logoalt Hacker News

pdimitarlast Wednesday at 11:06 PM0 repliesview on HN

Thanks for entertaining the discussion, the civility and your willingness to expand is very welcome.

> while I believe a change in direction for Rust would be beneficial, the ecosystem advancement and entrenchment of Rust makes it basically a non-starter as of 2025

May I ask why? I think I am gathering from your comment that you think the language is too big (which I don't understand, could you please clarify?) and that maybe we as an area become too dependent on too few PLs / frameworks? Is your worry an increasing centralization perhaps?

> If every company was use to and had to invent at a minimum their own dialect of a broadly defined language types and then train employees to function within their language environment I would be thrilled.

I was not there but I heard from folks on HN that during the LISP era a good amount of companies did this: they built their own DSL that described their business perfectly (on top of LISP) and then even taught business people how to modify parts of the system by giving them access to only some modules. Result was reported to be a crushing success.

But if we go by your reservations, would you then say LISP is too big and at a risk to entrench itself everywhere (I mean if you were there back then)?

> The above would do a considerable amount to stop corporations from treating programmer like replaceable/disposable cogs in a machine

I would really _love_ for that to be true but I remain skeptical. In my 24 years of career I have only always noticed how businesses always work hard trying to replace us. It's a constant tug of war and we are more or less tolerated because they can't do without us. The moment they feel that they could, most of us would be fired in a heartbeat -- and some of that, in a much smaller scale than they wanted, already happened with the advent of good-ish coding LLM agents.

> where it really hurts for me is that Rust’s pervasiveness prevents moving to a better option in the space due to moneyed interest and cultural buyin

Here we agree at 100%. I do like Rust a lot (though I don't work with it for a while now, I keep using it for personal projects and a little bit of portfolio work) but I believe it's a local maxima that we would all be stuck with for a while.

But that's not an indictment on Rust in particular IMO; it's a judgement towards the risk-averse nature of the area and something I personally don't blame people for (you should not rewrite your business code once every 5 years after all).

--

All that being said, to me Rust is an objective improvement of the state of affairs. It's going to be the new C++ and maybe even the new C, we'll see. And I agree with you that it's not without faults (`async` could have been done better; it's a huge slog to learn it properly and that really did not need to be the case).