Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

RustBelt: securing the foundations of the Rust programming language

dl.acm.org

Post reply on HN
Print view View original

71–80 of 109 posts

Re: RustBelt: securing the foundations of the Rust programming language

#71

Post by justinpombrio »

sifoo wrote: post #69
ktpsns wrote: post #25

By the way, this is a project funded by the European Union. It got 2 Million EUR by the European Research Council ( https://cordis.europa.eu/project/rcn/200802_en.html ). It is only one of the many projects which give back a lot to the open source community. Thank you, EU!

I'm not quite as enthusiastic about their choice of funding target. There are plenty of other worthy languages and tools out there that could use that money better in my mind. It's the C++/Java thing all over again, but stepped up several notches; to the point where it's creeping me out since I can't even mention C online without being attacked by the Rust Brigade; and they just don't give up, it's like one of those…

This funding was for verification work, which was applied to Rust. Rust is a relatively new language, and hasn't been the focus of much research yet. C and C++, in contrast, has already been the subject of tons of research and tooling. For example, Wikipedia lists 28 static analysis tools for C, and none for Rust:

https://en.wikipedia.org/wiki/List_of_tools_for_static_code_...

Is there a particular project for a particular language you think is promising and just needs some funding? Or does it just irk you when other people are happy?

EDIT: I'm also not sure why you assume that none of this EU funding is going to your language(s) of choice; have you checked?

Re: RustBelt: securing the foundations of the Rust programming language

#72

Post by namelost »

dbaupp wrote: post #65
namelost wrote: post #61

Earlier quoted context omitted.

> > The most ironic part is that so many restrictions and problems are likely to provoke people to rely on whatever option happens to work, which might not be the best/safest one > Programs written in Rust are empirically safer than programs written in C or C++. But Rust defines safety as being safe from the kinds of errors that the Rust compiler is capable of making one safe from (buffer overflows, race conditions,…

In this case, safer == memory safety (which is not a Rust-specific thing), not general forms of correctness, and the comparison in the original comment is to C and C++, not statically typed languages with a GC.

Unless it has been edited, the original comment doesn't say memory safety, it says "so many restrictions and problems are likely to provoke people to rely on whatever option happens to work, which might not be the best/safest one". The reply to that comment then interprets safety as memory safety and compares to C and C++. This is kind of my point.

Rust defines safety as memory safety, but unfortunately memory safety is only the tip of the iceberg when it comes to safety. Logic errors are just as pernicious.

Programmers have a natural bias towards solving problems with technical solutions, but for me, Rust's memory safety is achieved at the cost of a high productivity penalty and a moderate readability penalty. Both of these reduce my ability to understand and manually verify the code.

Whereas languages that achieve memory safety via a GC, allow me to focus my mental energy on writing correct code, not on memory management.

Re: RustBelt: securing the foundations of the Rust programming language

#73

Post by ktpsns »

justinpombrio wrote: post #71
sifoo wrote: post #69

Earlier quoted context omitted.

I'm not quite as enthusiastic about their choice of funding target. There are plenty of other worthy languages and tools out there that could use that money better in my mind. It's the C++/Java thing all over again, but stepped up several notches; to the point where it's creeping me out since I can't even mention C online without being attacked by the Rust Brigade; and they just don't give up, it's like one of those…

This funding was for verification work, which was applied to Rust. Rust is a relatively new language, and hasn't been the focus of much research yet. C and C++, in contrast, has already been the subject of tons of research and tooling. For example, Wikipedia lists 28 static analysis tools for C, and none for Rust: https://en.wikipedia.org/wiki/List_of_tools_for_static_code_... Is there a particular project for a part…

From my knowledge, the ERC funds fundamental science. In my personal area (theoretical physics), people tend to move from Fortran to C++ nowadays. As an effect, mostly readability decreases (without linear algebra libraries) or the compiler doesn't help anymore with compile-time bound checks (so people use run-time bound checks with assertions instead). Not even to mention the years of wasted time for people finding bugs in their parallel codes. And all codes with the "HPC" badge are parallel nowadays. I think supporting a modern ecosystem is a good decision if people move away from C++ towards Rust or Julia (another ERC... http://opendreamkit.org/).

(Note: Please don't feed the trolls. Nobody has time for language rants)

Re: RustBelt: securing the foundations of the Rust programming language

#74

Post by sifoo »

ktpsns wrote: post #73
justinpombrio wrote: post #71

Earlier quoted context omitted.

This funding was for verification work, which was applied to Rust. Rust is a relatively new language, and hasn't been the focus of much research yet. C and C++, in contrast, has already been the subject of tons of research and tooling. For example, Wikipedia lists 28 static analysis tools for C, and none for Rust: https://en.wikipedia.org/wiki/List_of_tools_for_static_code_... Is there a particular project for a part…

From my knowledge, the ERC funds fundamental science. In my personal area (theoretical physics), people tend to move from Fortran to C++ nowadays. As an effect, mostly readability decreases (without linear algebra libraries) or the compiler doesn't help anymore with compile-time bound checks (so people use run-time bound checks with assertions instead). Not even to mention the years of wasted time for people finding…

How is questioning the choice of project to fund feeding trolls? I was under the impression that diversity was mostly considered a good thing, that has to go for perspectives as well. I just don't get why Rust desperately needs a lot more money than any other tool that's equally useful, they were already pretty well funded by other organizations.

Re: RustBelt: securing the foundations of the Rust programming language

#75

Post by sifoo »

justinpombrio wrote: post #71
sifoo wrote: post #69

Earlier quoted context omitted.

I'm not quite as enthusiastic about their choice of funding target. There are plenty of other worthy languages and tools out there that could use that money better in my mind. It's the C++/Java thing all over again, but stepped up several notches; to the point where it's creeping me out since I can't even mention C online without being attacked by the Rust Brigade; and they just don't give up, it's like one of those…

This funding was for verification work, which was applied to Rust. Rust is a relatively new language, and hasn't been the focus of much research yet. C and C++, in contrast, has already been the subject of tons of research and tooling. For example, Wikipedia lists 28 static analysis tools for C, and none for Rust: https://en.wikipedia.org/wiki/List_of_tools_for_static_code_... Is there a particular project for a part…

Right, question something and you just don't want others to be happy. Good luck with that. Every C compiler and major C library out there to begin with, given that it's the most important and fundamental language in the real world right now. C++ already pulled this trick, it lead nowhere but in circles.

Re: RustBelt: securing the foundations of the Rust programming language

#77

Post by pornel »

Asdfbla wrote: post #5

I think I can understand and appreciate many different aspects of Computer science, but programming language theory truly is a bit too arcane and formal for me. I still try to follow it somewhat though (like when the Morning Paper had its nice writeup about RustBelt) and I enjoy learning Rust (which exposes you to the theory underlying its design more than other languages). Maybe it's good that the results of formal…

I find Rust's RFC/design discussions fascinating. They're a mix of serious Programming Language Theory and practice. The same feature in Rust is seen by some as "algebraic data types" and "oh cool, I can add data to enums" by others.

Some feature proposals starting with "wouldn't it be nice to have…" end up designed with input from people who literally have a PhD in this feature (and are bikesheded until people without PhDs can use the feature).

Re: RustBelt: securing the foundations of the Rust programming language

#78

Post by master-litty »

sifoo wrote: post #75
justinpombrio wrote: post #71

Earlier quoted context omitted.

This funding was for verification work, which was applied to Rust. Rust is a relatively new language, and hasn't been the focus of much research yet. C and C++, in contrast, has already been the subject of tons of research and tooling. For example, Wikipedia lists 28 static analysis tools for C, and none for Rust: https://en.wikipedia.org/wiki/List_of_tools_for_static_code_... Is there a particular project for a part…

Right, question something and you just don't want others to be happy. Good luck with that. Every C compiler and major C library out there to begin with, given that it's the most important and fundamental language in the real world right now. C++ already pulled this trick, it lead nowhere but in circles.

This is a great place to engage in intellectually stimulating conversations. I, among many others, would love to read your objective thoughts about the fundraising choice.

Instead you've decided to facetiously label rustaceans as psychotic and mentally ill because they like sharing an upcoming technology.

I don't know what reaction you wanted but it won't be a positive one here.

Re: RustBelt: securing the foundations of the Rust programming language

#79

Post by pornel »

namelost wrote: post #61
pcwalton wrote: post #33

Earlier quoted context omitted.

> The most ironic part is that so many restrictions and problems are likely to provoke people to rely on whatever option happens to work, which might not be the best/safest one. Programs written in Rust are empirically safer than programs written in C or C++. > The higher the freedom, the better the results delivered by a sensible/knowledgeable person. If that were true, then C and C++ code would be safer and more se…

> > The most ironic part is that so many restrictions and problems are likely to provoke people to rely on whatever option happens to work, which might not be the best/safest one > Programs written in Rust are empirically safer than programs written in C or C++. But Rust defines safety as being safe from the kinds of errors that the Rust compiler is capable of making one safe from (buffer overflows, race conditions,…

> If one if distracted by trying to satisfy the borrow checker

That distraction goes away with experience. Once you "get" how to structure ownership in the program in a borrow-checker friendly way, it stops being a problem.

Similarly, if you are used to dynamically typed languages, you may find static typing a distraction that tests your patience. But OTOH if you think in types and use design patterns for them, they become a tool to improve code reliability, not an obstacle.

Re: RustBelt: securing the foundations of the Rust programming language

#80

Post by sifoo »

master-litty wrote: post #78
sifoo wrote: post #75

Earlier quoted context omitted.

Right, question something and you just don't want others to be happy. Good luck with that. Every C compiler and major C library out there to begin with, given that it's the most important and fundamental language in the real world right now. C++ already pulled this trick, it lead nowhere but in circles.

This is a great place to engage in intellectually stimulating conversations. I, among many others, would love to read your objective thoughts about the fundraising choice. Instead you've decided to facetiously label rustaceans as psychotic and mentally ill because they like sharing an upcoming technology. I don't know what reaction you wanted but it won't be a positive one here.

Every C compiler and major C library out there to begin with, given that it's the most important and fundamental language in the real world right now. Did you get it this time?

Have you been living under a rock lately, or are you just pulling the oldest trick in the psychopath book; condemning others for reacting to abuse? Sharing and attacking others for not agreeing is not the same thing, I hope we agree so far.

Post reply on HN