Talk:Rust (programming language)/Archive 4

Archive 1

Archive 2

Archive 3

Archive 4

OO allusions

@War: @Ovinus: About the OO allusions (and there was also a statement about this in the GA review), I disagree that Rust is not OO. Rust is multi-paradigm and can be used for OO. See e.g. here. For that reason, I don't think we should emove mentions of objects, object system, etc. Is there something else we can do to clarify this better for readers? Caleb Stanford (talk) 13:42, 16 July 2022 (UTC)

See also Object-oriented programming, Composition over inheritance. Although I don't think we should call them as "object"s because that is not the correct name, and not used except for trait objects (which are just vtables). 0xDeadbeef 14:38, 16 July 2022 (UTC)

Adam Savage once said that "Every tool's a hammer." This does not mean that it is good idea to describe a tool by its hammer-like properties. As @0xDeadbeef mentioned, Chapter 17 of the Rust docs discusses the issue of how Rust is, and is not, OO. I think we should strive to not confuse the reader into thinking that Rust is a syntactically different C++, when it clearly is not. This came up on the lifetime section concerning "destructors". Again, the Rust docs talk about destructors, but in Rust they are not the same as C++ destructors. Suggesting that they are is inaccurate and confusing to the reader.

I get that with a particular background one thinks of these things from that background. A person that comes from a C++ background will associate features of Rust "as" or "like" various C++ features or C# features or Haskel features, etc. I think we should avoid these associations except where there is a precise and unambiguous correspondence. War (talk) 16:28, 16 July 2022 (UTC)

I confess I'm a bit confused by your response. I didn't say anything about C++. P.S. regarding influences in the lead, do you mind if we keep them in alphabetical order? The source doesn't explicitly state they're in order of how major they are and I am reluctant to make any sort of delineation between them, as we probably won't have good sources to back that up. Caleb Stanford (talk) 20:05, 16 July 2022 (UTC)

As a C++ programmer but not a Rust one, a destructor in Rust looks quite similar to one in C++ and other OO languages, and is aptly named as such. My suggestion would be to state what has been said here: that Rust has some features usually associated with OO languages, but is not regarded as one. Hawkeye7 (discuss) 21:00, 16 July 2022 (UTC)

I may have been sloppy in my edit about RAII by saying "object" rather than "value" (although [1] says "Rust enforces RAII (Resource Acquisition Is Initialization), so whenever an object goes out of scope, its destructor is called and its owned resources are freed"), but, even being agnostic on whether Rust is truly object-oriented, I don't think RAII is necessarily restricted to object-oriented languages, and several sources say Rust very frequently uses that pattern. Ovinus (talk) 21:26, 16 July 2022 (UTC)

@Caleb Stanford 'regarding influences in the lead'. I put them in the order in which they are mentioned in the Rust documentation which seems better than an arbitrary alphabetical ordering. War (talk) 21:30, 16 July 2022 (UTC)

I think we need a 3rd opinion then. The source doesn't say what the order represents, so it's giving more weight to the order than I think it merits. In particular, the current order suggests SML/OCaml had a higher impact on Rust than C++, but the source does not state that explicitly. The point of the alphabetical ordering is to be arbitrary and not assert anything in particular. Caleb Stanford (talk) 02:23, 17 July 2022 (UTC)

Caleb Stanford, third opinions only work for disputes between two people. 0xDeadbeef 03:21, 17 July 2022 (UTC)

I meant about about the major influences point. As far as I saw only War and I weighed in on that, did I miss something? Would you like to weigh in? Caleb Stanford (talk) 03:28, 17 July 2022 (UTC)

Sorry, I didn't read enough context when I posted that comment. Regarding major influences, I think we shouldn't mention any specific languages in the lead at all, instead we could refer to categories/families of languages. 0xDeadbeef 04:24, 17 July 2022 (UTC)

Okay, now we have 3 different opinions lol. Thank you for the input Caleb Stanford (talk) 05:09, 17 July 2022 (UTC)

I disagree with delleting information on anything related to the OOP. Though, it seems like (according to the docs) that Rust isn't supporitng inhertiance in usual sense. This should be covered by the articles. AXONOV (talk) ⚑ 11:31, 2 June 2023 (UTC)

a source for potential use

https://www.infoq.com/news/2019/03/rust-npm-performance/ 0xDeadbeef→∞ (talk to me) 13:54, 19 November 2023 (UTC)

Is infoq.com a reliable source ? Sohom (talk) 16:58, 19 November 2023 (UTC)

I've seen it used on many articles, so presumably yes. I know ZDNet and Ars Technica are considered generally reliable per WP:RSP, but InfoQ hasn't been subject to much discussion. I've seen InfoQ's articles and they are generally pretty good. (much better than self-published sources on Medium and the likes) 0xDeadbeef→∞ (talk to me) 06:31, 20 November 2023 (UTC)

I would personally not include it unless npm themselves publish something about this. The article is basically a summary of a whitepaper written by Rust devs, which is itself a primary source and I cannot find any other (unreliable/reliable) source mentioning this. Sohom (talk) 20:33, 20 November 2023 (UTC)

Added! Caleb Stanford (talk) 20:33, 20 November 2023 (UTC)

rv per my latest comments. Sohom (talk) 20:54, 20 November 2023 (UTC)

is the white paper itself a RS in your view? I think it's a real thing, still looking for other sources. Thanks, Caleb Stanford (talk) 23:24, 21 November 2023 (UTC)

IMO, the InfoQ source should be more reliable than the white paper itself. The white paper is a primary source. 0xDeadbeef→∞ (talk to me) 00:01, 22 November 2023 (UTC)