Navigate Select ESC Close

You’re all wrong

2025-11-21 Science & Technology
76.4k
2.5k
385
Theo - t3․gg
Theo - t3․gg
539.0k subscribers

Unlock all features

FREE: Get instant access to 10 AI summaries, chats, or transcripts per day.

Description

Developers are really bad at picking languages... Thank you Sevalla for sponsoring! Check them out at: https://soydev.link/sevalla SOURCES https://spf13.com/p/the-hidden-conversation/ https://x.com/spf13 Want to sponsor a video? Learn more here: https://soydev.link/sponsor-me Check out my Twitch, Twitter, Discord more at https://t3.gg S/O Ph4se0n3 for the awesome edit 🙏

Top Comments (10)

@doxologist 2025-11-21

Thanks for convincing me to use Rust for my next failed startup!🎉

406 4 replies
@JoelByrén 2025-11-21

"let an LLM decide for you, I don't even know anymore" was the perfect ending to this one ;)

121
@haydnbuilds 2025-11-21

I don't, but that's because I'm not a decent programmer

62 2 replies
@diadetediotedio6918 2025-11-21

This is not a productive article*. Most of the time there's not a "THE RIGHT SOLUTION" screaming. if you want concurrence you may think "well, Go is obviously the best candidate for that job, it has the right primitives and the a good balance between X or Y", but is it? You can probably do just the same in Scala, or any other functional language with a good support for structured concurrency, it could be even BETTER and more productive once the product reaches a certain scale, or it could be WORSE than Go. You will NEVER know that kind of thing beforehand, unless the scope of the product is extremely specific, and even so this is mostly a "probably better". This is not to say, whoever, that there are not objectively WORSE languages for a problem at a given moment, or as a starting point. I mean languages that would require you to build entire systems on top of it for them to be convenient to use for your problem, a lack of a mature ecosystem, etc. You can't even have an objective measure of "how much will it cost us" like the author would like, because this would require us to be able to evaluate ALL the repercussions of the language in development over time. But this whole article is more of the "I know better" crap, it's someone rationalizing the rationalization other people do, and thinking "well, now I know that, I'm better, because I KNOW the objective reasons for things, stay tuned for my next article!". The fact that person is using neuroscience ideas to justify their position on the topic is even more crazy, you don't argue that way, you can't simply start establishing connections between things without a proper further study like he did, and you need to take care of extrapolations on the conclusions of studies like these. And this person went even further, calling the decision predetermined, which is complete bullshit, biases don't work that way, they BIAS, they don't DETERMINE, this is a basic conceptual mistake. Having a bias is not the same as being unable to rationally think about something, or even change your beliefs, it only makes it significantly harder to do so. Even if I accept the treatment of languages as identities in this specific sense, this author is trying to give, it's not sufficient for me to say something like "all the discussions for which language to use ended the moment you hired a rust programmer". As for the classic "what do you propose, then?". I just don't care, languages are tools, some I prefer, some I don't, I'll use whatever people need me to use, and I'll still point out the crap all of them have when I have the opportunity to. If someone ask me, I'll probably say "what is the problem you are trying to solve?" and in the majority of cases I will say "use the same language your team is already comfortable and productive at", because this mostly does not matter. If I perceive an extreme gap in quality that could be mitigated by switching to a different language, then and just then I would recommend it, it's the reason even when I have a hateful relationship with C#, I work with it most of my time, it get's the damn job done. And for my personal uses, I'll use whatever pleases me more, currently it's swift, before it was C#, sometimes I like to use F#, this don't matter, it's mostly a convenience problem.

60 6 replies
@Gwentedness 2025-11-22

Thank you for validating my decision to make my first startup in Rust and force every sector of my flourishing company to use it.

48
@charlesdeuter 2025-11-21

I don't use Rails much, but I have enormous respect for it, sure it's the "wrong" decision, but it's the best wrong decision you'll ever make.

36 4 replies
@titato 2025-11-21

i'll pick whatever pays me the most

35 1 replies
@everdale8920 2025-11-21

About the bias study: if you read up on the opposite belief and still holds your belief strongly, that would suggest what you previously had a hunch about being right was not swayed when you received more information. It actually makes sense to be more convinced in your beliefs in this scenario. But yeah, it basically suggest that information creates polarisation, which could be an unfortunate truth (though I believe it does depend on the culture you live in).

6
@mfahz4067 2025-11-22

The intro is just "Primeagen, Primeagen, Primeagen, Theo"

4
@mlenthusiast6464 2025-11-22

"To consider an alternate view, you have to imagine an alternative version of yourself." That should be on t-shirt. Thanks theo, learned something new today.

4

Unlock the Data Inside
Turn Videos into Knowledge

  • Get FREE 10/day: transcripts, summaries, chats
  • Chat with videos, export text & PDF
  • $1 free API credit for RAG, chatbots & research

Free forever plan • All features unlocked

App screenshot