tim-kt

I don't know how to use WordPress, so I accidentally published this as a standalone site. I'm killing the submitted link and the actual blog post now lives here: https://timktitarev.wordpress.com/2025/10/11/continuous-petr...

The last time I tried Immich (a year ago or so), my impression was that Immich tries to imitate Google Photos as much as possible. This includes features such as searching by a person or by "cat", which requires some machine learning sophistication, which is done locally (you can also disable these features). This would be my guess, but I'm not entirely sure.

> vimtutor is to Babbel what this is to duolingo

It took me half a minute to realize that you probably meant "vimtutor is to VIM master what Babbel is to duolingo".

Meaning the first best language depends on the job?

> A safe community isn’t one where people are held in check by police. People are not roving around thinking “oh I’d break and enter and murder and rape but for the fact a police officer might shoot me.”

That's also not necessarily the point I'm making. Suppose you are in a society where a small part of people are bad actors, for whatever reason. They will break and enter, murder, and rape. You want to protect the rest of the society against these bad actors. You can now equip everyone with weapons so they may defend themselves. That also enables the bad actors to use said weapons because we don't know who really know who is a bad actor (at least not the ones that didn't commit any crimes yet). Or you give weapons only to a small part of society, where you enforce strict gun laws.

The alternative is to reduce the number of bad actors and this is, in part, fulfilled by the conditions that you are describing. But how do I reduce the number of state leaders that are willing to shoot each other? I guess it's what you are saying, namely constructive, open, and fair trade. But we're not really making progress in that direction it seems.

It's an interesting analogy. I'm opposed to guns in every household because we have the police which is meant to give security to people. There, we allow gun use, but under stricter conditions. The majority agrees that this is right, so the system works.

What is "the police" on the level of countries? There is no majority that agrees that, e. g., the NATO can serve as the police. It feels like on this level, we live in an anarchy with only very few actors who don't really want to live together. So maybe nukes are an option, although I don't like it.

Sure, thanks. I also saw this sentence, which was not there in the last version (I think):

> That is to say, with n nodes, gaussian integration will approximate your function's integral with a higher order polynomial than a basic technique would - resulting in more accuracy.

This is not really the case, Gaussian integration is still just interpolation on n nodes, but the way of choosing the nodes increases the integration's exactness degree to 2n-1. It's actually more interesting that Gaussian integration does not require any more work in terms of interpolation, but we just choose our nodes better. (Actually, computing the nodes is sometimes more work, but we can do that once and use them forever.)

> [...] almost all methods of numerical integration (in any case all those that are useful and I am aware of) are equivalent to approximating the target function with another simpler function, which is then integrated using an exact formula.

The key property of quadrature formulas (i.e. numerical integration formulas) is the degree of exactness, which just says up to which degree we can integrate polynomials exactly. The (convergence of the) error of the quadrature depends on this exactness degree.

If you approximate the integral using a sum of n+1 weights and function evaluations, then any quadrature that has exactness degree n or better is in fact an interpolatory quadrature, that is, it is equivalent to interpolating your function on the n+1 nodes and integrating the polynomial. You can check this by (exactly) integrating the Lagrange basis polynomials, through which you can express the interpolation polynomial.

What about it?

No, the integral cannot be "expressed" as a sum over weights and function evaluations (with a "="), it can be approximated with this idea. If you fix any n+1 nodes, interpolate your function, and integrate your polynomial, you will get this sum where the weights are integrals over (Lagrange) basis polynomials. By construction, you can compute the integral of polynomials up to degree n exactly. Now, if you choose the nodes in a particular way (namely, as the zeros of some polynomials), you can increase this to up to 2n+1. What I'm getting at is that the Gaussian integration is not estimating the integrals of polynomials of degree 2n+1, but it's evaluating them exactly.

What exactly is the mathematical model? What are the mathematics involved? I couldn't find it on your blog.

At least in Africa, Facebook is responsible for a large part of internet accessibility. One can say that, in Africa, "Facebook is the internet" [0]. I can't comment on Asia.

[0] https://www.theguardian.com/technology/2022/jan/20/facebook-...

I'm a mathematics graduate student. A good textbook about a topic in math is gold. Sometimes I even prefer reading a book than going to a lecture because I can skip things that I know or take more time on difficult sections. One time, in a theoretical physics course, I just didn't like the lecturer's style. Fortunately, his lectures were based on a very good book (Kuypers, classical mechanics) which I then read. But I don't think that I have ever "finished" a textbook from front to back. There are always things that are more important.

Well, there is a saying about spin of an electron. Imagine that you have a ball and it's spinning. Except, it's not a ball. And it's not really spinning.

Sure, I'm using the word "particle" loosely.

Sorry, my last sentence wasn't formulated well. Yes, a wave with higher amplitude (or one could say "intensity") has a higher energy. The photoelectric effect happens when you shine light with "enough" energy on some material such that the atoms of the material are ionized, i. e. electrons are freed. You need a minimal energy for this and if you use dim light with a low frequency, you will not see the effect. Now, if you increase the frequency of the light, you can measure electrons. If, instead, you make the light brighter, that is, increase the amplitude of the wave (if it were a wave), you don't see electrons. So at least in this experiment, light does not function as a wave.

What is the name of the better model which you are talking about?

The photoelectric effect [0] can be explained if light behaves as discrete particles, but not when it's a wave since a higher amplitude does not imply a higher energy transfer.

[0] https://en.m.wikipedia.org/wiki/Photoelectric_effect

I once built a prototype to a "Goodreads but for classical music", e. g. playing Chopin on a piano. I have thought about doing something like what you are talking about. My concerns are:

1. Moderation: you need to moderate content that people create and moderation is hard. If you don't, the service will be abused.

2. Order: Goodreads has "Librarians" to create some order in the chaos. I think that a meaningful search through "every tokenizable piece of human content" is very non-trivial and would require much work.

3. Monetization: ads are possible. Are they sustainable? I don't know enough of profitability to say something about that. The more specific the type of content is, the more options you have for monetization. For classical music scores my ideas were to make deals with sheet selling companies and make ads for (classical) concerts.

4. Market share: isn't Reddit pretty much a superset of this, minus a built-in rating system?

Care to elaborate? I think they are just trying to put into perspective how much resources LLMs are using?

For an everyday use calculator? Sure. It's still fun and challenging to create a calculator that can handle "as much" math/arithmetics as possible.

To sum it up in one sentence: You don't need to explain words that I can look up in a dictionary, unless you are commenting on that explanation. I know what the word "joke" means but you are forcing me to either read the definition or look for where you continue your comment. It's also slightly demeaning since I can read this as "you don't know what the word joke is, let me explain it to you". You do this again for the words formula, garden and absurdity. If you really need to give extra details to a word that you are using, you can do it in brackets (like this) or with a footnote [1].

The definitions of the words formula and garden also seem out of place because they don't add anything to what you were saying -- at least I don't see any analogies that would make the joke funnier.

[1] Like this.

That sounds plausible. I think that definition is equivalent to a number being rational.

> To me, this pretty much captures what people understand the numbers to be used for outside of college math (so no transfinite, cardinals etc.)

I'm in my fourth year of mathematics right now. I guess I'm not in the target group of articles such as these :P

It's difficult to read this comment. Why are you writing like this?

> Another way to look at what irrational numbers are is to say that they sort of don't really exist, they are like limits, or some ideals that cannot be reached because you'd need to spend infinity to reach that exact number when counting, measuring etc.

Depending on your definition of "existence", rational numbers (or any numbers) don't exist either.

Yes, this happened to me

I don't think that computers have an advantage because they can navigate search spaces efficiently. The search space for difficult theorems is gigantic. Proving them often relies on a combination of experience with rigorous mathematics and very good intuition [1] as well as many, many steps. One example is the classification of all finite simple groups [2], which took about 200 years and a lot of small steps. I guess maybe brute forcing for 200 years with the technology available today might work. But I'm sceptical and sort of hope that I won't be out of a job in 10 years. I'm certainly curious about the current development.

[1] https://terrytao.wordpress.com/career-advice/theres-more-to-...

[2] https://en.m.wikipedia.org/wiki/Classification_of_finite_sim...