@Mithical I think that, at least for me, it's...yet another reminder that the old Stack Exchange is gone has been killed off.
I made my Stack Overflow account when I was 12 (don't tell the lawyers), and SE ended up being a critical influence in my personal growth and development. For most of my teenage years it was the only place that I got to work & socialize with other software developers.
@Aibobot yeah, I knew I was underage when I joined...every online service has a box saying "I promise I'm 13", and of course I checked the box anyway
I had a blast; I got involved in SOCVR, SOBotics, and Charcoal, where I got to write code, hang out, clean up the sites, goof off, and make mistakes (and learn from them). I learned how to write, how to communicate online, how to work on programming projects with others, and how to be a part of a team.
I think they should be notified because long-term suspensions are levied for serious offenses, and it's possible they may commit recidivism right after expiration
@NobodyNada but most important were the people who put up with me, helped me grow, and acted as role models. I was never nearly as involved as Art, Undo, or Mith -- but I knew they were my age, and so I could relate to them and really learn from how they conducted themselves.
What would you be proposing? A community flag on their next post, or perhaps just an inbox notification when the suspension is about to end? Also, how would it work for cases like network-wide suspensions?
@SonictheK-DayHedgehog ... if they repeat the behavior, they'll get noticed. The intent of suspensions is that you should be able to come back and start relatively fresh.
Meaning no offense to your idea, but I always thought that the end of a suspension was a beginning of the "No grudges" ideology. They served their time and all.
Telling the mods that the suspension is ending soon just leaves the mods looking for bad behavior. Also... outside SO, the mods generally remember users who made it to a year... and on SO, they suspend enough people that it could be annoying or even missed in all of the other notifications.
But if the risk of recidivism is on the higher side, it may be worth letting moderators know somewhere aside from a notification, of impending suspension expirations so they can pay attention and watch for immediate recidivism the moment it expires.
If I were a mod, I imagine I would have my finger on the pulse of such behavior, and the users who typically participated in it. But, I'm no mod, so I don't really know how much an addition like that would affect the workflow, as it were.
I understand. I still think it'd be noisy and not particularly helpful if all long suspensions triggered that. They issue a lot of year-long suspensions before deleting socks.
So... Something the SO mods do a lot for spammers is suspend them for a year and then destroy them. Do mods have to send a message along with something like that, or is sending a mod message optional in cases of spammers? I imagine if it's not, they have a template they use for spammers, but...
and on the long term - if spammer mitigation stratergies are effective, we won't need to worry about them in a year, they'll move on to easier pickings
@Spevacus Technically, yes. All suspensions other than network suspensions must come with a mod message. The SO mods generally just send "goodbye" as the message.
Depending on what the suspension was for, a lot of the accounts may have been abandoned completely and never return, so notifications about them are noisy.
I suppose a Charcoal admin could propose a feature request for a particular block addition, or mention it in the TL/Mod Team and see if it could be done on the down-low. Knowing them, I'm sure it'd have plenty of evidence to ensure its accuracy and what have you.
Eh. I showed it to Shog before we ran it and he pointed out that it was good but somewhat... I think he said naive... so I can only assume that Art's explanation isn't complete.
Still a great post but don't assume it knows everything;)
@Spevacus In some cases, there can be an advantage to catching that spam after it's posted instead of before. There's so many built-in checks already that it's really hard to post spam -- so once the post goes through, the spammer thinks they won and goes on to spam somewhere else, while we hit 'em with autoflags.
@NobodyNada It's nice to think that, of the spam that gets autoflagged into oblivion, not a single one of the bots or bot owners that posted it realizes that it got blown away in the blink of an eye.
@Catija Once someone claimed that their brother had used the family laptop and the user's account to post something suspension worthy, one of the comments from staff said to use the contact form and that "they had spider on the user's wall" so they should be certain to make their complaint accurately. So I assume that the reality is that SE employs modern technology (not a spider) to examine usage patterns and add the results to SpamRam. --- Perhaps someone recalls the conversation / comment.
@Spevacus There's an automatic 14-day suspension levied on account destruction. I believe the SO mods are aware of this, and do this because they've found that to not be enough.
The big challenges for spam are cyborgs, people with tools to assist them. You gotta have better tools assisting your own people, or you'll spend more energy defending than they do attacking. This is the real value of SD
Interesting, thanks. At the time, I think I classified anyone older than me as an adult.
@Shog9 I think the irony to me is the value of SD, yet six years later (as far as I know) it remains an unfunded project built by a bunch of mostly teens bored of pushing buttons.
I think that the community-run aspect is part of the reason for its success -- we can experiment and break things without causing outages or angering managers, we don't have to justify our existence during layoffs, and we can opening the project up to the whole community without requiring anyone to make huge time commitments
while, at the same time, we get instant priority support from devs and CMs whenever we have questions or problems with bugs/changes on SE's end
that being said, "community-run" only works if you invest in your community. The environment that brought a bunch of teens together to write a spam-detection engine was not a product of the kind of decisions that were made today
I mean... I'm pretty sure there are a lot of questions about whether a ToS is even enforceable? unless I'm confused about things... maybe I'm thinking about EULAs.
But there's a huge difference between something that most people aren't even aware of and is mostly just a check box that people tick without looking at the 20-50 page doc... and a concise (though longer than it used to be) document of expectations that any moderator is required to accept before being granted access that, for many companies, only staff would have.
If PII is misused - which has happened - the company is liable for that and needs to be able to show that they are making a best effort to protect that PII. This is, in particular, something that is likely of greater concern after GDPR than before.
To me at least, the solution is then to reduce PII sharing. The experience is probably different on other sites, but I rarely look at PII unless I'm verifying that an account is actually a duplicate.
Many (most?) mods never use their PII access. The SO mods use it constantly. In the grand scheme of things, reducing PII access is definitely something that's been discussed but would require a lot of work to do in a way that didn't reduce a mod's ability to identify socks ... or trolls or whatever. And while it certainly looks lazy, the reality is, as painful as this change is... it requires very little dev work.
We have a huge backlog of things we need to build to make the site more usable and improve the user experience - and we have a tiny team of people to build that - there's ~5 devs for the entire public platform. This excludes teams like engineering who deal with the entire product but, for the team that actually builds stuff, there's only five.
... It is. We're doubling that team over the next year.
Just like we have plans to increase the CM team, too... but it takes time to get people hired, on-boarded and actually adding to the product in a big way.
Yeah, I mean I can't fault the network for taking time to solve its problems. But I'm sure this has been known for years. Just the problem of being a poorly-funded company relative to other large tech firms I suppose.
I think the company has been struggling to find a solid source of income for years. The funding this summer is making stuff possible that wasn't but we also need to show those funders that we were a smart investment - so we're investing in the core public platform but we also have to continue to improve our paid platforms and products.
And, well... I don't know if any of y'all have been paying much attention to how many jobs are on offer during a global pandemic... but it's not been great for SO Jobs.
SE is a SaaS company now, there's no way that doesn't affect the priorities for development (and the mod tools were low on the list even before that). The only reason I think it's still possible that SE will spend resources on making mods without PII work is if the current situation is declared too much of a legal liability
That's likely to come more from our Security department than Legal. Though, our head of Security seems to be a really solid and reasonable person. I've appreciated the interactions I've had with her.
@JourneymanGeek True, although I'm just thinking like what if you cut off PII access for mods on say the bottom 50% of SE sites by traffic? How long would it take to notice? (It would honestly take me a while.)
@hichris123 hypothetically - I would trade off knowing an IP for reliable ISP/Host recognition, and I've found socks at least once cause the user was john.smith@outlook.com and stackechange@johnsmith.com ...
so the tradeoff is building extremely robust tools that would replace that functionality
@Catija I like security folks in general :D Other than the hacker wannabees :D
Ha, yeah. I think it's probably because ES is low traffic, but I've seen maybe two or three socks in my years moderating. Mostly, I find unregistered and registered users on the same question and see that they have the same email, or john.smith@gmail.com versus john.smith@hotmail.com.
I think you could get to 80% functionality with a relatively simple tool. The basic idea is to not show emails and IP addresses, but have a tool that tells you if a certain user matches those with any other user.
If you get e.g. "userX was seen 15 times from IPs used by userY, and none of these IPs are associated with any other user", that should be enough information
@JourneymanGeek :D From the little of Security training I've done (in a prior job) the guidance on risk assessment is really regimented which, I think, helps them be thoughtful about risk rather than reactionary and easily spooked.
I haven't played that game for a week now. I've been crocheting quite frantically as I always finish a blanket before the year ends, and this one needs work still XD
I'm just saying, I'm just saying. I've seen people so religiously commit to this "age of consent/drinking/whatever" like everything before it is illegal and everything after it is what a normal healthy functioning adult would do.
Well I come from a place where it's so weird you could say the age of maturity is considered 18 or 16 or 15 or even 13, depending on which law you're looking at. I think some things are as wrong when you do them when you're 17 as when you're 19, and some restrictions don't make any sense, unless you set the bar pretty low.
Guys, what do you think about this answer about selfharm/suicide that is under the same question that the official policy links to an answer of. I feel it is outright wrong, and perhaps even harmfull to keep on the site. Yet I generally object to deleting wrong answers, but here there are severe real-life consequence possibly. meta.stackexchange.com/a/243701/361484
TL;DR: Do not copy/paste. Do not suggest a suicide hotline. Don't lie. Don't treat them like a baby. Don't try a quick-fix.
Suicidal people are encouraged to be open, and talk about their suicidality---it helps. But the instant you do, people begin acting very weird. Anyway, here's my tips:
@Luuklag Well, not every person is alike. As far as I'm aware the policy hasn't changed that much, though on reading it twice now there's no more mention of actually handling the post by deleting it.
At least a suicide hotline has people trained and willing to handle this stuff, instead of it being dumped on people that don't have.
I've dropped Cesar a comment too, it would be nice to know if we can at least still clean up messes ASAP instead of being expected to play suicide hotline.
@rene Not really fair though, the post does say that we can choose not to engage... ;) So we're not really expected to play suicide hotline, but it's still a weird choice if the choice is between 'do nothing except to escalate, then wait hours until a CM has time to handle it', leaving the posts up for all and sundry to see and stuff, and 'play suicide hotline'.
@JourneymanGeek Have you ever had any credible ones?
I guess I was slacking when proofreading the policy XD I now see it also includes "Please do not discuss these situations in chat rooms, particularly in The Teachers’ Lounge." which to an extent makes sense but is also... TL is kinda supposed to be there to discuss these issues if people need it? Especially if they're supposed to sit twiddling their thumbs until a CM arrives.
@SonictheK-DayHedgehog I suspect its a matter of different means of support
TL's the primary means of peer support across the network
so it would be strange, without context, to specifically discourage using the room for support
On the other hand, in situations like this, there might be a time urgency, and its certainly going to be nerve wracking until you know someone is looking at it
@Tinkeringbell I think that "don't discuss these cases in the TL" is there to prevent moderators dogpiling into a threat or chatroom. Usually, the moment someone discusses something contentious, the chatroom in question get bombarded by moderators. Well-meaning moderators also can't help responding to comments, deleting comments, etc etc. Generally it creates noise in a situation where calm is of prime importance.
@Tinkeringbell That's very true, and I've used TL in the same way myself. But I'm talking about situations like this, where too many well-meaning moderators can cause more of a problem outside of the TL.
@Snow I've luckily never seen mods pile on onto a threat of self-harm... but that might be because such things would be closed and deleted before they got the chance under the old policy.
Most often it was 'provide link to policy, then chat a little to blow of steam'
@Luuklag Probably a combination of excuustruus, making them feel good about involving a junior developer, and having a developer from an outside team ;)
@Luuklag Haven't seen any that is. Some are just less bad than others. The ones I'm at now love to segregate people based on age gender into 'interest groups'.
