> Pattern-matching website in body, pattern-matching website in title ---------- Title - Position 1-21: healthyminimarket.co Body - Position 1-21: healthyminimarket.co
> Pattern-matching website in body, pattern-matching website in title ---------- Title - Position 1-21: healthyminimarket.co Body - Position 1-21: healthyminimarket.co
> Pattern-matching website in answer, pattern-matching website in title ---------- Title - Position 1-21: healthyminimarket.co Body - Position 1-21: healthyminimarket.co
> Pattern-matching website in answer, pattern-matching website in title ---------- Title - Position 1-21: healthyminimarket.co Body - Position 1-21: healthyminimarket.co
> Pattern-matching website in answer, pattern-matching website in title ---------- Title - Position 5-25: healthyminimarket.co Body - Position 5-25: healthyminimarket.co
> Pattern-matching website in answer, pattern-matching website in title ---------- Title - Position 5-25: healthyminimarket.co Body - Position 5-25: healthyminimarket.co
For example, using one of the links from above, you'd run an API call to get user 363661 on server fault and pull the account id. From there you'd make a call to associated-users.
Two API calls for a single user, because the shallow_user object doesn't contain account id
The only bad thing: Bad in the olden days spammers usualy kept making new accounts before they were going to spam, so the associated-users list didn't include new pending sites
Spammer makes account 1, spam site 1, spammer makes account 2, spam site 2, spammer makes account 3, spam site 3. If we did that we probably only catched 1 or 0 posts more
@Andy more/less that the spammer already spammed on that site/ still need to create more accounts for new sites, we probably only catch sites were he already posted on
@TIPS I think this is reasonable: when the URL parameter of addblu is network-wide profile, use the API to retrieve accounts and blacklist them. Don't have time right now, though...
also, @Åna I left this out of that email, but do keep trying for a pro-tem appointment. You remind me a lot of... myself, about three years ago - extremely eager.
For some reason I want to say this'll be the third or fourth try.
@Åna Indeed. In the meantime, find yourself a niche and figure out how to flag stuff in it. I really wouldn't mind having someone who is really good at finding bad reviewers on SO, for example.
@Undo don't you need 10k rep to be effective at finding bad reviewers? Even if you bump into a bad review, you'd have to look at other reviews of the same user
unless there's a lucky kind of disaster when a user floods so much crap into the review queues that you can follow that trail and see what's approved:)
If the number of changed characters is available, the lower the better for wild guesses, right? I'd expect a smaller ratio of minor edits to be approvable than the global ratio
This shouldn't be difficult, really... The hard part is collecting the data, when you've got it you can just throw it at a database and run tests on it
@Undo it demonstrates that there's a significant number of users who do not know how to use the system and accordingly this would be translated across other areas of the UI
@Yvette I don't see the argument for it necessarily transferring across other areas. Let's find those other areas and fix them instead of sinking dev time into this.
@Yvette Oh, of course we'd try to fix it in a perfect world with unlimited dev resources. At the moment, though, I don't have any evidence that says this should be prioritized over anything else.
And if there's no evidence for that, why should it be done at all?
Basically, I just need to see a better reason than "people are doing it wrong". Show me a number of spam posts that lived for a significantly longer period than they should have because of this, or another tangible impact.
it's not a big deal, nothing really is, any mistakes people make, the thing that amazes me is the number of higher rep users (people with close vote priv) who don't know how to use the system. That's the amazing thing. To be on the site for so long and not know how to use the system. There is a failure in the system with than I think.
The flag menus should (and probably will) be condensed in the future, but until we have a reason represents a majority or a breaking issue, it's best if we leave this out of the dev pile
@Yvette In my experience, it is a very underused feature across platforms. SE is successful at spam due to the size of the community and because of teams like us.
@Yvette I think it's more like this: people who think of flagging as something that only happens to other people, won't flag if something flaggable stares them in the face
