Also, for some stats:

[misc@ajani freenode]$ grep -ri "you guys" \#fedora-* |wc -l
275
[misc@ajani freenode]$ grep -ri "guys" \#fedora-* |wc -l
840
[misc@ajani freenode]$ grep -ri "hey guys" \#fedora-* |wc -l
55
[misc@ajani freenode]$ grep -ri "hey folks" \#fedora-* |wc -l
148

that's on the log of my irc client, and I am not sure how old they are (but I think around 3 years old given the uptime on that specific server).

For the regex I most commonly see these phrases: hey guys, hi guys, any guys, you guys and morning guys

Examples like
"Hello, are any guys around?"
"Good morning guys"
"Are you guys going to the meeting?"
"Hi guys, how are you?"

The bot could reply with something like
how about saying "everyone" next time or "all" instead of "guys"? Inclusive language makes everyone feel welcome.

Discussed in 2020-08-27 meeting.

I dropped a comment in the GitHub Pull Request to let @ryanlerch and @nphilipp know about this ticket and our plan to cover this in an upcoming meeting.

In the meeting, I hypothesized this idea needs special consideration to the human factors, especially how an auto-correcting IRC bot might be perceived by other people. We should think through possible scenarios and reactions we might get, and figure out if there is a friendly/inclusive way to implement this without it feeling like a hand-smack.

The framing on avoiding a "hand smack" is IMHO not great.

It really give the impression that the focus is the feelings of the poor man (and I do not mean "man" in a gender agnostic way, gender stats in the opensource being what they are, that's likely a man who will be corrected by the bot) over the problem it currently cause WRT diversity (again: https://journals.sagepub.com/doi/abs/10.1177/0146167211406434 , I let folks spend 40$ or use sci-hub to read it ).

Thanks @jflory7 for pinging us, after I closed the other ticket I didn't notice the subsequent commo there.

I'll respond to both the comments in the supybot-fedora ticket and in this ticket here, to keep everything in one place.

So while I am sure that's not the intent, refusing the patch on technical correctness ground really send the wrong message on D&I given the context. …

You're right, the decision to not include the functionality in supybot-fedora is because it's out of scope, i.e. not specific to Fedora. If the "naked ping" thing weren't part of it yet but submitted today, I'd refuse on the same ground. Regardless of my opinion on it, some people find it worthwhile having, so I won't rip it out just for the heck of it. But as far as I'm concerned, its presence is an exception and not a good argument for including more functions that aren't specific to Fedora but potentially useful for a wider audience.

What message this decision sends regarding the objective of your patch, that this is a tool to further inclusion, is coincidental and secondary to the question of "is this in scope for the project or not".

The same goes for "it upset people". Sure, some folks are likely upset, but not enough to submit a patch or even a ticket it seems from a quick reading of that repo. And while folks are upset, we should also take in account that folks likely changed their habits, and I guess i could do some analysis from my irc logs to see that.

I mentioned that I know someone whom the naked ping messages upset to show that not everybody takes being admonished by a bot equally well. Some people might change their habits due to it, others might feel being condescended to, especially in cases where the bot gets the situation wrong.

Unlike naked ping, I think it is fairly uncontroversial to say that gender imbalance is a huge problem in the industry, and worst in the free software movement.

Here's where I guess I should state that I'm all for inclusion and making people conscious how their use of language affects it.

However, I'm deeply skeptical that bot-driven automated messages are universally helpful tool to accomplish that:

• Some people might gain insight and be more conscious about how they communicate.
• Some others might be annoyed to a) adjust their behavior just enough so it doesn't get caught by the bot or b) leave.
• Some might ignore it completely, unless we give it teeth to say kick people from channels (which I'm wholesale against, a bot can't fairly judge the situation).

Only one of these potential outcomes is really desirable IMO.

I'd go as far as saying that a bot—even some AI-driven Siri- or Alexa-like thing which is way beyond what we're talking about here—is no substitute for human interaction especially regarding a nuanced topic like this. I fear that automating this conversation, delegating it to a piece of software sends a message that we, i.e. the community, don't care enough to put the effort in to hold a potentially difficult conversation ourselves. It feels about as real and convincing as these automatic "Happy Birthday" mails I get from a number of online services I have accounts with.

We ran out of time to cover this ticket in today's meeting, but we will revisit. A lot of team focus is currently on getting Fedora Women's Day events ready for October.

I was curious how much engineering work would be needed to opt for a 1x1 private message with the bot when someone uses the "hey guys" regex above, with an option to opt-out from receiving those specific messages.

Reading the above, I agree that an admonishment in a public channel from a bot is not a very friendly feeling. I think a 1x1 message is less confrontational. We should give the person receiving the message the option to opt-out from being messaged if they choose. This gives the people in User Story no. 2 (above) the option to disable the annoying behavior if they desire.

I do not think the D&I Team has engineering resources to own this work right now, but we would be happy to advise or offer early testing for such a feature in our channel if someone had the time, interest, and capacity to add this to supybot-fedora.

Discussed in 2022-04-25 team meeting.

We agreed to close this ticket as stale, in the interest of narrowing the scope of the team and keeping our work focused. We felt this was still a good idea, and @riecatnor brought it into the fold with the moderation guidelines discussions happening in the Council ticket tracker. But for our purposes, we will close the ticket as stale so we can keep momentum and focus our work in a few key areas.