Let's take a second to ponder how marvelous the human brain is in its versatility and its ability to learn things it never evolved to do. ⤵️ •1/16
You're probably reading my words as squiggles on a computer screen. The absolutely incredible fact, here, is that these squiggles have ❋meaning❋, and your brain is able to decode these squiggles at an incredible speed to extract said meaning. •2/16
So the meaning travels from my brain to yours through an incredible convoluted, almost rube-goldberg-esque path of my moving my fingers to type keys on a keyboard and generate signals which then enter a very sophisticated electronic system which mankind designed, … •3/16
… to appear at the other end as little LEDs flashing or not, emitting light which your retina sensors detect and which your brain decodes as meaning. ✳︎Every✳︎ step along this path is its own special wonder. •4/16
But what really amazes me is that our brains never evolved to do this “decoding-the-meaning-squiggles” magic, yet we're essentially all able to learn it — and able to do it at a pretty incredible rate. •5/16
And if rather than reading my words you're hearing them instead, the magic is no less impressive: the meaning is now encoded in very minute variations in the pressure of air molecules. We've evolved to detect such variations of pressure, of course, … •6/16
… but the density of meaning we're able to pack into them is still impressive. The entries “Vibratory Telepathy” and “Psychometric Tracery” in Eliezer Yudkowsky's description of “Mundane Magic” are relevant here: lesswrong.com/posts/SXK87NgE… •7/16
I like to learn languages. Or rather, I like to ✳︎start✳︎ learning languages: I never get too far, but the start is the most rewarding part I think, when utterances in the target language cease to be gibberish and the brain realizes there is meaning to be extracted, … •8/16
… after listening repeatedly to the same utterances and thinking about their meaning (initially communicated through translation) the brain connects the two and rewires itself accordingly. This is an almost ecstatic experience. •9/16
But here's another mundane experience which our brain did not evolve for: driving. I thought about it while steering my motorcycle at a speed that rivals the fastest cheetahs: we need a machine to achieve the speed, of course, … •10/16
… but so far, most of us (and certainly those on two wheels) still use our own brains to guide the mechanic. It's incredible that we're able to do this: it requires a lot of complex actions which are nothing like the environments in which we evolved. •11/16
Amazingly, almost everyone can be taught to handle these incredibly fast vehicles and do it safely enough that we put our lives at stake. Almost everyone can be taught to drive with a gear stick, which involves further unnatural complications and “feeling” the engine, … •12/16
… and almost everyone can learn to drive a motorcycle, which involves an even more complicated dynamic equilibrium than a car (look up “countersteering” if you don't know what it is!), and it becomes completely natural in very short time. •13/16
I learned to drive at a fairly late age (three years ago), so I realize perhaps more acutely how strange the rewiring was: the task initially seemed impossible (like making sense of foreign gibberish) and somehow, voilà, my brain rewired itself! •14/16
And of course, the last example I could give is my own profession (not that it's in any way unique or exceptional, but I just happen to know it): our brains did not evolve to think about mathematics beyond, perhaps, the smallest natural numbers, … •15/16
… yet somehow I find myself able to form a mental representation and an intuition of ordinal numbers or finite groups or L² functions, deduce things about them, and communicate them to others. I never cease to find this wonderful. •16/16
• • •
Missing some Tweet in this thread? You can try to
force a refresh
Why are the movies ‘Logan's Run’ and ‘City of Ember’ almost exactly the same?
(Except that the former is so-bad-that-it's-good and the latter is genuinely rather good. Also, the former's æsthetics gave me a 70's-retrofuture-gasm while the latter is steampunk. And ‘Logan's Run’ has Peter Ustimov in in, which is always good. Anyway.)
(Also, ‘Logan's Run’ is a bit like Beethoven's symphonies in that you think you've reached the final resolving chord, but no, it goes on to another resolving chord. And on. And on. And on. And the more it does so, the more it looks like ‘City of Ember’.)
«Une société autoritaire qui enferme 67M de personnes pendant des mois sur simple décision du gouvernement, c'est un peu grave, quand même, même quand on ne se suicide pas. Pas sûr qu'on soit encore une démocratie après.»
L'histoire des mutations, c'est compliqué. (Celles qui nous inquiètent semblent surtout venues d'individus immunodéprimés.) Les mesures de contrôle diminuent la population virale, mais pas le nombre de générations, et ajoutent une pression sélective.
Je suis franchement sceptique sur l'idée de contrôler les mutations d'un virus qui s'est répandu sur toute la planète, a sans doute infecté des centaines de millions de personnes rien qu'en Inde; car, non, l'Inde ne va pas jouer au jeu de «zéro covid».
C'est amusant comme certains qui aiment tenir le discours «il faut écouter les scientifiques!» (et quand on commence à critiquer une publi, «tu n'es pas un expert!») tiennent tout d'un coup un autre discours quand il y a une publi qui les dérange. Quelques remarques. ⤵️ •1/23
— mais, et c'est le point très important, elle ne vaut pas spécialement moins (ni plus) que celles qui prétendent montrer l'efficacité des confinements. •2/23
Il y a d'autres publications, tout aussi peu convaincantes, qui prétendent utiliser des corrélations et comparaisons entre pays montrer l'efficacité ou l'inefficacité des confinements. J'avais par exemple mentionné ceci:
Il est temps que je le reconnaisse publiquement: j'ai eu tort. J'ai eu, de façon répétée, par optimisme excessif, complètement tort. ⤵️ •1/36
Quand la Chine a confiné des dizaines de millions de personnes, il y a un an, j'ai été horrifié par la barbarie du procédé. Je me suis dit «ce sont des méthodes dignes d'un régime totalitaire: quand l'épidémie viendra en Europe, nous réagirons différemment». •2/36
Quand l'Italie a confiné toute la Lombardie, puis tout le pays, en mars 2020, avec des mesures de plus en plus fortes, j'ai été terrifié. Mais je me suis dit: ils sont débordés, ils ne savent pas ce qu'ils font, mais la raison l'emportera sûrement. •3/36
Given a topological space X, can we find another topological space ∇X (ideally in a functorial way…) so that continuous real-valued functions on ∇X coincide with locally constant real-valued functions on X? mathoverflow.net/q/382011/17064
In more sophisticated (i.e., cryptic) terms, does the inclusion functor of the full subcategory of P-spaces (in the sense of Gillman-Henriksen) have a left adjoint?
But with or without sophistication, I don't know whether the ring of locally constant real-valued functions on ℚ can be written as the ring of (all) continuous real-valued functions on some topological space!
De nombreux médias ont repris cette information selon laquelle le covid a fait baisser l'espérance de vie de plusieurs mois. Ce n'est pas exactement «faux», mais c'est EXTRÊMEMENT trompeur.
L'espérance de vie est quelque chose qui a un sens en régime permanent. Calculer l'espérance de vie sur une année exceptionnelle, c'est techniquement possible, mais ça donne une impression extrêmement fallacieuse.
J'ai expliqué ça un peu plus longuement dans ce fil (en anglais):
— disons pour résumer que cette baisse d'espérance de vie correspondrait à ce qui se produirait si la même chose se produisait CHAQUE ANNÉE qu'en 2020.