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

Why AWS loves Rust, and how we’d like to help

aws.amazon.com

Post reply on HN
Print view View original

101–110 of 338 posts

Re: Why AWS loves Rust, and how we’d like to help

#101

Post by _greim_ »

LukeEF wrote: post #32

Having developed and launched a database (TerminusDB) built on a Rust core, this makes me slightly nervous. Feels like the evil empire will bring both great riches and great peril to the fantastic Rust community (part of the reason we picked Rust). Guaranteed sponsors for conferences, but also a certain malignant cynicism (am I being too harsh?). AWS will also aggressively compete for the best talent, which is tricky…

> AWS will also aggressively compete for the best talent, which is tricky when you are small and open source.

Barring surges in demand at month-scale timelines, this isn't a zero sum game. Competition for Rust talent will result in an increase in the amount of Rust talent. Even better if you're open source; a growing population of Rust devs roaming the planet can only help you.

Re: Why AWS loves Rust, and how we’d like to help

#103

Post by loeg »

wolco2 wrote: post #69
saagarjha wrote: post #51

Earlier quoted context omitted.

> the browsers are all C++ You mean one is already shipping Rust and another seems to be on track to shipping it soon?

In fairness the one using Rust has fallen apart on my windows 7 machine since the rewrite. It would be unfair to blame rust. But that example doesn't provide confidence that switching from c would be a great idea.

I think it's weird to blame Rust for that. I don't want to rush to blame Windows 7, but do consider that Windows 7 is massively EOL (2015; extended support also ended early this year) and I would not fault Mozilla for not doing any testing on that platform anymore. I don't think this has much, or anything, to do with Rust.

Re: Why AWS loves Rust, and how we’d like to help

#104

Post by dagmx »

andrewmcwatters wrote: post #50

Am I just not "with it?" I don't see the appeal of Rust, but it gets the hottest talk in HN-town. Maybe it's just because I haven't worked long enough with C or C++ to complain about them (only about a decade or so), but I don't have any issues with those languages. And I find it wholly unappealing to abandon the existing corpus of literal decades of knowledge having been poured into those language ecosystems all in…

> I haven't worked long enough with C or C++ to complain about them (only about a decade or so), but I don't have any issues with those languages

This may be part of the reason. Rust improves on those languages in many ways including better safety, package and build setups as well as doc setups. It's genuinely a breath of fresh air setting up and maintaining rust projects versus C++ or even Python.

> I find it wholly unappealing to abandon the existing corpus of literal decades of knowledge having been poured into those language ecosystems

Rust builds on these ecosystems. It uses LLVM as its backend compiler, much like clang does for C++.

Additionally it has great support for CFFI so it can integrate very well into existing codebases with other languages.

> I just have too much experience in the industry to throw away my time and play with toys.

I read this as prejudice. Rust isn't a toy as can be seen by the large number of companies and projects that are putting significant efforts into it.

Conversely, perhaps having entrenched experience in the industry means you're predisposed to approaching new paradigms in the field with higher cynicism than may be warranted?

> It's ugly as sin, and I know I'm not alone in that opinion.

I think that's very much subjective. I do know many programmers who find it ugly, but many like myself who find it write clean to read.

Certainly I find it significantly easier to grok than C++ or Objective C for example. It's possible to write very elegant rust code that can verge on being Pythonic in nature. It's also possible to write very ugly code too.

> there was literally no one using Servo, the site didn't tell you how to actually use it

Servo is a research project from which portions have been taken and integrated into Firefox. That means a lot of people are using servo whether directly or indirectly.

Re: Why AWS loves Rust, and how we’d like to help

#105

Post by k__ »

andrewmcwatters wrote: post #50

Am I just not "with it?" I don't see the appeal of Rust, but it gets the hottest talk in HN-town. Maybe it's just because I haven't worked long enough with C or C++ to complain about them (only about a decade or so), but I don't have any issues with those languages. And I find it wholly unappealing to abandon the existing corpus of literal decades of knowledge having been poured into those language ecosystems all in…

The Rust community alone is so much better than anything I've seen in my 15 year programming career.

The communities of statically typed programming languages were especially elitist and gatekeeping.

The Rust community is welcoming.

Re: Why AWS loves Rust, and how we’d like to help

#106

Post by blitblitblit »

steveklabnik wrote: post #68
blitblitblit wrote: post #63

Earlier quoted context omitted.

Does this mean the "Media Guidelines" is going away? If we want to redistribute cargo or rust with some patched changes, I'd like to be able to do it without coming back and asking for trademark approval every time. Even if the resulting binary says "unofficial/unsupported build".

I can't make any particular statements about what an organization that doesn't even exist yet may or may not do with its property. Who is "we", in this case, out of curiosity?

Any OS distribution, e.g. Debian maintainers should be asking permission to patch rust per Media Guidelines, but it doesn't always happen.

Re: Why AWS loves Rust, and how we’d like to help

#107

Post by foldr »

geofft wrote: post #36
johnnycerberus wrote: post #27

Earlier quoted context omitted.

Applications of Java and Rust do not overlap. As a big data engineer, Rust does not provide any advantage to me. Good luck rewritting all the big data tools from Java (Elasticsearch, Spark, Kafka, Hadoop, Neo4j, Deeplearning4j, Cassandra, Solr, Arrow, OrientDB) and all the scientific stuff from C++ (Eigen, Tensorflow, PyTorch, Jax, Halide, OpenCV and the multitude of resources for CUDA and OpenCL) and databases (MySQ…

Why do you need to rewrite them? Rust can call C++ code, just like Java and Python can. (I note you list Python libraries like Tensorflow and PyTorch as C++, which is accurate because they're written in C++, but they're certainly heavily used from a non-C++ language.)

It's easy to call C++ code from Rust but it's often not trivial to wrap a C++ library with a safe Rust interface. C++ APIs often have implicit ownership structures that are difficult to encode using Rust's lifetimes.

To be fair, that doesn't put Rust in a worse position than any other language binding a C++ API. It's just that Rustaceans tend to have higher expectations in terms of safety.

Re: Why AWS loves Rust, and how we’d like to help

#108

Post by brandmeyer »

cogman10 wrote: post #77
andrewmcwatters wrote: post #50

Am I just not "with it?" I don't see the appeal of Rust, but it gets the hottest talk in HN-town. Maybe it's just because I haven't worked long enough with C or C++ to complain about them (only about a decade or so), but I don't have any issues with those languages. And I find it wholly unappealing to abandon the existing corpus of literal decades of knowledge having been poured into those language ecosystems all in…

I can give you a few reason you might like rust. - Instant boot time. Because rust is statically compiled, there's no runtime startup or anything involved with it. From dead to alive is pretty much instant. - Low memory footprint. Because rust does not use a GC, it doesn't need all the bits and pieces of memory storage that modern GCs require. The memory allocated is (close to) the minimal memory needed to run the ap…

> Instant boot time. Because rust is statically compiled, there's no runtime startup or anything involved with it. From dead to alive is pretty much instant.

Dynamic linkage is not the slow part of application startup. The slow part (for applications which are slow) is almost always the gathering of non-compiled assets from disk.

IMO, the lack of dynamic linkage is Rust's #2 weak point, second only to slow build times.

Re: Why AWS loves Rust, and how we’d like to help

#109

Post by johnnujler »

steveklabnik wrote: post #78
johnnujler wrote: post #47

Earlier quoted context omitted.

Great thing but I feel these companies are just too clever for Rust. Rust is a developer’s delight and I don’t want it to become like C++, and I feel the C++ standard got worse especially after these so-called great companies became a part of the standards committee. Or take the example of Web, made worse after the Googles and the Apples of the world got in. So I have mixed feelings about too many clever people doing…

The language team is the sole group of people who decides how the language itself changes over time. Entry into that team is determined by the consensus of the team itself. Anyone can comment on proposals, and of course, a company could hire those individuals, but it's not possible to become a decision maker (or stay one) without the explict consent of the existing set of decision makers. That being said, we (that is…

I am genuinely curious to know more about this. Some of these companies dedicate multiple teams towards an OSS project, which I’ve seen first hand with projects like llvm. So isn’t it possible that the involvement of these companies could actually systemically turn things as most of the people working would be their employees and that the scales could be manipulated easily by these companies(need not be intentional). For example I remember “std::string_view” in c++ being one example which Google pushed and even got into the C++17 standards despite its bad design. Although I might be wrong, but my guess here is that Google had a use for it(which we know it had) and that lot of people on the standards committee being from Google led to a systemic bias. Couldn’t that happen with Rust?

Re: Why AWS loves Rust, and how we’d like to help

#110

Post by gpm »

andrewmcwatters wrote: post #97
wwright wrote: post #61

Earlier quoted context omitted.

> but still came away with the feeling there was literally no one using Servo Code from Servo is in every single copy of Firefox, so it's pretty popular as code goes. > And I find it wholly unappealing to abandon the existing corpus of literal decades of knowledge having been poured into those language ecosystems all in the name of whatever it is that Rust delivers on. I just have too much experience in the industry…

> Code from Servo is in every single copy of Firefox, so it's pretty popular as code goes. This is akin to saying the DWM is in every copy of Windows so it's the most popular graphical compositor of all time.

It really isn't. The entire purpose of servo was to be a vehicle for r&d with the eventual goal of merging code back to firefox. It succeeded. No one used it because that really wasn't the point.
Post reply on HN