[WIP] Scribble: Advocating Rust #4
Conversation
_posts/2017-04-05-promoting-rust.md
Outdated
|
|
||
| Let's start by things that you definitely shouldn't do, if you have any interest of convincing someone. It's good to always have these in your head, just to ensure to not do them accidentally. | ||
|
|
||
| First of all, don't ask random people to consider Rust or an _existing_ project. They already chose the language they want to use for implementation and as long as they don't come up with a specific problem that brings up the language discussion on again, the topic is through. No one wants to waste more time on topics already discussed. |
There was a problem hiding this comment.
Might be worth distinguishing between random people and a project you're involved in.
_posts/2017-04-05-promoting-rust.md
Outdated
|
|
||
| Second, don't kick people when they are down. The next security incident in libinternet? This is not the place for drive-by advocacy. Sometimes, this is called [Ambulance Chasing](https://en.wikipedia.org/wiki/Ambulance_chasing). "Rewrite in Rust" just isn't a practical suggestion for existing projects with no Rust programmers around. I'll get to that in a minute. | ||
|
|
||
| Third, don't make the impression you take project maintainers for amateurs for their choice of technology. While there's certain things many people would like to go away, it's not like people thought about their choice of technology for building something. The modern Tor probably shouldn't be written in C, but Tor was born at times where it was a good solution. |
There was a problem hiding this comment.
Calling Tor out specifically here feels like it may unnecessarily start a flame-war argument. We could just mention "applications which are security-sensitive have better options for programming languages, but many were written before these options were available"
There was a problem hiding this comment.
I think an example would be good. Would you have one?
There was a problem hiding this comment.
I wouldn't give a specific example because a) it's picking favourites and b) it gives people something to mentally latch onto.
|
|
||
| Finally, here's a few things that you definitely shouldn't do, if you have any interest of convincing someone. It's good to always have these in your head, just to ensure to not do them accidentally. | ||
|
|
||
| First of all, don't ask random people to consider Rust or an _existing_ project. They already chose the language they want to use for implementation and as long as they don't come up with a specific problem that brings up the language discussion on again, the topic is through. No one wants to waste more time on topics already discussed. This is a little bit different if you are already part of a project, but even there, take care to waste project time and make the discussion fruitful. |
|
Just to be clear, this isn't even an incomplete draft, more a scribble. |
|
Sorry for letting that slide. I'd like to get this blog post finished this month or decided to not publish. Maybe a fresh version with "advocating Rust in 2018" makes sense? @rust-community/community-team do you think this is still a worthwhile text? |
This is just an incomplete scribble currently, but if you have good techniques to advocate Rust, please feel free to change them.
At the end, it should have a "Do:Don't" ratio of about 3:1. :)