Buying a module carries much less responsibility than selling a module.
There are scam possibilities from both sides. I am dealing with lot's of requests to delete unjustified downvotes which is annoying because I don't want to interfere in private conversations so if someone comes up with a better, easy to implement idea I am all ears.
-- modulargrid
I've been thinking about a system to replace the current up/downvote options with and I think I have a pretty solid idea here. Besides being more reliable, I think this system will also reduce the amount of requests to remove negative feedback.
Whenever a user has bought or sold a module and decides to up/downvote to the other, the user on the receiving end is sent an automated request to return feedback to the user that initiated it. The catch is however, that only after both parties have submitted their feedback, the score is shown. Neither user can change their feedback after placing it.
This approach eliminates any worries about receiving retaliatory downvote simply because a user is dissatisfied with another user. At the same time, if a transaction didn't happen or if a user is being unreasonable, a user can decide to not leave feedback and thus not getting any in return either.
An optional one-time opportunity, in the form of a single non-editable comment, could be given to any user receiving the downvote as a way of explaining themselves. This comment would be shown as a comment on the users profile, for example when hovering over the username of someone who left the feedback. In any case, comments can not be replied to or edited after placing them. If a solution has been agreed upon by both users, there will be an opportunity to change the negative feedback into a neutral feedback and with that also removing the comment from the users profile. In no way however, will negative feedback be allowed to turn into a positive feedback unless submitted by mistake.
Since there are several aspects to this system, implementation can be as simple and straight forward, or as elaborate and complicated as desired. Simply withholding information about the type of feedback left by one user to another creates a far more reliable system than is currently in place and completely eliminates any fear of retaliatory feedback.
I know this system is not entirely fool-proof, but it would be very difficult to pull that off without enforcing transaction verification.
The current feedback score for users can be held in the form of a legacy score in order to smoothen the transition to a new system.