Please make sure to visit this Steemit blog’s front page where a link to the current policy on how I handle copycats will be provided.
Dear fellow Steemians,
It has now been three months since I started my own private campaign to search for, educate, and, if necessary, flag and report Steemit abusers, most importantly, plagiarists and other producers of copy/pasted content (who I here refer to as “copycats”).
Even though my mission has not changed since I have started it, and the means to achieve my goals are roughly still the same, I have over the last weeks and months adapted my strategy slightly to be more effective and productive.
In this post, I want to quickly update you on how my mission works, for the sake of transparency.
This is a non-profit project
I believe I have not even mentioned this before – it just seemed too obvious to me. Still, let me say this loud and clear here:
I have always intended this to be a non-profit project. I have no plans to ever “cash out” the profits I make with my posts or comments here. Instead, I’m planning to keep them locked into SteemPower to amplify my voice.
I believe that by keeping people from plundering the Steemit rewards pool, everyone on Steemit will profit, including myself, as I am a Steemit content creator as well. This way, I will certainly indirectly profit from this project, even if the profit is of course less than marginal, and only part of a far bigger picture.
A short recap
Let me state here that I deem my project a great success already. Here come a few quick stats!
(Note: when I say “a user has stopped posting” I mean here that for at least the last month, there was no new post.)
- Over the last 3 months, I have checked and added to my watchlist 38 users.
- 6 of them have not even made it on my blacklist since they have stopped abusive behavior after receiving my comments, but before receiving my first flags. Sadly, 5 of those have stopped using Steemit altogether.
- 28 I have put on my blacklist.
- 17 of the blacklisted users have meanwhile stopped producing abusive content. 11 of those have stopped using Steemit altogether.
Mission focus - UPDATE
I’ve always seen it as my main goal to not only address individual Steemit abusers, but whole networks of copycats who upvote each other’s abusive content mutually. I want to emphasize this direction even more in the future.
The past three months have shown that it really is not enough to flag a single user’s posts, even if done persistently. As long as the support of other abusive users is unbroken, a user can always come back again.
Also, I have started to widen my focus on other cases of abuse aside from plagiarism / copy/paste. A few users I have flagged in the past have later fled into using other abuse strategies, including comment spam and bid bot abuse. While my main focus will remain plagiarism and copy/paste, I will now take a close look on all actions of a user once he’s on my blacklist.
How my mission works - UPDATE
Again, the basics of how I am doing my work have not changed since my initial post on the topic. Here, I will outline how my processes work at the moment. (Items marked with (New) have changed since my original post, or I have not explicitly mentioned them earlier.)
How I find copycats
- I still have a huge, ever growing backlog of potential copycats which I am slowly working my way through.
- (New) I’m now more dynamically looking for users who participate in the same upvote network as known copycats. The votes graph by steemreports.com is of great help here.
- (New) As another way of discovering related accounts, I also examine known copycat’s wallets and transfers to discover accounts which are closely financially tied to them. These are typically alt accounts of the same physical user.
- (New) Based on my workload, I may add ~1-3 users per week to my (internal) “watchlist”.
How I identify copycats
In order to decide which user to put on my (internal) “watchlist” next, I take a few factors into account to make sure that I use my time most efficiently by only watching users who have a sufficient negative impact on Steemit economy.
- I’m typically dealing with users in a reputation range - / + 10 of my own reputation range, until the next higher decimal step. For instance, on my current level of 41, I would address users up to reputation 49.
- A user needs to be currently active and receiving a relatively significant payout (~$1+ on at least some individual posts).
- A user needs not to be actively engaged by other users or anti-abuse bots (except cheetah).
- (New) I may anonymously report accounts with higher reputation if I consider them to be mass-plagiarists and/or a vital part of a network of abusive accounts.
How I handle non-copycats
- (New) In my original post, I have suggested that I may “award” genuine original content creators by commenting once on one of their post – also, of course, to increase exposure for my mission. I have never done that, in fact, and I don’t see it necessary anymore now.
How I handle copycats – Phase I: Warning
In this phase, a user is on my (internal) “watchlist”. This list is not public.
- In this first phase, I’m only looking for plagiarism / copy/paste offenses.
- I start by commenting on one of the user’s most recent abusive posts (within 7 day payout period) with proof that the work is copy/pasted. If appropriate, I report the post in question.
- (New) If payout of multiple recent posts is comparatively high (~$3+) and the copy/paste / plagiarism intention very clear, I may comment on and report a second post right away.
How I handle copycats – Phase 2: Exposure
- Users who fail to change their behavior after two warnings and continue to post at least one more abusive post I will put on my public blacklist.
- Additionally, I will write a blog post about them to increase exposure of their abusive behavior for others to see.
- (New) This blog post will also include information on suspected related accounts (that either participate in the same upvote network or are financially tied).
How I handle copycats – Phase 3: Flagging
- Once a user is on my blacklist, I will comment and flag and, if appropriate, report all of his future abusive posts.
- (New) Posts with very small payout (< ~$1) I may ignore.
- (New) Once on my blacklist, abusive posts I am looking for will also include other offenses, including bid bot abuse and comment or post spam. I will check with other anti-abuse projects to make sure I really only comment on posts that are commonly seen as “abuse”.
- (New) I’m currently experimenting with tools that help me stay up to date with new posts of known copycats, for instance, steemstats.com.
How I handle copycats – Phase 4
- (New) In my original post, I have suggested that there may be a forth, more forceful phase of dealing with mass-copycats (10+ abusive posts). Judging by the successful application of my current workflow, I deem these plans no longer necessary.
How I handle copycats – Updates
- I will update my blacklist regularly with my newest findings.
- (New) About once a week, I will publish a summary of my recent findings.
How I handle copycats – Discussions & dispute
- As always, I stay very open for discussions with other users. This project, at its core, is about education. I would much rather like to see a user stopping plagiarism and creating interesting genuine self-made content than having to flag someone down until his voice is stilled.
- Revenge-flagging I will forward to anti-abuse organizations.
- (New) There is zero tolerance for insults and threats, against my account or others. “Watchlisted” users who commit these actions will immediately be publicly exposed and put on my blacklist.
Lastly, I want to say a big thank you to everyone who kindly supports my mission by upvoting my posts, and especially to all those of you who leave such kind and encouraging comments.
Thank you for everyone who gives feedback and first and foremost thank you to the guys at steemflagrewards for all their support and everything else they do, for the Steemit platform in general!
Together, let’s continue to make Steemit fairer for all genuine original content creators.