Learning Rust
Learning Rust
Posted Feb 5, 2025 18:01 UTC (Wed) by corbet (editor, #1)In reply to: Learning Rust by Trainninny
Parent article: Resistance to Rust abstractions for DMA mapping
So these posts are increasingly beginning to look like trolling, and do not appear to be adding anything useful to the discussion. Perhaps it's time for both the posts and the responses to stop?
Posted Feb 5, 2025 18:42 UTC (Wed)
by Trainninny (guest, #175745)
[Link] (4 responses)
I do claim that I am putting effort into my comments. For instance, in https://lwn.net/Articles/1008103/ , I put effort into discussing programming language design.
If you are referring to the holes in Rust's type system, there are several examples of that and associated problems, discussions, mitigations and fixes.
github.com/rust-lang/rust/issues/25860
github.com/rust-lang/types-team/issues/117
blog.rust-lang.org/inside-rust/2023/07/17/trait-system-refactor-initiative.html
github.com/rust-lang/rust/issues/25860#issuecomment-1455898550
github.com/orgs/rust-lang/projects/44/views/1
github.com/rust-lang/rust/issues/75313#issuecomment-672216146
github.com/rust-lang/rust/pull/75443
github.com/rust-lang/rust/issues/75992
github.com/rust-lang/rust/issues?q=state%3Aopen%20label%3A%22A-type-system%22
github.com/rust-lang/rust/issues/129844
github.com/rust-lang/rust/issues/25860#issuecomment-1579067138
Rust has an advanced solver and type system checker, but it is also very complex, and the Rust language developers are working on a new solver and type system and trying to make it as backwards compatible as possible.
What should I do in regards to this discussion? Are my comments off-topic? Too long? Something else?
Posted Feb 5, 2025 19:13 UTC (Wed)
by Cyberax (✭ supporter ✭, #52523)
[Link] (1 responses)
Posted Feb 5, 2025 20:02 UTC (Wed)
by Trainninny (guest, #175745)
[Link]
Posted Feb 5, 2025 20:01 UTC (Wed)
by daroc (editor, #160859)
[Link]
There's nothing wrong with long, detailed technical comments. One of the things that I love about the LWN comment section is that we will occasionally get real experts coming in and writing detailed responses to stuff. That's also why the comment submission form has no maximum length.
But valuable comments are usually written to directly address the comments or articles they're responding to — something that just inserting long quotes or lists of links to other resources pretty much cannot do. After all, you're responding to something written after the source you're quoting, in most cases.
Instead of quoting large sections of text, it is usually better to say something along the lines of "I disagree with [specific point X] because of [specific reason Y, stated in your own words]. In support of that, [here is a brief and narrowly targeted quote on the topic]."
Generally, when we say that comments should be "informative", that means they should be both information dense, and presented in a way that makes people more informed. Ideally, someone should be able to come to your comment, read it, and feel like they understand more than they did before. If they bounce off because of a lengthy quote, or because you're not responding directly to their previous points, that makes your comment less valuable even if the topic is otherwise interesting.
Posted Feb 5, 2025 21:51 UTC (Wed)
by Wol (subscriber, #4433)
[Link]
And this has exactly what to do with the price of tea in China?
Rust developers are writing theorem provers so they can move code out of "unsafe" into "safe". Are you saying they shouldn't be doing that?
That is why you're seen as trolling. You're treating other peoples' attempts to improve the language - which is what Rust expects of its developers - as evidence that Rust is not fit for purpose.
Cheers,
Learning Rust
Learning Rust
Learning Rust
Learning Rust
Learning Rust
Wol