More than 1000 merit abusers found here (clearly motivated) - red trust time

[claudiotiego]

how can i find out my profile number?

[1715400121]

1715400121

[1715400121]

1715400121

[1715400121]

1715400121

[mdayonliner]

The script is showing those who have exchanged minimum one merit with each others. While the whole list might not be even 50% accurate (a wild guess) but it's obvious that scanning the list will make the life easier if you want to dig down the merit abusers rabbit hole.

I wonder with the earlier poster...
Why did you created this ALT? Why not you use your real account. Scared about getting RedTrust. Man... Look at my trust page LOL

PS: This is really a good shit but since it's from an alt I stopped myself sending an sMerit 

[coinlocket$]

Op, I want to merit you for this post but I think this can be an alt so let me know (I don't waste my precious merit if they are not useful)

I started off thinking like that, but now I don't think that it matters. I feel that it is the post that is getting the merit, and not the poster, so we should not be influenced by rank or other factors. The exception is where a post has been made to gain merit to pass on to another account. Hopefully these will be highlighted in the current research, and we will be able to ignore these posters.

Yes you are probably right but I know the feeling to be one not English native and try to post here, so I prefer to use merit to help people (when they deserve). For this reason usually, at least for now, because merit is harder and harder to get for the "decay"(-40% from March to April), usually I don't merit legendary even if sometime they deserve a lot of merit cause they don't need it.

[squatter]

How have you worked that out? I've given out hundreds of merits, and I've received hundreds. I'm sure that I have given merits to members who have also given merits to me. I haven't checked, and don't remember as the awards ween't linked.

But what are the chances that those users have given you merit back within one hour? The chances of two random users giving each other merit is already a big coincidence, but then doing it within the hour is very unlikely and very suspicious.

If someone gives 100% of their sMerit 100% of the time to another user, obviously it's suspicious.

And if we assume that any two users are random, then your assertion would be correct. But social media functions are not random. One example is the Wall Observer thread. A lot of the regulars in there seem to merit each other -- and re: the time question, the conversation actually moves really quickly. Merit distribution is probably nowhere near random even if we ignore the alt accounts, farming, etc. This isn't 4chan; users aren't anonymous. There are clearly established social groups on the forum.

I guess I would just stress caution when the pattern isn't extremely clear. People around here really love their witch hunts, though.

[Expert3]

It's a shame to found out those. Though this is not a clear indication that they we're the same person who is abusing the merit system for themselves. What we need to do is to post better and don't mind these fraudulent activities because obviously this is just a waste of time.

[digaran]

What we need to do is to post better and don't mind these fraudulent activities because obviously this is just a waste of time.

What we need to do is to let people to rank up normally after receiving at least 100 merits and never let them to send any merits to others. when you have the ability to send merits and rank up other members, you'd go with your nature to rank up your alts.

Monitoring 80 or 100 people is easier than monitoring thousands. now if you already have received 100 merits, you should rank up with no merit requirement and you shouldn't have the ability to send or receive any merits.

EDIT: disclaimer: the tongue of my language is not my English mother.

[LoyceMobile]

   - go through the merit file and simply checks for any direct exchange of merits between 2 users within 1 hour

I can't really view all on my phone, bit since I'm on your list, I can say with 100% certainty your method is at least inaccurate.
What you should do, is manually check all your positives and see if the merited posts don't deserve it.

[jmigdlc99]

How have you worked that out? I've given out hundreds of merits, and I've received hundreds. I'm sure that I have given merits to members who have also given merits to me. I haven't checked, and don't remember as the awards ween't linked.

But what are the chances that those users have given you merit back within one hour? The chances of two random users giving each other merit is already a big coincidence, but then doing it within the hour is very unlikely and very suspicious. I haven't looked through the results in the OP but if it works as stated it would be helpful to go through and manfully look for potential abuses for those that can be bothered or have the time. People will get wise to this though and just wait a day or two to return merit.

The chances may actually higher than you think. I believe time between merit isn't such a good indicator of abuse because legitimate people/accounts tend to give and return merits as a means of saying "thank you."

I understand it's not designed to work this way but some people will use the following logic:
gee thanks for the merit buddy, here let me return the favor and merit one of your posts as well.

[deisik]

I have coded a script to find abuses in the current merit system automatically, this is briefly what is does:
   - go through the merit file and simply checks for any direct exchange of merits between 2 users within 1 hour
   - count how many of the merit given in total have been used for possible exchanges and produce a percentage
   - write down who received the merits, how many and how many they sent back as part of the possible exchange.

This is bullshit and witch-hunting at its best. There are a lot of excellent posts which go unmerited (up to a point where you can safely say that the merit system doesn't work), so if someone gives you a couple of merits, you feel like you should give them a couple of merits in return. And if both of you are good posters, I see nothing wrong or bad in giving a merit (or two) as an expression of your thanks. Is this a red-trust time for you, OP?

Obviously, cases with a lot of merits exchanged is a completely different matter

How have you worked that out? I've given out hundreds of merits, and I've received hundreds. I'm sure that I have given merits to members who have also given merits to me. I haven't checked, and don't remember as the awards ween't linked.

But what are the chances that those users have given you merit back within one hour? The chances of two random users giving each other merit is already a big coincidence, but then doing it within the hour is very unlikely and very suspicious. I haven't looked through the results in the OP but if it works as stated it would be helpful to go through and manfully look for potential abuses for those that can be bothered or have the time. People will get wise to this though and just wait a day or two to return merit.

The odds are really low and so what? Is there a rule which forbids you awarding a merit to someone who merited you as long as the posts of both of you are good? Okay, now if someone gives you a merit, you should think twice and wait at least a week before meriting back so as not to raise the suspicion of the forum's "merit police"

[coinlocket$]

~SNIP~
I see nothing wrong or bad in giving a merit (or two) as an expression of your thanks. Is this a red-trust time for you, OP?

I understand you point of view but Merit are not designated to Thanks people but for improving the quality of the forum. It's not the Like from facebook.

While we will not be directly moderating this, I encourage people to give merit to posts that are objectively high-quality, not just posts that you agree with.

[DdmrDdmr]

As shahzadafzal reminds us in a post somewhere above, the official position is that any sMerit malpractice will (eventually) decay and become proportionally insignificant in the bigger picture.  Being that so, chasing Merit malpractice, aside from being extremely difficult to prove, doesn’t really get us too far regardless of whether we like it or not and the sheer ethics behind it, where effort should be the core driver, and not cheating the system.

Distinguishing malpractice from friendship, liking or affinity is likely to be a non-feasible task, and since sMerit is a free currency to use once earned (with ethical and tactical goals, yes, I know), who’s got the overall right to judge and act upon?

Having said that, I figure that the OPs intention is not to assert that the 1k list are merit abusers, but rather a narrowed down list of potential abusers by the 1 hour timeframe sMerit swapsies criteria he used. That is, if someone were to look for merit abusers, that is a better staring point that having nothing. Unfortunately, there is no strong focus put on this, so persuing this will probably render a face-on crash against a wall.
In full disclosure, I too created a post sometime back with narrowed down lists by other criteria. I believed, at the time, that merit abusing was being pursued adamantly, but soon came round to letting it be.

What is against Forum rules, as I understand it, is having multiple Alt Account participating in the same campaign simultaneously. That is being pursued and, being it a difficult task that needs to be treated manually on a one per one basis, could benefit both from reporting on threads (which does take place regularly) and/or to moderators, as well as from lists that help narrow down suspect cases.
Now sMerit transactions could lead to helping the latter, since farmers are likely to try to uprank their Alt accounts for a larger gain. With the Merit System, these Alt accounts will require to receive sMerit at some point in order to uprank, thus perhaps some lists base on sMerit TXs could help pursue this goal (account farming).
Even so, it will not be automatic and will require someone with experience, criteria and power to do a one by one check and act upon what he sees.

To prove or deny my point, I went through the sMerit.txt file crossed with user profiles and though that a potential (underline potential, not certain in any way) set of cases to look into would be:
-   Jr. Members that have receive high amounts of Merit from one single user (50 or so to start off with).
-   Likewise with Members.
(there are plenty more potential criterias we could come up with, but these are easy top of mind).

Now the sMerit can be received as a one shot deal, over multiple transactions, etc. The suspect could receive himself the sMerit from one source alone or multiple sources that allow to understand the sMerit he receives. These factors all help to narrow down the study of the case.

I gave it a go with a quick drawn list of 10 Jr. Members and 5 Members:


From User   From Rank   ToUser   ToRank   Merit   numTx   Comment
152126   Legendary   366535   Jr. Member   50   1   50 sMerits for a single link to an Article
168114   Legendary   289275   Jr. Member   50   1   50 sMerits for a deleted post
192898   Legendary   1130764   Jr. Member   50   1   50 sMerits for reserving an Ann thread
211531   Legendary   1593656   Jr. Member   50   1   50 sMerits for an article summary; no more sMerits received from others
211641   Hero Member   232828   Jr. Member   50   1   50 sMerits for a 2016 2 liner comment
244104   Legendary   319826   Jr. Member   50   1   Initial Monero 2014 Ann thread; account not active for over a year
307978   Legendary   1227588   Jr. Member   50   1   50 sMerits for a quick one liner (no other receives sMerits)
319535   Hero Member   1308930   Jr. Member   50   1   Phore workbuddies
339752   Legendary       769913   Jr. Member   50   11   ann ICO bounty OP
412599   Hero Member   1163497   Jr. Member   50   1   50 sMerits for a pre Merit kick-off short post

From User   From Rank   ToUser   ToRank   Merit   numTx   Comment
377987   Legendary   1580039   Member   73   64   Deep connection here
105730   Hero Member   1348903   Member   72   2   Deep connection here
85601     Hero Member   1393407   Member   60   2   Can´t understand the language, but merited threads are oldish
668854   Hero Member   1162829   Member   55   3   Not very clear to me
672155   Hero Member   1201858   Member   50   1   encouraging for newbies: 50 sMerits for 12 words and a wink (on backdated post)


Do I consider I have proof of Account Farming through the above ? No, not prove, but suspicions.
In all the cases ? No, just a couple, but those cases are visible thanks to analyzing the sMerit transacted, and then performing a case by case study.
Do I want to become a Forum Cop? No, not really.

The point is that we can use sMerit to help detect Farming Accounts as an additional way of approaching the matter.

[LoyceV]

~SNIP~
I see nothing wrong or bad in giving a merit (or two) as an expression of your thanks. Is this a red-trust time for you, OP?

I understand you point of wiew but Merit are not designated to Thanks people but for improving the quality of the forum. It's not the Like from facebook.

Those two can be combined: after receiving merit, I sometimes end up reading someone's post history. If I find posts that deserve merit, I merit them. Not because someone merited my post, but because I ended up reading them. You could argue this is because of the first merit transaction, but I don't think that matters as long as the posts are worth it.

Having said that, I figure that the OPs intention is not to assert that the 1k list are merit abusers, but rather a narrowed down list of potential abusers by the 1 hour timeframe sMerit swapsies criteria he used.

The problem with this approach is that chances of a positive increase as someone has sent and received more merit. Eventually, almost all quality posters who use merit as intended will show on this list.

[deisik]

~SNIP~
I see nothing wrong or bad in giving a merit (or two) as an expression of your thanks. Is this a red-trust time for you, OP?

I understand you point of wiew but Merit are not designated to Thanks people but for improving the quality of the forum. It's not the Like from facebook.

They were designed specifically to thank people for writing quality posts

Unless their real purpose was in fact to find and weed out account farmers, of course. Anyway, you still have to prove that giving a merit to a useful post (which should be merited on its own but wasn't) as a kind of thank you is a violation of any forum rules or ethics. Mind that I'm not talking about thoughtlessly giving merits back just for being merited (kind of an-eye-for-an-eye approach in reverse). If you have difficulty interpreting or perceiving the meaning of my words, imagine that a merit given is an invitation to evaluate the posts of the giver. Since there are many threads across the forum offering just that (i.e. evaluating and meriting quality posts), you will have difficulty challenging this view

[TheQuin]

Those two can be combined: after receiving merit, I sometimes end up reading someone's post history. If I find posts that deserve merit, I merit them. Not because someone merited my post, but because I ended up reading them. You could argue this is because of the first merit transaction, but I don't think that matters as long as the posts are worth it.

I've done that a few times when it has been an account name I didn't recognise. I think the best use of the information is for anyone who has time and the inclination to do it to go through and see if there are other links to the accounts. Like are they in the same bounty campaigns etc.

On a side note while Vod's site rebuilds its database, from the information it has so far I seem to be an alt of TMAN and achow101. Who knew?

[DdmrDdmr]

Having said that, I figure that the OPs intention is not to assert that the 1k list are merit abusers, but rather a narrowed down list of potential abusers by the 1 hour timeframe sMerit swapsies criteria he used.

The problem with this approach is that chances of a positive increase as someone has sent and received more merit. Eventually, almost all quality posters who use merit as intended will show on this list.

True. The OPs idea is just one criteria, that is bound to get plenty of false positives. Unfortunately, OPs subject heading is not too well chosen, since it leads to assume that the 1k list are indeed merit abusers, when the real deal is potential abusers (he does clarify in the post this fact though).

I would in any case favour filtering by tighter criteria, creating shorter lists ordered descending by that criteria, being less false positive prone (for example the two lists I took extracts from in my previous post above). Favoured criteria would scout for:
-   Injective cases: High amounts of sMerit awarded to a single user from a single user (1 or n TXs).
-   Bijective cases: High amouts of sMerit awarded to a user and back again (1 or n TXs).

The former seems more logical from an Account Farming point of view.

All will surely throw false positives, and even if they didn’t, actually proving that account farming is behind is highly difficult just due to suspicion derived from sMerit TXs.

[LoyceV]

I think the best use of the information is for anyone who has time and the inclination to do it to go through and see if there are other links to the accounts. Like are they in the same bounty campaigns etc.

In that case, I must be an alt of about half the users in the ChipMixer campaign.

Unfortunately, OPs subject heading is not too well chosen, since it leads to assume that the 1k list are indeed merit abusers, when the real deal is potential abusers (he does clarify in the post this fact though).

The clickbait title is what got me to read this thread. I couldn't believe he identified more than 1000 out of just 14500 merit receivers as being abusers.

I would in any case favour filtering by tighter criteria

I'd say leave out everybody who sent/received merit to/from more than a certain number of unique users, to leave out the users who turned positive just because they have had so many merit transactions.

Favoured criteria would scout for:
-   Injective cases: High amounts of sMerit awarded to a single user from a single user (1 or n TXs).
-   Bijective cases: High amouts of sMerit awarded to a user and back again (1 or n TXs).

This will probably still tag me as an alt of nullius, so I'll have to disagree.

[DdmrDdmr]

<...>
I think the best use of the information is for anyone who has time and the inclination to do it to go through and see if there are other links to the accounts. Like are they in the same bounty campaigns etc.

Exactly, and what would then be the final proof is seeing if the ETH (generally speaking) addresses are linked through transfers from one to the other. If anybody has a function to query Etherscan ETH and Tokens someone could work on that at some point (there are APIs, but I'm wondering if there's any easy way to pull data into SQL Server to cross over there) . Although hinting it here is probably not my best idea...

[squatter]

I understand you point of view but Merit are not designated to Thanks people but for improving the quality of the forum. It's not the Like from facebook.

While we will not be directly moderating this, I encourage people to give merit to posts that are objectively high-quality, not just posts that you agree with.

But that's exactly how people use it -- like a Facebook "like" button. We should approach merit based on how the system is actually used, not the intent. I agree with Theymos' words above, but there are other social dynamics at play. Getting merit often seems more about saying what people want to hear -- and doing it in the right venue (e.g. Meta vs. Bitcoin Discussion) -- than anything else.

Unfortunately, OPs subject heading is not too well chosen, since it leads to assume that the 1k list are indeed merit abusers, when the real deal is potential abusers (he does clarify in the post this fact though).

Yeah, the OP ought to edit the title.

Favoured criteria would scout for:
-   Injective cases: High amounts of sMerit awarded to a single user from a single user (1 or n TXs).
-   Bijective cases: High amouts of sMerit awarded to a user and back again (1 or n TXs).

This will probably still tag me as an alt of nullius, so I'll have to disagree.

I guess the solution is: don't ever give or receive merit. But since you can't avoid receiving merit, allegations of merit abuse will always surface. C'est la vie.

[LoyceV]

Favoured criteria would scout for:
-   Injective cases: High amounts of sMerit awarded to a single user from a single user (1 or n TXs).
-   Bijective cases: High amouts of sMerit awarded to a user and back again (1 or n TXs).

This will probably still tag me as an alt of nullius, so I'll have to disagree.

I guess the solution is: don't ever give or receive merit. But since you can't avoid receiving merit, allegations of merit abuse will always surface. C'est la vie.

I don't worry about the allegations (if any), especially if they're coming from an automated algorithm that tags 1000+ accounts as abusers.
I guess the solution is: don't cheat. I know why I give merit, and if anyone is bored, feel free to review it.

[shahzadafzal]

bitcointalk.org account got hacked ??

No OP extracted the public merit data and analyzed based on some rules like merit exchanges between two users within specified time... which could be possible if they agreed to exchange merits... 

Technically to me it won't be merit abuse if they give merit to the deserving posts...