xorand

Finished, see especially "How not to read these notes" (section 1) and "How you can contribute" (section 2) https://arxiv.org/abs/2003.14332

"We can program a computer to do anything. What if we had the same power over the molecules of our bodies?"

Almost finished project, you can spend hours in there. https://chemlambda.github.io

I found the previous Hamiltonian Neural Networks [0]. If the authors are here, I'd be interested to use [1] for a version with dissipation.

[0] https://greydanus.github.io/2019/05/15/hamiltonian-nns/

[1] https://arxiv.org/abs/1902.04598

Yes, like linux liberated the pc from windows. Now is much more complicated, because a liberated harware is needed as well. Sigh, this is too much, but at least as you say: a normal computer in the smartphone form factor, where I can install the anti-android OS, including the game where you replace the android logo with the ...? Of course the anti-android would allow me to access anything from the android world, but easier.

I upvoted this, but it is not accurate. If you're French then you may be misled by the combination of two signals: "Ecole Polytechnique (X)" (aka "social status") and birth date (aka "old"). This gives "boomer" hence the sarcasm.

Now that memories are mentioned, I'm not a grand-pere, nor French. Grand-peres are proud of their '69 youth. Yes, they stopped the social elevator and I share your opinion about them. "X" was relevant only for the availability of (mostly empty) rooms full of "unix workstations" and the friends I made in Paris. During this time, your grand-pere probably was vaguely proud about the minitel and a bit disturbed about the fact that people like me are allowed to study at X.

Oh sweet memories. I was born in 1967, so I'm older than the author, but the memories are somehow the same. I started with a Sinclair ZX Spectrum clone, called HC85, and with Basic as first language. I discovered the Mosaic browser in 1994 at Ecole Polytechnique, Paris, and understood that the future of research, publication, collaboration will be very different than the past. I'm a mathematician. I installed my first Linux in 1995. All in all in my life I bought exactly 0 bits of proprietary programs.

I had for a very short time a FB account, which I deleted. I deleted my Twitter account about two years ago. I was hooked by Google+ and I had a popular collection about artificial life (now is public again). But I deleted my g+account before they closed it.

All my family has smartphones, but I can't stand the limitations. When the smartphones will be liberated I'll have one.

My overall impression is that, some details excepted, now that everybody has a computer in the pocket, people pass through the same learning process as we nerds did some years ago. Today is harder, because less freedom. On the other side, the new thing is that today everybody is online.

Same here. This "it was a dark and stormy night" style is bad writing. I read a lot, I like reading novels. This is just an empty structure. In this particular case the bad writing almost killed an interesting subject.

I don't quite understand this conflation of mathematics with category theory. This obsession of some programmers with mathematics, actually with a tiny part which is category theory, looks to me very strange. By chance, there is this recent comment of Scott Aaronson, a strong mathematician who is a rising star in quantum computing, which contains what is it probably the more balanced view.

I quote from the source [0] the relevant part: "With some things I don’t understand well (nuclear physics, representation theory), there are nevertheless short “certificates / NP witnesses of importance” that prove to me that the effort to understand them would be amply repaid. [...] And then, alas, there are bodies of thought for which I’ve found neither certificates or anti-certificates—like category theory, or programming language semantics [...] For those I simply wish the theorizers well, and wait around for someone who will show me why I can’t not study what they found."

[0] https://www.scottaaronson.com/blog/?p=4616#comment-1830447

I suppose the same context as serving a drink in a jar.

Waiter delivers a shovel of spaghetti [code], client happy.

:) Probably is a sort of revenge on people more interested in semantics than geometry.

A scenario more interesting than boundless self-replication is Ackermann goo [0], [1]. Grey goo starts with a molecular machine able to replicate itself. You get exponentially more copies, hence goo. Imagine that we could build molecules like programs which execute themselves via chemical interactions with the environment. Then, for example, a Y combinator machine would appear as a linearly growing string [2]. No danger here. Take Ackermann(4,4) now. This is vastly more complex than a goo made of lots of small dumb copies.

[0] https://chemlambda.github.io/collection.html#58

[1] https://chemlambda.github.io/collection.html#59

[2] https://chemlambda.github.io/collection.html#259

Let me check it. If you see this pls vote or reply.

Some numbers: There are 264 posts with animations, a bit more than 1/2 of the original collection, with the possibility to rerun in js the simulations.

Whenever possible there is a mol file attached to the animation. There are about 490 mol files. If they are too big to be used without stalling the js reduction, this is signaled by the message “mol too big” in the post. If there is no mol which matches, this is signaled as “mol unavailable”.

Of all 264 posts, 36 of them fall in the “mol too big” category, 46 in the “mol unavailable” and there are 6 posts which don’t have a chemlambda simulation inside. So this leaves 264-88=176 posts which have matching mol files to play with.

The original revived collection is here [1], but since 2 weeks the traffic is too big, or something strange happens [2], I made a copy with smaller images, to fit the github constraints.

[1] http://imar.ro/~mbuliga/collection.html#153

[2] https://chorasimilarity.wordpress.com/2020/01/29/ddos-attack...

From http://www.perseus.tufts.edu/hopper/text?doc=Perseus:text:19...

opp. to a professed orator, opp. a professed philosopher

Comes from http://www.perseus.tufts.edu/hopper/text?doc=Perseus:text:19...

where you can find: opp. public, my personal opinion, unique and different from others

There's a great book by Truesdell, named "An Idiot's Fugitive Essays on Science". According to Truesdell, the initial meaning of the word "idiot" was one who does not have preconceived ideas.

https://www.springer.com/gp/book/9781461381877

If you're into lambda calculus or artificial life, then here is a landing page I made for several of my (and contributors) projects. You have here a lambda calculus parser and graphic rewrites reducer, a collection of artificial life animations which you can replay in js, graphical quines, presentations slides, you name it.

https://chemlambda.github.io/index.html

No, all insects have 6 legs https://en.wikipedia.org/wiki/Insect. Spiders have 8. https://en.wikipedia.org/wiki/Spider

Btw, you know that spiders are not insects?

Some trivial mathematical facts. No matter what algorithm you use, you want to make a quasiisometrical embedding of a tree into a finite dimensional space. Or, for a tree the no of nodes at distance R from the root is like exp R, but in any space of dimension N you can cram about R^N points which are at distance R from the origin. Hence the problem.

But you can embedd any tree into the Hilbert space :)

I don't get it, is "classic" bad? My page is online since 1995 and is still the best way I share my work. There surely are many "classic" style sites, with unique information. I'd encourage everybody who has something rare to share, to do this in a "classic" way.

http://imar.ro/~mbuliga/

I second this. Conserved quantities (by extension symmetries) do not compute. Car engines and refrigerators are both thermal engines, but the same conserved quantities are engineered (say programmed) to do different things.

I'm interested to know more about it, especially if it can be used for Open Science. As a researcher, I look for the less bloated way to replace a scientific article with a corpus of work which can be independently used by other researchers.

Looks like this: https://imgur.com/a/S0kG880

There is life outside of Google, and always was.

I had a collection of animations related to lambda calculus and alife on g+. Well, now is back online and moreover it is better, because I can add the option to remake these animations live, with js.

http://imar.ro/~mbuliga/collection.html

Paris is a big city, that's all. The Eiffel Tower and Moulin Rouge are not among the most interesting things to see in Paris. People are not like in a Disney movie.

Sci-Hub does not have the opportunity to sell its platform, there is no advertising, or ‘social networking’ to obtain vital user data that it can monetise; it is a pure and unashamed ‘pirate’.

Black is white and white is black. They are not pirates and they have no connection with the 'pirates' political movement AFAIK. Feel free to contradict me, I'll learn something new.

Programming is now a bureaucracy.

[1] "In any bureaucracy, the people devoted to the benefit of the bureaucracy itself always get in control and those dedicated to the goals the bureaucracy is supposed to accomplish have less and less influence, and sometimes are eliminated entirely"

[1] https://en.wikipedia.org/wiki/Jerry_Pournelle#Pournelle%27s_...

Thanks for the link to Kipf's paper, I was not aware of it. I'm a mathematician working on graph rewrite systems, I would like to explore in the NN direction, in case you're interested contact me.

I think, similarly, that there is an underground graphical rewriting formalism for all of mathematics, category theory included. Category theory is somehow a step towards it, but we don't keep categories in our heads, only the graph reducer.