RustConf 2019 - The Symbiotic Relationship of C++ and Rust by Isabella Muerte

RustConf 2019 - The Symbiotic Relationship of C++ and Rust by Isabella Muerte

For better or worse, Rust and C++ have a shared future ahead of them. C++ will affect the design of Rust, and so too will Rust affect the design of C++. Rust's future is most likely the same path as C++, but what does that mean exactly? In this talk, we'll learn about the evolution of these languages, how much they have in common, the power their respective communities have to shape them, and the responsibility these communities have to each other.

[tialaramex]

It's weird to discuss standards, on the Internet, and not notice that hey, lots of the standards around here didn't come from ISO, or IEC, or EMCA, or W3C, or numerous other bodies with the mentioned problems. Standards like IP, or TLS came from the IETF. What's the membership fee for the IETF? Trick question, the IETF does not have members and so cannot charge them a fee. IETF Working Groups are open to all individuals. Only people can participate. Google can't participate in the IETF, but Google can (and does) pay people to do that as part of their job. China can't participate in the IETF, but it too can (and perhaps does) pay people to do so. Hobbyists, professors, journalists, anyone who wishes to participate for any reason or none at all can do so if they wish. Since there is no membership, it doesn't make sense to vote, so, the IETF doesn't vote, everything produced is a result of consensus. The IETF has no enforcement powers. Don't want to implement IP the way it says? No problem, your stuff doesn't interoperate of course, but that's fine. Don't fancy obeying the MUST conditions in TLS? No problem, perhaps now you don't have any security, but I guess you didn't want it. This would be a bad way to run a country, but the IETF doesn't run a country it just represents an ongoing consensus about how the Internet works.

Rust can, and perhaps should, adopt a very similar approach.

[EduardoMengesMattje]

A very important talk, especially the end message.

[alexmc]

Going to be honest, a lot of this went over my head

[richsadowsky]

I really enjoyed this talk. I once worked on a C++ IDE product called Symantec C++ (evolved from Zortech) with Walter Bright. Also was in the Borland and Microsoft C++ landscape before I went to Symantec. I've been in puppy love with Rust since 4th of July this year. I dedicate about an hour a day at least to either reading from books, code, or examples, and other resources. I agree with your core premise that these two languages will be working together for some time to come. I'm old and still like to learn programming languages printed on dead trees. I am not deep in the language community for 25+ years now, but I like to understand the systems level, programming language, and how to run on embedded metal up to triply virtualized cloud abstraction. Rust caught my attention for a fun learning adventure. And wow indeed it has. I also collected SoCs and MCUs during the pandemic so I find fun little things to in Python, C++, or Rust. I've been working on edge computing in my day CTO job. Good stuff.

[NotherPleb]

C++ at this point it's a mess of bad decisions and the fixes for it. It tries to do everything modern languages do but it's bad at it.

[honza_kriz_bass]

Haven’t laughed so hard while watching a tech talk for a loooong time 😂 awesome talk!👌🏽

[KeBerds]

Bravo. Thanks for sharing your experience!

[robertoprince]

Where can I find the presentation disp?

[nextlifeonearth]

If you try to make the if let Some(x) = function() {.. to c++ I wouldn't go for if (auto x = function) {.. because it simply isn't the same. x isn't the object you intend in the C++ example while it is in the rust example.
The c++ example would be something like:
if (auto optional_x = function() && (auto& x = optional_x, true) {
Is it janky? Yes. The compiler will take care of the if true at the end and you get the x declaration inlined when it's determined to be correct.
Then there's the fact that if the c++ version returns a ObjectType& it will be copied if you use an auto without a &.
Magnificent creature, C++ that is, but it's jank.

[Voy]

that tweet was right :) I co_guarantee it