a57721

> quirky California college radio stations

I listen to SomaFM (https://somafm.com/) and FIP (https://www.radiofrance.fr/fip), they have online streams by genre. When something gets me interested, I look up the artist, and I keep discovering lots of new names, independent labels, etc.

Normally you don't want to format automatically generated code, you adjust the code generator instead.

The text on all pages looks very much LLM-generated, and I think all "user comments" there are fake too. (Edit: I mean "what people said" sections.)

I like the idea of the site, but not the execution.

It feels weird to read "nostalgic recollections" that pretend to be human, while in reality come from gen AI.

It is obviously AI-generated, lots of filler text that communicates nothing of substance. Disappointed to see this on the front page, what a waste of reader's time.

Usually when someone decides to share code with the world, they don't want to publish the actual development history. They publish the first version that is ready to go public as the first commit. With enough functionality etc.

Isn't it normal for the initial commit to be large?

I have a similar experience, I learned English much later than my first programming languages, and picking up some keywords and basic APIs was never an issue (it was BASIC and C/C++ at the time). Maybe I would occasionally look up in a dictionary what is 'needle' and 'haystack' in a code snippet, and I was puzzled by the ubiquitous "foo, bar, baz", which to my relief turned out to be equally cryptic for the native speakers. I still don't feel about code as a kind of English prose, it occupies a separate part of my brain, compared to the natural languages.

Scala and Kotlin have 'tailrec' annotation/modifier, though not as sophisticated as you describe.

You mention it was 8 years ago, at that point a typical Java dev would be already using Spring Boot for requests and deserializing JSON to POJOs (with Jackson under the hood).

I get your point, but the natural numbers do not form a ring or field.

It looks like a parody of LLM delusion, but the PR is oddly specific to be just trolling, and the author also submitted his work to HN: https://news.ycombinator.com/item?id=45982416

I am doing backend in Kotlin, but I must admit that Java has been catching up quickly, and it seems like Kotlin has been shifting its focus to Kotlin Multiplatform. Modern Java is a good, pleasant language and a safer bet.

Gradle with Kotlin DSL is nice, what's annoying is Gradle's constant API reshuffling for the sake of it that breaks plugins. Some plugins also introduce pointless breaking changes just to have a fancier DSL.

The IDE support is not an issue in practice, in my opinion, because IDEA is the best IDE for both Java and Kotlin. The official Kotlin LSP was released 6 months ago, but I haven't tried it.

A vector is always a vector -- an element of something that satisfies the axioms of a vector space. The author starts with the example of R^n, which is a very particular vector space that is finite-dimensional and comes with a "canonical" basis (0,...,1,...,0). In general, a basis will always exist for any vector space (using the axiom of choice), but there is no need to fix it, unless you do some calculations. The analogy with R^n is the only reason the "indices" are mentioned, and I think this only creates more confusion.

> and they aren’t irrational (i.e. they have a finite precision)

No, if you want only rational "indices", then your vector space has a countable basis. Interesting vector spaces in analysis are uncountably infinite dimensional. (And for this reason the usual notion of a basis is not very useful in this context.)

I had a similar experience. When I bought a device with faulty electronic components on Amazon, I wrote a negative review, and almost immediately I was notified that it had been flagged and removed for violating the "community guidelines". Apparently, a seller can do that. My review was a polite explanation of the issues, obviously not violating anything and not accusing the seller of anything, but now I'm sure they had refurbished units or a batch that was known to be faulty.

One factual thing that looks off is "the UK is imprisoning thousands for their tweets". I'm not in the UK and not following closely the situation there, but "thousands", really? Genuine doubt, would love to see some evidence.

Otherwise, the "doomer manifest" is OK, but the comically inflated ego of Durov is annoying, him thinking that such banal and commonplace sentiments are worth pushing as an alert message to all users, wrapping everything into announcing his birthday (that he doesn't want to celebrate, oh no).

It is used for things like "Foo x = new Foo()" where the type is obvious.

> humans decided they were "prime", i.e. most important, based on arbitrary considerations

No, not arbitrary considerations.

The term goes back at least to Euclid who investigated factorization of integers in his "Elements". He used the greek word "protos" that was later translated to Latin as "primus". It doesn't mean "most important", rather "first".

The idea is that primes are the "multiplicative building blocks" for other numbers, the "origin" or "first principles", because every integer factors into primes. When a mathematical object can be decomposed in some way, it is very natural to study the irreducible blocks, because many questions boil down to them.

Oh I think you're right and both took it from Common Lisp, which largely predates namespaces in C++ and R.

C++ also uses (::), by the way. R probably borrowed it from C++.

> you still need a higher % of trustworthiness than LLMs can provide for parts of the business and data layers

For many domain-heavy systems, it's not even the trustworthiness; just getting the business logic right requires a lot of work and lots of iterations with in-house domain experts and clients, there's no way LLMs can do that.

I think remasters are uploaded by the record labels, it's not some kind of algorithm run by Spotify.

It has loudness normalization, which can be turned off in the settings.

Among well-known mathematicians, Gabriel Lamé claimed a proof in 1847 that was assuming unique factorization in cyclotomic fields.

This was not obvious at the time, and in fact, Ernst Kummer had discovered the assumption to be false some years before (unbeknownst to Lamé) and laid down foundations of algebraic number theory to investigate the issue.

I'm sure this feature is a joke, not intended to be usable. It's a personal blog.

You can use LLMs to check grammar and style and manually go through each suggested correction; it will take more time, but your readers will appreciate the effort and your own voice and style will be there. People get upset by LLM-generated content not because they are against LLMs, but because of the awkward style that is very recognizable and leaves you wondering "If the author couldn't be bothered to write it, why should I bother to read it?"

> it was generally more brittle and harder to maintain

It depends on the system in question, sometimes it's really worth it. Such setups are brittle by design, otherwise you get teams that ship fast but produce bugs that surface randomly in the runtime.

My apologies, I didn't want to sound patronizing and wasn't making assumptions about your work and experience based on your examples, I am happy that generative AI allows you to make such apps. However, they are very similar to the demos that are always presented as showcases.

> and before you discount it as too trivial, somehow none of the other ones did what I wanted

No offense, it's really great that you are able to make apps that do exactly what you want, but your examples are not very good to show that "software projects that somehow are 100% human developed will not be competitive with AI assisted or written projects" (as someone else suggested above). Complex real world software is different from pomodoro timers and TODO lists.

In the JVM projects, I use Karate for integration tests: https://github.com/karatelabs/karate

It gives you full control of constructing requests and assertions because test scenarios may include arbitrary JavaScript.

It really depends on the context, e.g. if you need texts for a database of word frequencies, then the answer is a clear "yes", and LLMs have already ruined everything [1]. The only exception from your list would be OCR where a human proofreads the output.

[1] https://github.com/rspeer/wordfreq/blob/master/SUNSET.md

As a mathematician, I think it is very hard to write popular expositions of mathematics. It is acceptable and even necessary to skip the technical details and replace rigorous definitions with intuitive explanations, but it should be done with great care to avoid statements that are plainly wrong and only introduce more confusion. I think the majority of texts on complicated mathematics that target the general audience suffer from this issue.

The review mentions some very unfortunate errors that are not "passing over a point in less detail": suggesting that Cantor's diagonal argument depends on the AC and that CH is equivalent to c = 2^ℵ₀.

I wasn't familiar with the book and had just read the review; I wouldn't blame the author, but the publisher should have contacted someone for proofreading.