@Spevacus- sorry for the delayed response. Got ppulled into something.
I'm glad you are finding the post interesting. I've got more questions still to come. We're gonna try to do that weekly.
I think the work that's happening here is fascinating, though I will admit that I'm probablyl the wrong person to engage too deeply there. However, as we get my trust and safety team staffed up, I'm going to suggest strongly that they spend some time here.
Also, for anyone who might be interested, I've just posted two roles for community managers, specializing in trust and safety.
When I was at Wikipedia, though, I did a regular 'office hours' series, twice a week (for the first couple of years, and then much less frequently, as we onboarded more people who were doing more interesting work than I was.)
@thesecretmaster- thanks. and @RyanM as well. :). I hope you guys know that my (virtual) door is open to you. If there's anything you need me to hear, or anything I should know, please feel free to email me.
@Ryan M - you'll find a friend in me as regards that feeling toward spammers. I enjoy killing spam, with fire. When we got to the point of eliminating about 95% of it at reddit, i broke open a bottle.
Reddit spam must be an interesting beast with such a variety of topics and formats among the various subreddits. The more rigid format of Stack Exchange makes the task a bit easier, though still a challenge in some cases.
reddit spam is wicked. We trained a couple of different ML models and turned them loose on "warn" rather than "active" status and pitted them against each other. It was a fun time.
In the end we bought an office the shelf tool and heavily customized it. (Spamurai)
...and then came the day that Spamurai shut down. And reddit wept. Or at least the T&S and Community teams did.
To add what Spevacus said earlier, in addition to people always being around to answer questions, we've also got a website that would give a pretty decent summary of how our bot works and how we catch/response to spam.
Hey @JourneymanGeek, totally. It's more like what we had at Wikipedia. (For a good time, go check out wikipedia's recent changes feed but turn bots ON. You'll get a feel for how close we are to bots truly running the world.)
@JourneymanGeek, thanks for weighing in on MSE on the questions page, btw. I appreciate it.
Absolutely. Generally speaking, I think CMs should have (or quickly develop) a proficiency in meta. Hopefully faster than I did, given that my first post there was my introduction.
(I say 'or quickly develop' because i don't rule out the possibility of outside hires when it makes sense to bring us skills we don't already have, or that are difficult to get internally.)
Nic's interesting, he's built a community, but wasn't that visible until after the whole international site initiative fizzed out cause TPTB of the time lost interest
When I was hired, initially as a CM, i told them to expect me to be very low profile for 90 days while i learned which buttons to not step on. Then they went and made me VP, so I didn't really get the operational time as a CM that I would have ideally liked.
/me nods. Pity. But I understand. I've heard some of the history.
And we dont make it totally easy, I know.
Journeyman Geek, if you have the free time someday and are willing, I'd love to sit and chat about your experience and see if there are ways we can make it better for people who come after you.
As you may have seen,I have two jobs posted now. I'm going to be watching the hiring process like a hawk to make sure we dont have any repeats of that ....incident
That's part of why I'm here. :). Hoping that I can sucker.....er..... convince.... someone here to apply for the roles I've got up now, working with Cesar.
I hear you. I'll tell you this much, and I've shared it with recruiting as well.... I'm much less interested in formal qualification than in knowing if you've had a big idea and chased it.
Now, I'm not the hiring manager. But as you might guess I have some influence there.
But it also includes specialists in trust and safety or community building or governance who have done it before.
I'll just say that the team that I have - while absolutely amazing, each and every one of them..... well, I wouldn't cry if they all got cloned overnight and doubled the size of the group.
The first time I threw up a "work at wikipedia" banner, I had high hopes that i might find an anthropologist that knew online communities. We found her in the first 25 MINUTES.
@Mast on the contrary, I'm gonna bump it up to the blacklist - the only instance on the network is electronics.stackexchange.com/a/151827/261257, which has 13k rep and wouldn't be caught
@SmokeDetector user has a low-quality but not abusive question on the site, giving benefit of the doubt and picking NAA
@SmokeDetector third time this post has been reported, can only assume that it's survived spam flags five years ago. Unclear if it's spam, it's not a clear match for the clear spam, but also there was a lot of variety.
[ SmokeDetector | MS ] Bad ip for hostname in body, bad keyword in body, blacklisted website in body, body starts with title and ends in url, pattern-matching website in body (487): There are numerous fragrances to browse by Matt Curan on stackoverflow.com
@Xnero That pattern looks like it's already caught by Mostly punctuation marks in answer, Mostly non-latin answer, Mostly punctuation marks in body, and Mostly non-latin body Append -force to the command if you really want to add this pattern.
@Makyen That pattern looks like it's already caught by Potentially bad keyword in answer and Potentially bad keyword in body Append -force to the command if you really want to add this pattern.
