Leib: There are things that you can valuate without some personal bias

Chapek: Deroos: how does Rust and ATS negatively impact programmer productivity?

Shipps: And for sure C show how old is

Staenglen: Granted Rust is too young, there are lots of things that can be improved, and there’s almost no tooling

Deroos: Dolby: Yeah if that’s where you’re at there’s likely no convincing you anyway

Coykendall: But one has also to think that the area of application of C the only alternatives are C++ and I think Ada

Orlowsky: Hannibal_Smith: What exactly did you want to show me there?

Hovnanian: Madsy, that there are objective values when one has to think about a language

Ansoategui: Hannibal_Smith: In a perfect world, sure. But reality isn’t perfect

Heppe: Hannibal_Smith: hopefully Rust can be another alternative in the distant future.

Claggett: What’s stopping you from using Rust other than the tooling ecosystem?

Briehl: Hannibal_Smith: For example, security checks aren’t always cheap or free memory-wise or performance-wise. Sometimes they might even stop you from doing what you need to do.

Stagliano: Havvy, for me nothing but I’m a a CS student

Gullickson: Havvy: nothing, but it doesn’t really replace C/C++ right now, for some of the cases where C/C++ are used. Sure it replaces a lot of C++ for application-level stuff.

Deroos: I can write my application in other languages and achieve the kind of stability I require a lot faster than I could with Rust where I’m forced to resolve a bunch of technical issues upfront which delays the time from me having an algorithmic idea, and me typing it out and hanving it work

Mininger: I can see why today Rust is not a viable language to areas where C or C++ are used

Ackerson: Dolby: Ah, so the #no_std stuff then.

Clune: Hannibal_Smith: Choosing a programming language to use in the real world isn’t just about the pros and cons of the semantics of the language, and I think you know that. Even office politics plays a role :

Estep: Deroos: that’s a tooling problem.

Chiprean: Haskell has the same problem, in fact.

Deroos: That’s a lofty empty promise

Jakubek: Madsy, yes I remember that the Mars Rover code is written in C because is what the people in that team was familiar with

Deroos: I don’t think you can tool away the Rust programmer’s responsibility and burden to consider much more upfront than in other languages

Deroos: There are burdens on rust programmers that there are not on programmers of other languages in even just considering basic structure

Eanes: Deroos: you’re thinking that Rust requires the programmer to do that in the first place, which isn’t exactly the case.

Deroos: This increases the time between algorithmic insight and implementation

Deroos: And for some uses cases this cost is totally justified

Deroos: Because the memory safetey of the program is its largest source of value

Deroos: Like, space stations and medical equipment

Chomicki: Sure types enforce additional constraints. But tooling can solve that. There’s nothing really preventing you from having an environment that is as productive as Smalltalk’s while using Rust.

Deroos: If I’m going to experiment with some silly science exploratory code, I’m going to use python

Deroos: Since its just as fast as Rust but I don’t have this cost, and Nim is nice and productive like Python

Fegurgur: Well, I don’t consider Python to be a productive language :

Bisard: Deroos, keep in mind that in cases where at least C or C++ is used today begin fast is not always the primary reason why are used

Umezawa: Mostly because tooling around Python is pretty bad too.

Deroos: Not if you need stability levels near rust

Deroos: You will have a bad time making sure there are no bugs in your python code

Deroos: Hannibal_Smith: absolutely