3 ms·
I'll probably be writing a slightly longer response post to this later, but for now... EDIT: here it is: http://words.steveklabnik.com/fire-mario-not-fire-flowe
by steveklabnik 10y ago
I'll probably be writing a slightly longer response post to this later, but for now... EDIT: here it is: http://words.steveklabnik.com/fire-mario-not-fire-flowers http://words.steveklabnik.com/fire-mario-not-fire-flowers
I think the core of it is this:
> Safety in the systems space is Rust's raison d'être. Especially safe concurrency (or as Aaron put it, fearless concurrency). I do not know how else to put it.
But you just did! That is, I think "fearless concurrency" is a better pitch for Rust than "memory safety." The former is "Hey, you know that thing that's really hard for you? Rust makes it easy." The latter is, as Dave[1] says, "eat your vegetables."
I'm not advocating that Rust lose its focus on safety from an implementation perspective. What I am saying is that the abstract notation of "safety" isn't compelling to a lot of people. So, if we want to make the industry more safe by bringing Rust to them, we have to find a way to make Rust compelling to those people.
1: https://thefeedbackloop.xyz/safety-is-rusts-fireflower/ https://thefeedbackloop.xyz/safety-is-rusts-fireflower/
- kaoD 10y agoAs someone watching from the sidelines... maybe it's the cynic in me, but I think you're giving too much credit to the "but I don't have safety problems" folk. Judging by how defensive they get, it's just a post-hoc rationalization to why they won't invest time learning it. That's fine but, IMHO, not a marketing problem and just good old resistance to change. There's an endless stream of excuses to choose from, no matter how you pitch Rust to them. I'm all up for a marketing shift though. I agree Rust is much more than safety and I think graydon2 missed your point there. The fireflower simile is excellent. Plus I've seen lots of people confused about Rust, including here on HN. Those are the people to whom marketing failed and should be targeted better.
- bluejekyll 10y ago> I think "fearless concurrency" is a better pitch... I would go a step further, "fearless programming". Though I would hesitate on 'easy'. Rust gives you fearlessness in all the things, but it does mean learning new style and discovering new solutions to old problems. To fully understand 'Send' vs. 'Sync', for example, means really groking the Rust type system. Once you get the type system, then fully utilizing it with the expressive generics becomes unlocked, and then at that point you've transcended from Rust dabbler to fully fearless Rust user. Once this world of development is unlocked to you it is mind-blowing, but it is a journey to get there, and not everyone will have the heart to make it. It comes in stages, is wonderfully rewarding, will make you a better programmer in you other favorite languages, but I think we should be careful with statements like 'Rust makes it easy'. It does make hard things easy, but only after you've fully embraced Rust. This feels more accurate: "Hey, you know that thing that's really hard for you? Rust makes it fun."
- nullc 10y agoMaking software free of data races and memory safe (assuming you don't use any unsafe code...) is still a long way from being free of serious defects. Rust is cool enough that it doesn't need to be promoted with excess hype.
- GolDDranks 10y agoI would argue that if the Rust project would have just one mission statement, it wouldn't be "create a safe systems programming language". It would be "move towards a world where safe systems programming is the norm". What's the difference? Both of the statements have the premise that Rust is – and ought to be – a safe systems programming language. However, the latter captures not only the REAL goal, but also the nuances and tensions: while safety is indispensable, we must do something else too, for the programming society to accept the safe tools we are trying to promote. That means ergonomics, that means performance, that means ease of use, that means wide availability – and that might also mean advocation of visions of a better world, which is what this blog post of Graydon's does.
- steveklabnik 10y agoI really, really like this. Thank you. Well put.
- btown 10y agoIt's important to get the nuance of this statement right. Consider, for instance, the PSF's mission statement: > The mission of the Python Software Foundation is to promote, protect, and advance the Python programming language, and to support and facilitate the growth of a diverse and international community of Python programmers. "Facilitate the growth" does not imply that dominance of a certain practice (whether that is use of Python, or adoption of its core principals) is a goal. So you have an incentive for the PSF to say "our international community is growing at a sufficient rate" and focus instead on "advancing the language," which may or may not be aligned with adoption. In such a framework, it becomes easy to justify backwards-incompatible major releases that, regardless of whether their opinion is justified, many users consider to be user-hostile. Framing Rust towards a larger mission that implies user adoption of core principals as a fundamental goal seems much cleaner in that regard, and could conceivably avoid similar pitfalls.
- nickpsecurity 10y agoI think the problem is that Graydon is better at technology than marketing. Getting language adoption is largely a marketing problem combining what those in control of the language push and how those in the community pull. Rust's success comes from doing that dynamic correctly. If it was just safe and not C/C++, it would be another Ada, Modula-2, D, etc. It's important to market all the key benefits of it in a way that lets potential users know it will help them solve problems faster and with less trouble down the road.