@Ollie That pattern may be a North American number. If it is, please use a format which starts with an optional 1 followed by possible separator text and has the main number in the format 887-512-4365 where - could be a single alpha character or any [\W_]*+. Alternately, you can add the comment (?#IS NorAm) to the end of the pattern to force also using the alternate normalized form, or (?#NO NorAm) if it's not a North American phone number and it's incorrectly recognized as one. Perhaps try !!/watch-number 887-512-4365. Append -force to the command word(s) if you really want to add the pattern you provided.
@Ollie In order for a "number" to make an exact match, the pattern must end with a digit. Append -force to the command word(s) if you really want to add the pattern you provided.
@Laurel Even with the question, the answer looks like spam, especially since it's posted by a new user, with a suspicious username, and there's no other hits for that domain on SE.
> Mostly non-latin body, mostly non-latin title ---------- Mostly non-latin body - Text contains 1 non-Latin characters out of 1 Mostly non-latin title - Text contains 1 non-Latin characters out of 1
All but two of those posts were written at least four years ago (at a quick eyeball). Given the persistent QuickBooks/Coinbase/etc. spam, wouldn't that be acceptable?
I'm also more concerned with how effective blacklisting that emoji would be.
That there are existing legitimate posts containing it tends to indicate that there may be more such posts created.
What would it gain us beyond the existing detections? Are there posts which it would detect which we're not detecting? Are there posts which would end up autoflagged which we're not currently autoflagging?
Are there other things associated with the emoji which exist in spam, but which don't in non-spam? What might be used to narrow the focus of a detection to something more than just a single ☎️?
@Makyen If there's more that's used to discriminate between TP posts and FP posts (e.g. that the detection is nearly entirely TP), then there doesn't need to be all that much gained. Basically, the benefit needs to outweigh the cost, but that can happen in a variety of circumstances.
I was more concerned with if blacklisting that would add more flagging weight to the posts, but didn't think about it being at the expense of 50+ legitimate posts (and possibly more), and it's only a single character. It wasn't a very good notion. Sorry.
OTOH, that doesn't mean there's not a pattern there which we could be looking for. That emoji is fairly commonly used by spammers, but in order to avoid FP, we'd need to be looking for patterns which are more involved than a single character.
@RyanM You can execute the !!/allspam command again in 30 seconds. To avoid one user sending lots of reports in a few commands and slowing SmokeDetector down due to rate-limiting, you have to wait 30 seconds after you've reported multiple posts in one go.
(also, if people think me dumping previous spam posts is unhelpful, let me know, but I figure it's useful to get them in the DB for future repeats of similar stuff)
(and also to have examples of what SD doesn't currently catch)
@RyanM IMO it is particularly useful for things that contain domains that we don't currently have detections for or have borderline, but uncaught, trollish language or offensive language. So, thanks :D
At some point, I might try to make the process a little less manual - e.g., it'd be great if AF would show me which posts are reported from a search, rather than needing to click into every single one.
Also, major props to AdvancedFlagging for making this even remotely practical.
I'm currently working on a way to find old spam posts that weren't caught as they were far to old, but the SO dump is almost 100GB, so currently testing on one of the smaller sites
@SmokeDetector Profile being extremely spammy heavily suggests this isn't just NAA. I've custom flagged to explain that I don't believe it's just NAA considering the profile page.
@SmokeDetector fp I guess it is Seeking Recommendations, so I flagged as such. Still, a rather poor question and I'm a bit suspicious given the email so I'm blacklisting them
@SmokeDetector Simply saying "cope" out of context seems like NAA, but in response to someone having a problem, it can be seen as unfriendly. For example, "I've tried everything!" - response: "Cope" is a meme-y way to be dismissive. I am not sure if I would rude flag it (a mod seems to have handled it) but I would still go with tp personally.
@Spevacus Feedback changed to tp... I thought it about it more, and I agree. If I ask for help, and someone says "Cope.", then that isn't an appropriate answer for reasonable discourse, so I agree that R/A would be fair game. I did flag the other one as R/A though
Is anything going to change with SD when the staging ground is launched? For example, is there going to be an option within SD for minor editing instead of fp if the question triggers SD?
@Ethan A lot of conversation (appears) to have occurred between Charcoal admins and Stack Exchange CMs or devs regarding what access we (Charcoal) are going to have in regards to the staging ground posts. What I know for sure is that Charcoal will definitely be able to see and parse the posts using Smoke Detector, and I find it hard to believe we wouldn't be able to flag them.
However in regards to feedback, I don't think we'd really need to adjust how we currently handle things.
If it's spam or utter nonsense, that's still a true positive. If it's not, it's still a false positive.
NAA's not relevant since the staging ground is all about asking a question (unless I've missed something)
I suspect that the Charcoal admins haven't disclosed much since all of that conversation occurred in private rooms and taking that information out of there would be... Well, a bit of a breach of trust, for one, and possibly a violation of their moderator agreement for two (in respect to the admins who are also moderators, most of them are, but not all).
This will be covered more in the second workflow post. But to answer your question here: I have already talked with the Charcoal team to discuss the access that they need, and the plan is to make sure that they have this access (including relevant API routes) from day one of the MVP release. We will try to make sure that all regular users of these chat rooms have a way to get access to the SG as Reviewers. Flags on SG posts will follow the normal flag flow to get mod attention (and will show up in the relevant mod dashboard). More on the next post. — Yaakov Ellis ♦Mar 28 at 11:52
That's about what we know thus far. We'll be able to access the SG as Reviewers.
@forest I don't know how much you can share, forest, but since you've somewhat recently become a mod... How much access do you feel you have to upcoming changes to the site wrt. feedback moderators can provide on the Moderator Team? If you can only answer in low tones/hushed voices, could you give me a ballpark answer like "I wish I was able to provide more feedback" or "I'm satisfied with what is disclosed at the moment"
@Spevacus The Moderator Team is pretty low-volume. It's mostly mundane things like "how should I deal with this particular event", and occasionally new moderator features being published.
But I don't feel like moderators are given particularly early access to new features such that they can reasonably change the direction things are going. It's not uncommon to see sentiments like "this is being done this way yet again despite us collectively speaking out about it". Obviously the specifics are not something I can share.
Actually, @Slate could probably say how much detail I'm allowed to talk about, since she's a CM. Otherwise the answer boils down to "I can't give details but it's not like we have special early access to everything exciting".
@Slate @Spevacus curious about how much feedback mods can give to SG in the Moderator Team, and naturally I don't want to go into detail lest I run afoul of "don't disclose what happens in mod-only spaces" policies.
More broadly I was curious how much feedback moderators can provide to up-and-coming features that do not directly relate to their mod tools (since the upcoming changes to their tools makes sense to post about on the Mod Team or in the TL) and more specifically, some of the others chatting above were lightly concerned about what access Charcoal is going to have to the SG.
It says on this Meta post that you need at least 500 rep to be a reviewer... can exceptions be made? For example, I don't even have 200 yet, but I have over 700 helpful flags (on SO) and am on SO flagging questions every day
