Project idea - "friendly chat"

[Evan Czaplicki]

Motivation

I was reading a book recently, and the main character wanted to send a telegraph to his employer that basically said "I quit" in less friendly language, and the telegraph operator said no. That sort of language was not permitted in telegraphs! The idea that standards of discourse could be built in to a medium was both shocking and kind of obvious.

Why doesn't this exist?

Right now, I think there are structural reasons that such a tool does not exist:

1. More anger means more "engagement". More "engagement" means more ad revenue. So many companies would take a serious financial hit if people got along better.
2. Many communication tools are designed for workplaces, like slack and email. In a workplace, there are expectations of behavior and mechanisms like "You are fired" if someone is derailing projects, wasting time, etc. In that setting, it is fine if the chat app does not have any help at all for that sort of behavior. There are other mechanisms.

So it sounds like there is a need here that may be best filled by open-source projects because our needs do not really match corporate needs. There are also lots of folks who specifically prefer to use open-source tools and would love to have something like this.

Specific Ideas

Timeouts - I think video game streamers with 5k+ people in their chats are on the frontier of these issues, and it seems like timeouts work reasonably well. They give folks a couple minutes to take a walk and reset. They are also much lighter than outright bans. It's just a way to say "cool it" without a massive discussion. It is crazy to me that discord does not have this.

Pre-reviews - I think it would be better if conversation guidelines could be applied before a message is sent. For example, rather than have a bot try to do this once it is too late, the chat app itself could give some suggestions of how to edit before letting the message through. In my dream version of this, it could help me tone down the emotions that can sneak into my writing sometimes. I think that could help me be more effective in discussions where I may be quite upset. (This also transforms an adversary "Don't say that" into an assistant "Maybe X would sound better?")

Anyway, my goal here is not to start a debate about whether these are good ideas or not. I just wanted to put some of these ideas out there and see if it inspires anyone!

[Robert Cooper]

Hey Evan,

I have a startup that is working on this exact thing. My goal is to have a "self moderating" chat for thousands of users to use at once. I have an alpha version out now being used by a large sports community, and it seems to be going pretty well. The front-end is in Elm, and the backend is Elixir.

Let me know if you're interested in knowing more about it!

[Evan Czaplicki]

That is really interesting!

Do you do anything like making messages with profanity unsendable? I wonder if the tradeoffs of that vary by the shared goals of a given community? E.g. I feel that profanity lacks the specificity for clearly communicating frustrations in a goal-oriented technical setting, but in a community that is more about emotional expression, perhaps it is really helpful.

Anyway, is there somewhere I can take a look and check it out? And feel free to message me privately if you don't want to share info publicly at this stage :)

--
You received this message because you are subscribed to the Google Groups "elm-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to elm-dev+unsubscribe@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/elm-dev/c0324f05-a454-4c0c-8c78-c50225c47f9c%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.