's music. This tune, “Eclypso”, is a composition of the pianist Tommy Flanagan recorded in 1957 but only published in 1959 after the end of Coltrane's contract with Prestige.
The album is called *The Cats*, and features four musicians on the cover: Tommy Flanagan (piano), John Coltrane (tenor saxopohone), Kenny Burrell (guitar) and Idrees Sulieman (trumpet). With those four cats, we have Doug Watkins on bass and Louis Hayes on drums.
The song has an AABA form, where the A is played in a latin style while the B is played swing in ‘“trading fours” style: two bars with the full band, two bars of drum solo. The choruses are also played swing.
(Twenty years later, Flanagan would offer a second recording of that tune with George Mraz on bass and Elvin Jones on drums.)
https://youtu.be/agQPRYup0Ck

2026-10-05 17:10:20 UTC

- reply

non, je n'avais pas vu ! Tu te rappelles quel livre ?

2026-10-05 17:01:03 UTC

- reply

13. Blue Calypso #JohnColtrane100
In 1957, Coltrane's drug addiction led Miles Davis to end up his collaboration with the saxophonist, despite the growing reputation of the later, and the recording company Prestige took its chance to have him on several sessions as a side man.
Here he plays under the lead of pianist Mal Waldron, with Hackie MacLean on alto saxophone, Bill Hardman on trumpet, Julian Euell on bass, and Art Taylor on drums.
The theme, “Blue Calypso”, is a straightforward theme in the genre that had been made popular by Sonny Rollins's “St Thomas” (1956), with an extraordinary Max Roach on drums. Jazz critics found the recording quite uninteresting, but for Coltrane's choruses. Probably with a definite bias for the instrument, note the role of the drums in that tune, the way Taylor starts the dance without any other accompaniment, the drum calls that he makes before Coltrane's chorus, as well as his own short chorus at 7:28, to the fade-out ending.
https://www.youtube.com/watch?v=KxsWforP2eA

2026-10-04 16:05:04 UTC

- reply

12. Afro Blue #JohnColtrane100
This is a song that percussionist Mongo Santamaria composed (and recorded) in 1959. Then it had a 12/8 rhythm with a clave playing 4 beats of three pulsations per bar, while the snare drum plays 3 beats of four pulsations.
In October 1963, John Coltrane played it live at Birdland and it gave the tune a totally different feeling: now the main rhythm is that of the syncopated jazz waltz which is reinforced by the piano playing 2 notes per bar, and this is the style that is chosen for most standard transcriptions of that tune.
Also: while Santamaria played the melody in a smooth and mellow manner, Coltrane attacks it with a shrieking soprano saxophone.
I suppose that the music started that night, the audience of Birdland (where you can eat or chat) must have stopped anything they were expecting to do because the musicians didn't leave a moment of peace during the 11 minutes of this ferocious afro blue waltz.
https://www.youtube.com/watch?v=HIH3fNUsbnA
https://media.mathstodon.xyz/media_attachments/files/117/383/475/443/797/042/original/c7201632510f859d.png

2026-10-03 16:58:33 UTC

- reply

11. The Night Has A Thousand Eyes #JohnColtrane100
The latin rhythms of yesterday's tune make me wish to explore the latin jazz aspects of Coltrane's music. It won't be so easy to select tunes in Coltrane's own repertoire because he didn't record so many latin themes.
This song, The Night Has A Thousand Eyes, was originally composed 1948 for a movie.
Don’t whisper things to me you don’t mean
For words deep down inside can be seen by the night
The night has a thousand eyes
And it knows a truthful heart from one that lies
Coltrane's quartet transforms the song by playing it very fast, the first section in afro/cuban style, while the second one reverts to swing. (McCoy Tyner is on piano, Elvin Jones on drums, Steve Davis on bass.)
A few years, in 1962, Sonny Rollins would record a version of that song in the style of a mellow bossa nova.
Although released in 1964, Coltrane's album *Coltrane's Sound* was recorded in 1960, and I don't know how both saxophonists influenced each other.
https://youtu.be/ioPRCwpB1Ak

2026-10-03 14:45:43 UTC

- reply

quel est le nom du restaurant ?

2026-10-03 10:53:26 UTC

- reply

profite bien de ce beau moment ! En plus, ça a l'air si calme…

2026-10-03 10:28:08 UTC

- reply

où est-ce ?

2026-10-02 17:15:28 UTC

- reply

10. Wise One #JohnColtrane100
Let us keep the album *Crescent* playing and listen to another composition by Coltrane, “Wise One”.
The first part of the tune is a shriekingly sad melody that is sung on saxophone, with a rubato accompaniment by Jones, Garrison and Tyner. Then the musicians switch the mood to a calm but rhythmic latin dance on which Tyner starts a chorus. Coltrane then plays a chorus sometimes flying over the repetitive rhythm, sometimes sustained by it. After the chorus ends low, with a cymbal roll, Coltrane recaps the initial melody and concludes, leaving the final notes for the piano.
https://youtu.be/bXd55keSsdk

2026-10-01 20:48:41 UTC

- reply

9. Lonnie's Lament #JohnColtrane100
A beautiful composition by John Coltrane, recorded in 1964 for the album *Crescent*.
Coltrane plays tenor saxophone and is playing within his classic quartet that he had assembled around 1963: McCoy Tyner is on piano, Jimmy Garrison on bass, an Elvin Jones on drums.
https://www.youtube.com/watch?v=japCux4wals
https://media.mathstodon.xyz/media_attachments/files/117/367/612/080/165/820/original/397c05a18f69f31f.png

2026-09-30 19:23:49 UTC

- reply

8. Aisha #JohnColtrane100
This is a composition of McCoy Tyner — Aisha was the name of his wife. Together with Tyner on piano, we have Reggie Workman is on bass, Elvin Jones on drums. Three horns accompany this ballad. On tenor saxophone, Coltrane exposes the theme and takes a first chorus. His is followed by Freddie Hubbard on trumpet and then by a quite free Eric Dolphy on alto saxophone. Tyner also takes a chorus before Coltrane ends it beautifully.
The piece is written in 4/4, the structure is AABA, but the very slow melody is sustained by the rhythmic section with some interesting unbalance : the rhythmic motive on the piano is syncopated, and using brushes, Jones sometimes plays the 4 beats of the bar, and sometimes plays 6 triplets.
https://www.youtube.com/watch?v=wi7X-7QDyfE&

2026-09-29 20:54:52 UTC

- reply

7. Everytime We Say Goodbye #JohnColtrane100
Featured in *My Favorite Things* album, this is a 1944 song by Cole Porter that Coltrane also recorded in October 1960 with the same quartet as yesterday's theme: Steve Davis on bass, McCoy Tyner on piano, Elvin Jones on bass. Coltrane is on the soprano saxophone, and all musicians play that beautifully sad melody in a very restrained way.
There's no love song finer
But how strange the change from major to minor
Every time we say goodbye
https://www.youtube.com/watch?v=Zllu7zlL6DM

2026-09-28 19:12:02 UTC

- reply

6. Central Park West #JohnColtrane100
This is a composition by John Coltrane himself, recorded in October 1960 but only released in the 1964 album *Coltrane's Sound*. The recording had taken place in the studios of the Atlantic disc company, but when Coltrane left the label for Impulse, he abandoned recorded material that the company released, apparently without Coltrane's approval.
Nevertheless, the recording is very good. This is a quartet with McCoy Tyner on piano, Elvin Jones on drums and Steve Davis on bass. John Coltrane plays soprano saxophone, an instrument that he had been offered by Miles Davis when they were touring together in Europe.
https://www.youtube.com/watch?v=euE0wyRmyZs
https://media.mathstodon.xyz/media_attachments/files/117/350/245/410/899/519/original/9901e4a08298c54c.png

2026-09-27 16:18:36 UTC

- reply

5. Lush Life #JohnColtrane100
Yesterday, we listened to a version of “Lush Life” with vocals by Johnny Hartman, but Coltrane had already recorded that tune in 1958, in a quintet version.
The song starts with Coltrane on tenor saxophone, Red Garland on piano, and Paul Chambers on bass. Louis Hayes on drums enters only after the first exposition of the theme. Then Garland starts a chorus. Then Donald Byrd, on trumpet, starts playing a chorus.
I would like to understand what brought him in that recording session where they apparently only recorded that tune! An explanation might be that Red Garland's trio (with Art Taylor on drums and George Joyner on bass) made four recordings in the end of 1957 for which he invited both Coltrane and Byrd. So maybe Coltrane liked the quintet mood and simply returned the invitation.
In his *Downbeat* review, Dave Rickert notes that Byrd helps pushing the tempo a little bit. On the other hand, he goes back to the initial tempo at the end of his chorus. Anyway, this gives Coltrane another opportunity for a short chorus. Then both Byrd and Coltrane play the melody together and conclude the piece using descending harmonies.
https://www.youtube.com/watch?v=-JqTFAIiAn4

2026-09-26 16:19:26 UTC

- reply

4. Lush Life #JohnColtrane100
In 1963, Coltrane recorded an album of ballads with singer Johnny Hartman with what is now understood as his classic quartet: McCoy Tyner on piano, Jimmy Garrison on bass and Elvin Jones on drums.
This piece, Lush Life, had been composed by Billy Strayhorn between 1933 and 1936, but it was first performed in 1948, by Duke Ellington's band with which Strayhorn has composed and arranged many classic tunes for almost all of his life.
It starts with Hartman singing only accompanied by piano; the lyrics are full of unfilled promises and both Tyner and Hartman play it with both tension and sensitivity.
Life is lonely again
And only last year everything seemed so sure
Now life is awful again
A troughful of hearts could only be a bore
and Garrison and Jones enter while Coltrane simply responds after some phrases of Hartman.
And there I'll be
While I rot with the rest
Of those whose lives are lonely, too.
The lyrics could almost end there, and Coltrane starts his own rendition of the song, staying very close to the melody. The sound of the quartet is immediately more dense, with a double time feel; Jones, on brushes, adds a lot of musicality to the interpretation. When the riveted ride cymbal ends the final bar of the melody, Hartman comes back and sings the last verse once more, slows the tempo to the end.
https://youtu.be/q_xbjOCPgkU
https://media.mathstodon.xyz/media_attachments/files/117/338/241/582/013/577/original/8fa75692c1161ed3.png

2026-09-25 18:41:44 UTC

- reply

3. I's easy to remember (and so hard to forget) #JohnColtrane100
As promised yesterday, I continue with a ballad composed by Richard Rogers with lyrics by Lorenz Hart that was written for the movie *Mississippi*. This piece, feature in the album *Ballads*, was recorded in December 1961 with McCoy Tyner on piano, Reggie Workman on bass, and Elvin Jones on drums. The next year, Reggie Workman would be replaced by Jimmy Garrison for the rest of the recording that took place one year later in November/December 1962.
This is the kind of album that those high skilled musicians can do. Apparently, they had never played those tunes before, they discussed a little bit on the way they would play them, what chord substitutions they would choose, rehearsed no more than half of an hour, and make an un-cut version which is fit for the album.
Please appreciate the final seconds of the piece, when Jones summons the thunder and the piece comes to its end.
https://www.youtube.com/watch?v=RZ_NiGFIUGc
https://media.mathstodon.xyz/media_attachments/files/117/333/141/398/846/458/original/57bf96f4f0d092fb.png

2026-09-24 17:48:37 UTC

- reply

2. Naima #JohnColtrane100
I will stay in the ballad mood for some days, probably because the ballads Coltrane played have some immediate reach on me, while the faster pieces, and I'm not talking about the furious ones, sometimes keep me distant.
Here is that piece, Naima, composed in 1959 for his wife. The melody is simple, the harmony does not vary much, suggested in a simple way by Wynton Kelly's chords on the piano, each chord lasting one full bar, and by a striking bass line that leaves engaging silences. The piano chorus hides the sophistication of its modulations. This is a beautiful piece of love.
Preparing this post and reading Wikipedia's page on that composition, I discovered a poem by Kamau Brathwaite which I will simply copy here.
“Naima” for John Coltrane
Propped against the crowded bar
he pours into the curved and silver horn
his old unhappy longing for a home
the dancers twist and turn
he leans and wishes he could burn
his memories to ashes like some old notorious emperor
of rome, but no stars blazed across the sky when he was born
no wise men found his hovel, this crowded bar
where dancers twist and turn
holds all the fame and recognition he will ever earn
on earth or heaven, he leans against the bar
and pours his old unhappy longing in the saxophone.
https://www.youtube.com/watch?v=bPAC6zt_1ZM

2026-09-23 19:10:00 UTC

- reply

#JohnColtrane100
1. Violets for your furs
I'll start this thread with a beautiful ballad from the first recording that John Coltrane recorded as a leader. That song, written by Matt Dennis in 1941 was first recorded by the Tommy Dorsey orchestra, sung by Frank Sinatra. The lyrics by Tom Adair apparently referred to Lana Turner.
You smiled at me so sweetly
Since then one thought occurs
That we fell in love completely
The day that I bought you violets for your furs
Red Garland is on piano, Paul Chambers on bass and Albert “Tootie” Heath is on drums
https://www.youtube.com/watch?v=uQo1-CzyNuM
https://media.mathstodon.xyz/media_attachments/files/117/321/912/512/980/244/original/ca49a3d1c2a72977.png

2026-09-23 19:01:25 UTC

#JohnColtrane100
On this day, 100 years ago, John Coltrane was born. His musical life has been much too short, from his debuts in 1945 just after he came back from the Navy, to his death in 1967, just aged 40, but offered a variety a musical styles of jazz music, from be bop and then a fruitful companionship with Miles Davis, to the quartet he led, to the birth of free jazz, sometimes forceful and noisy, sometimes lyrical as a prayer.
Let's discover his music for the upcoming 100 days.
You can follow that hashtag #JohnColtrane100. The pieces will also be featured on the independent webpage
https://webusers.imj-prg.fr/~antoine.chambert-loir/Musique/JohnColtrane100.html
as well as on the YouTube playlist
https://youtube.com/playlist?list=PLAflVSKN2kc0
https://media.mathstodon.xyz/media_attachments/files/117/321/953/198/612/369/original/29d21d022c342554.png

2026-09-10 08:11:47 UTC

#AI #Math
The most recent case of the finite time solution of the forced Navier-Stokes equation has been discussed a lot in the recent days, on this network and in the mainstream press,
but I will try to give a perspective on what happened recently.
This is prompted by a friend informing me this morning of two annoying additional stories that taint the recent public success of AI generated proofs and will hopefully lead the mathematical community to address the question of predation and plagiarism.
However, I won't address here the even more important questions posed by AI companies beyond merely quoting some of them:
- its energetic cost;
- its impact on rare earths and other matters;
- its direct and immediate consequences on the ecosystem (water heating, pollution, land waterproofing, etc.);
- its impact on the work market, in particular those prompted by (actual or alleged) productivity gains;
- its impact on learning, cognition, etc.;
- its impact on the economic structure, tremendous announced gains/losses of companies that look like a Ponzi scheme;
- its impact on the power structure, who owns what, who gives access to what;
- its impact on the political structure, the surveillance capitalism, etc.
These are fundamental issues that should be addressed more clearly in the mainstream medias, and that, in my view, suffice to object to the existence of AI.
Anyway, what does it mean for academic math?
1/n

2026-09-09 12:53:43 UTC

Tiens, wikisource conclut son recueil de citations de Sam Altman par une citation d'Aaron Swartz,
« You need to understand that Sam can never be trusted. He is a sociopath. He would do anything. »
https://en.wikiquote.org/wiki/Sam_Altman

2026-09-06 17:47:33 UTC

- reply

je risque de proposer qu'on regarde Jay Kelly...

2026-09-06 17:34:33 UTC

- reply

quel film ?

2026-08-31 02:38:45 UTC

- reply

i was struck by the fact that you start from non-standard quaternions and not non-standard complex numbers!

2026-08-29 21:29:15 UTC

- reply

où est-ce que ça aura lieu ?

2026-08-29 18:26:26 UTC

- reply

c'est un unowhy13? On en a un à la maison que je voudrais aussi passer en linux...

2026-08-26 11:48:22 UTC

- reply

I had no doubt about your position about AI/LLMs. I just can't imagine AI-robots doing math without any humans interacting with them and imposing their speed on the mathematical humanity.
@npub1mnf…uvj0 @npub1juz…njnt

2026-08-26 11:32:23 UTC

- reply

In a world where math is mostly a professional activity, governed by academic norms of productivity that fuel the professional hierarchy (from grad student to PhD to post doc to assistant professor to professor to emeritus), “letting math go its own way” doesn't seem to make much sense to me. It will necessary follow the way of productivity.

2026-08-22 17:48:26 UTC

- reply

Qui aurait pu prévoir?

2026-08-22 08:26:07 UTC

- reply

Nice that people are showing it in the US!

2026-08-22 08:11:18 UTC

- reply

it always has been viewed as a metaphor for totalitary regimes that had plagued mid-20tg century Europe. It is so sad that this way of denunciation has not reached its goal.

2026-08-04 17:27:24 UTC

- reply

I meant it is easy to “understand some of the consequences of the Jacobian conjecture being false…” (assuming the thread mentioned some other proof/disproof).

2026-08-04 16:03:11 UTC

- reply

This case is easier. The Jacobian conjecture has little interesting consequences, and neither has its disproof. (The Dixmier conjecture is a noncommutative variant, and the one on moments of random variables seems pretty much unmotivated.)
@npub1gve…sduy @npub1nf4…nqe4

2026-08-03 17:04:41 UTC

- reply

Well, there's this quote from @npub1vzx…ncd7 which I find fairly disturbing, if not problematic.
“A few days earlier I had got an email from a professor in the maths department here at Imperial, expressing surprise that some of our graduate students were paying $200 per month to access models such as Sol and Fable. He said that he thought that these people were crazy. I did not immediately respond. But after meeting with Andrew I emailed the professor back and told him that in my opinion, any PhD student who was not paying $200 per month to access these tools was crazy. In fact during the workshop I learnt from Harvard PhD student Bryan Wang that Harvard were already giving free Fable access to all PhD students, post-docs and faculty at Harvard.”
https://xenaproject.wordpress.com/2026/07/20/human-mathematicians-are-being-outcounterexampled

2026-08-02 10:15:14 UTC

- reply

one thing that might be wrong in the math as chess metaphor is that math goes on, whatever theorems mathematicians prove. When a conjecture is proved, new questions emerge, when it is disproved, a quest for better intuitions has to start. And sometimes, mathematicians simply create a new field.

2026-07-29 07:38:23 UTC

- reply

je l'ai trouvée extraordinaire !

2026-07-29 06:31:53 UTC

- reply

quelle expo?

2026-07-03 16:02:55 UTC

Cognitive dissonance
https://media.mathstodon.xyz/media_attachments/files/116/856/883/646/400/412/original/1c53716d4e30704f.jpg

2026-07-01 12:39:31 UTC

The science preprint archive arXiv is leaving Cornell University to become an independent nonprofit organization.
https://blog.arxiv.org/2026/06/30/arxivs-next-chapter/

2026-06-28 15:06:40 UTC

Ce que je trouve lunaire dans l'attaque d'Emmanuel Lechypre contre Christophe Cassou, ce n'est pas la phrase d'introduction, « vous n'avez pas été convaincants », c'est la suite : « vous avez été abstraits », en prenant comme exemple « 2°C de réchauffement en 2050 ».
Emmanuel Lechypre est chroniqueur économique. Il manipule toute la journée, dans ses interventions, des chiffres de la dette, du chômage, du PIB, qui, eux, sont des chiffres abstraits, construits, et presque artificiels.
2°C, c'est un chiffre concret, qui correspond à une réalité physique immédiate.
Je suppose qu'il sait ce que signifie 2°C de plus dans le studio de radio ou de télévision.
Je suis certain qu'il sait ce que signifie 2°C de plus pour sa température corporelle.
Maintenant, s'il ne sait pas l'impact de 2°C sur la croissance des plantes, qu'il se renseigne ou qu'il se taise.
S'il ne sait pas l'impact de 2°C sur l'habitat des animaux sauvages, et des humains, qu'il se renseigne ou qu'il se taise.
S'il n'est pas capable d'envisager les perturbations de ces impacts sur la société toute entière, humaine et non humaine, sauvage et civilisée, qu'il se renseigne ou se taise à jamais.

2026-06-27 21:14:32 UTC

- reply

en fait, un arbre est tombé sur une caténaire et je n'ai pas bougé depuis mon pouët précédent.

2026-06-27 19:26:32 UTC

- reply

Je suis dans le train qui vient de Bordeaux, en pleine Beauce, à 30 minutes du sud de Paris. Dans la direction du nord, le ciel est tout noir, et on vient de voir un bel éclair. (Maintenant, le train est arrêté, c'est ballot, parce qu'il y a une coupure d'électricité à cause de l'orage.)

2026-06-13 14:08:19 UTC

- reply

je ne pourrai pas y être, je surveille un examen à la même heure...

2026-06-13 13:30:30 UTC

- reply

Vu le mot « intrusion » dans le résumé et l'opposition farouche du premier intervenant aux LLMs, je ne pense pas que cette journée sera sous le signe du côté obscur de la force.

2026-06-13 09:53:09 UTC

18 juin - Journée IA et apprentissage des mathématiques, organisée par la Société mathématique de France.
À l'occasion de son Assemblée Générale du 18 juin (à Paris), la SMF, en partenariat avec l'UFR de mathématiques de Sorbonne Université, organise une matinée table-ronde / débat autour de l'intrusion de l'IA dans les enseignements et nos cursus de maths au niveau universitaire. Il sera question aussi de faire le point sur l'usage des assistants de preuves (de type Lean) dans nos enseignements.
https://smf.emath.fr/evenements-smf/journee-ia-apprentissage-mathematiques

2026-06-09 16:18:50 UTC

- reply

Idem. Et mes week-ends sont hélas trop souvent trop studieux...

2026-06-08 08:54:38 UTC

- reply

pour une mauvaise audition, ou pour les acouphènes ? Un orl m'en avait prescrit, mais l'idée de porter un SMIC par oreille sans garantie de succès, ça m'avait un peu refroidi.

2026-06-01 07:42:08 UTC

Découvert via le Khrys'presso de ce matin de @npub17kp…gmwf cet texte à propos de Linus Torvalds qui « siffle la fin de la récré de l'IA ». En gros, y'en a marre de ces rapports générés automatiquement qui prennent des plombes à évaluer.
https://goodtech.info/linus-torvalds-ia-code-slop-linux-noyau-7-1-rc5/

2026-05-15 20:12:32 UTC

- reply

je crois que ca sera bien.

2026-05-15 20:10:41 UTC

- reply

je serais plutôt pour reporter, je n'ai pas l'habitude de randonner dans ce coin et je voudrais que cette expérience soit vaguement agréable.

2026-05-10 13:15:35 UTC

- reply

Bonjour, je suis très tenté : est-ce que c'est une balade difficile lorsqu'on n'a aucun entraînement ?

2026-05-09 11:26:27 UTC

- reply

I know that this is not the trend, but I have never been willing to propose PhD topics that would be at the level of the average thing somebody can do with the tools that are on the market, because this already requires a substantial time of the PhD project and in was already clear — at least in my field — that this would wouldn't be sufficient to get hired. So I manage to suggest questions that could be damn hard, but for which several approaches are likely to tell something at least mildly interesting, and let's see what comes out of it.

2026-05-05 13:44:10 UTC

- reply

👏 L'avertissement dit la chose suivante :
Il y a 1 + n fichiers de configuration de polices, l'un pour le système, et les autres pour les utilisateur·ices qui l'ont réclamé via `updamap-user`. Si d'aventure tu ajoutais une police pour tout le monde via `updmap-sys…`, il faudrait que tu fasses le même ajout pour toi-même via `updmap-user…` parce que ton fichier ne serait pas mis à jour par la commande `updmap-sys`.

2026-05-05 12:54:53 UTC

- reply

On dirait qu'il manque l'étape `updmap`, la moins drôle.
Donc, en ligne de commande, entrer :
updmap-user --enable Map=emerald.map
@npub1hnl…k2tt @npub12u6…q3pu

2026-05-05 10:38:38 UTC

- reply

(Je suis tombé dans TeX quand j'étais petit. Ce README est assez pourri.)
Pour savoir où créer le répertoire texmf, tu peux entrer (en ligne de commande) :
kpsewhich -var-value=TEXMFHOME
et créer le répertoire en question s'il n'existe pas déjà. (Probablement $HOME/texmf)

2026-05-05 10:20:12 UTC

- reply

L'installation est un peu casse-pieds.
Sur https://ctan.org/pkg/emerald tu peux télécharger un .zip, là, il faut le décompresser et — malheureusement — faire ce qui est indiqué dans le fichier README (qu'il faut lire, c'est pas comme dans Alice au pays des merveilles).
En particulier, les répertoires emerald/fonts et emerald/tex doivent être copiés dans le répertoire TeX soit standard, soit local ($HOME/texmf), et ensuite il faut lancer la commande `mktexlsr` pour en informer le compilateur TeX. (Je voudrais bien te donner la ligne de commande pour faire ça sans effort, mais je ne le puis,
l'idée c'est que emerald/tex/latex/emerald/* est copié dans texmf/tex/latex/emerald, et c'est encore plus pénible pour emerald/fonts.) Tu peux faire de même avec emerald/doc
Il y a enfin le répertoire emerald/other qui contient les polices au format TrueType et je ne sais pas comment elles sont gérées sur Debian.

2026-05-02 17:49:53 UTC

- reply

I'm no economist but my answer to those questions is twofold:
* What sense does it make to compare two quantities that are not meant to be compared? Why 100% as a reference? Why not 140% ? What if other productions of values are added to GDP?
* Do you have an idea of the private debt ratio to GDP?

2026-04-26 16:45:16 UTC

- reply

there's a slight (probably more philosophical than mathematical, at least now) difference between sets and groups, that the former were invented (as a formal theory) to furnish foundations for mathematics, at a time when it was inconceivable that not everything could be proved. “Wir wollen wissen, wir werden wissen!” In netherlands, mathematics is Wiskunde, the science of knowledge.

2026-04-20 15:53:14 UTC

- reply

la 14?

2026-03-01 17:27:54 UTC

- reply

so what discourse would you recommend?

2026-02-21 19:12:55 UTC

- reply

As said in the nlab page given above, continuous logic is a logic where “truth values” take values in the unit interval (or a bounded interval, or even R) . This doesn't mean at all that 1 would be absolutely true and 0 absolutely false, and all other values somewhere inbetween. In model theory, what truth values allow to do is building predicates, ie, loci where given properties hold. The classical quantifiers are replaced by supremum and infimum (or variants of). In classical logic, type spaces are totally discontinuous; continuous logic gives type spaces which have a richer topological structure and is pretty much adapted to a model theoretic approach to topological group actions, real-valued fields, and even some aspects of diophantine geometry. The main subtlety in the definitions comes from the (achieved) goal of having a compactness theorem.

2026-02-01 23:08:28 UTC

- reply

I had liked it, but that was some 30 years ago.

2026-02-01 22:19:43 UTC

- reply

I guess you have read David Lodge's book with that title!

2026-01-26 10:37:24 UTC

#ICM2026
The Société mathématique de France won't attend the 2026 ICM.
Their message mentions visas, personal safety, martial law, as well as anti-science behaviour and academic freedom.
https://smf.emath.fr/actualites-smf/icm-2026-motion-du-ca

2025-10-25 15:13:57 UTC

- reply

What you wrote is better than what she said, and lacks the contempt: if you're versed in that math, you recognize the stuff, if you don't, maybe you get an idea, and if you don't, the maybe you leave happy with the impression that somebody tried to explain something to you.
@npub1vqh…d49x

2025-10-04 16:19:15 UTC

Philosophy and mathematics: one upcoming conference and one upcoming seminar:
Lead by @npub1tuz…3hxk , Philosophie de la pratique mathématique, at Collège de France, will run every Monday, 14:00-15:00 from Oct 13 to Nov. 17.
https://www.college-de-france.fr/fr/agenda/seminaire/la-philosophie-de-la-pratique-des-mathematiques-10
and Oct 15-17, the joint 2nd Conference of the European Society for the Philosophy of Mathematics & Seventeenth French Philosophy of Mathematics Workshop will take place in Université Paris Cité and Université Panthéon Sorbonne.
https://philmath.eu/2025conference/
#mathematics #philosophy

2025-08-12 13:44:08 UTC

- reply

PhD level ain't what it used to be.

2025-08-09 17:01:01 UTC

- reply

it's actually easier than one might think since the complement of the circle has two connected components. So provided you know your blue curve is connected, you just need to know it doesn't touch the circle and contains one point within the disc.

2025-08-02 18:57:31 UTC

- reply

Fran Lebowitz, *The FL reader*
Michel Faber, *Listen*

2025-07-26 15:32:39 UTC

- reply

All apologies
https://www.youtube.com/watch?v=SEVu1tcUGnc

2025-07-25 03:45:07 UTC

- reply

I'm shamefully ignorant of those things, but what is is isometry group of S_8, as a Riemannian manifold? Or maybe you're ready to relax the kind of diffeomorphisms that act on it?

2025-07-14 13:36:10 UTC

#mastobada
Billie Holiday and Lester Young — I can't get started
I've been around the world in a plane
Settled revolutions in Spain
The North Pole I have charted
But can't get started with you
And at the golf course I'm under par
Metro-Goldwyn wants me to star
I've got a house and a show place
But can't get no place with you
You're so supreme
The lyrics I write of you
Dream, dream, day and night of you
Scheme just for the sight of you
Baby but what good does it do?
https://youtu.be/ZGAvnOSbJ_M

2025-06-01 08:18:25 UTC

- reply

L'hortensia toujours
taillé trop tard, négligé,
fleurit sans rancune.

2025-04-23 21:36:00 UTC

- reply

This book brings many details, anecdotes, on a rich topic, and that gives it some appeal. However, I am extremely doubtful of the mere project to summarize 3000 years of mathematics in one book, the more if it means conflating so many diverse mathematical traditions into one word (analysis). I skimmed through the Leibniz part, and I don't see a clear discussion of what infinitesimals meant to him (a useful fiction, did he wrote), but my reading may have been too hazy. As usual, the question with history is whether we want it to validate our own practices, to color them with the gilded dust of ages, or whether we are truly interested in what people did, why and how they did it.

2025-04-08 17:53:50 UTC

- reply

OK, so one takes the part where the probability density is above some threshold, and that is the shape?

2025-04-08 17:34:05 UTC

- reply

My question is half a joke, half serious.
I have been taught that there are such a thing as protons with electrons that turn around them as the Earth turns around the sun, and then that none of this exists, that everything is a wave function, and then that everything is rather a representation of SU(2). So I don't really understand anymore what physics is talking about.

2025-04-08 17:05:27 UTC

- reply

What is the shape of a wave function?

2025-01-31 20:49:11 UTC

- reply

Onyx Boox Note 2

2025-01-30 05:34:10 UTC

- reply

it will certainly be videotaped and put on yt, and maybe streamed live. you can send an email to the first of the two organizers on that page and ask for more information.
https://philmathulm.hypotheses.org/a-propos

2024-12-17 22:00:47 UTC

- reply

not quite. most mathematicians pretend they work with ZFC but can't state the axioms. they (we) just work with a naive notion of set and the habit not to write something blatantly inconsistent.
of course, every large formalization effort brings some technical debt.
in the 80s, Pierre Cartier explained in several interviews why Bourbaki didn't make the move to categories some 20 years before while they (Cartan, Eilenberg) had basically if not invented them, then making them prominent: that would have meant rewriting 2000 pages and they felt they had better write down their ideas about Lie groups and Lie algebras.

2024-12-13 15:46:38 UTC

- reply

Some philologists I know jumped off their chair at the view of this table. I just can't be correct. They pointed out to Bernard Vitrac's magnum opus and in particular to this website where all known editions of Euclid Elements are shown, with their relations.
https://thamous.univ-rennes1.fr/php/graph_open.php?idGraph=133&projet=grecs

2024-12-08 22:32:09 UTC

- reply

a few years ago, a tv movie needed a list of victims for a crash or something like that, and they took the list of participants to a 2010 math conference in Paris...

2024-12-03 22:18:39 UTC

- reply

Wow, that's rude. I don't feel that having such an implicit use (or abuse) of notation is good for science. And it's definitely not good for pedagogy.
@npub14a5…9fse