Nov 20, 2012 6:18am

Curious about some of the flagmod / flagging rules. Let's say I came across a profile like one of these:

- now, all of the above are fake / scam accounts, for fairly obvious reasons. Either photos in them link to porn sites via GIS, or in one case it's someone pretending to be supermodel Chanel Iman, or in another it's obvious the photos were stolen from another profile (they have the "double watermark") - anyway, all were flagged as fake profiles. Yet they're all still standing, for some reason. No one in their right mind actually thinks Chanel Iman is trolling for dates on OKC, or that the last one - which gets 110,000(!) GIS hits is in any way legit.

Yet... there are always those flagmods who, I guess, don't read the line "Is obviously a spammer (based on photos or essays)" - they jump in with "flag the pic" as if simply removing the pic of Chanel Iman, or the pic stolen from a porn site, will solve the problem. It's still a fake profile, someone pretending to be someone they're not, and they're simply going to upload new pics, only to get flagged again, ad infinitum, until the faker gets bored and gives up (if ever) or finally finds a set of stolen pics that don't register in GIS for some reason.

Now, the question is - if one was to flag the profile a second time, after it had passed through he flagmod queue and somehow it still stood, would that second flag register, and the profile re-enter the queue? It seems you can only flag something once. I can understand the need for this, as people could revenge flag someone into oblivion that they just personally didn't like. But it also seems like maybe there could be a solution to permanently delete these obvious fakes/scams. Maybe if, at the end of the queue, they're all passed on to someone who actually works for OKC who can look at such obvious things and right past all the "flag the pic!" clutter.

Some of us are just trying help out the community by cleaning up the site a bit when we've got some free time, but it's a bit like fighting a brick wall now and then. If you don't hit the queue right at the moment a group of sensible flagmods happen to be modding, things have a way of just passing through and lingering on the site forever. It makes it such that one doesn't really want to bother helping out anymore - we do it out of a sense of community, but if it's an exercise in futility, there doesn't seem to be any point.

Nov 20, 2012 10:17am

As you started the thread here, I'll answer here. I'm a bit n00b to flagmod, but from what I understand, when you have too many picture deletes, OKC will delete your account. So that kind of helps to make sure things don't spiral into oblivion, even if profile is kept & pic is taken. Not perfect, but then again — we only see a very onesided ... side of the whole system and I trust there's solid machines of metal somewhere, grinding hard statistics on flags, votes and the like, ever cautios of finding the balance between weeding out spam & getting too many false positives.


Nov 21, 2012 12:51am

Bear in mind that the company that owns OKCupid is literally worth billions. They just bought for $300 million a few months ago.

If they genuinely cared about fake profiles, they would hire a staff of professional moderators, rather than getting gullible people to whitewash the fence for them.


Nov 21, 2012 7:23am

If the company also owns then they already have a solution to this problem, If you don't want to see posts that violate a rule, join


Nov 21, 2012 11:59am

tyrebiter - do not name the one who cannot be named! But the point is valid — with freedom come things with pointy teeth and ungodly appetites, that is the basic logic of all computer security. OKC has struck a good balance of noise/signal ratio and flagmods with their plethora of opinions are one piece of the infrastructure upholding this balance.


Nov 21, 2012 5:12pm


If they genuinely cared about fake profiles...


Scam profiles are killed-off here by the thousands, sometimes thousands in a day.

Whether the amount that slip through the control measures are too much for the taste of Eeyore up there or anyone else, has nothing to do with "they don't care". Sites like this are continuously bombarded with miscreants trying to exploit the site for scamming and spamming.

The reason many of us are HERE rather than the hundreds of dating website competitors (or the dozens of other dating websites Interactive Corporation owns other than OkCupid), is because they have not clamped-down like the Gestapo requiring every single f*cking change to your profile to be vetted by some stupid nanny, and every single thing you write or post anywhere on the site to be moderated and sanitized by the thought-police before seeing the light of day, and so on and so forth. The mere fact you all can sit here on one of OkCupid's open/public forums and kvetch about all the stuff you don't like about the site is a perfect example of that freedom. Because many of the competitors ban that sort of thing, might even throw you off their site for it.

For those who prefer the nannying sort of thing, there are plenty of alternatives, including various other websites owned by the same umbrella corporation that now owns OkCupid. (but which does not run day-to-day operations here)




May 3, 2013 4:08pm

Ok, someone help me out.

I've just become a "mod" recently, to where I can vote on pics that are flagged.  I've read the guidelines... they don't mention anything about GIS.  So what is "GIS"?

I did a google search and found the following:

"A geographic information system (GIS) integrates hardware, software, and data for capturing, managing, analyzing, and displaying all forms of geographically referenced information."

Well, I'm not a computer programmer, and have no inspiration to become one.  So I don't know anything about this.  I suppose there's some type of software you can run a picture through and "analyze it"... and if the pic "fails", then it's considered a GIS and should be deleted?  A simplified explanation in the guidelines would be rather useful.

But alright, besides this GIS deal... how do you *PROVE* that a pic is not really the user?  Do we need to click on their profiles and read through, and try to determine if the person is real or not?  That's just a tad beyond what is indicated in the guidelines - that we're just voting on the pic itself.

I would like someone to get back to me on what to do when everyone has voted "gis", and I cant find anything wrong with the pic.


May 3, 2013 4:19pm

I see you already found the place I was going to direct you to. Never mind.

Post a comment