You are viewing a single comment's thread from:

RE: You cannot take away from the author that which does not belong to the author.

in Proof of Brain3 years ago (edited)

It's not my job to come up with solutions here. That's not my project. I don't feel like building or designing it either.

How about this. I asked logiczombie if they could find a flaw. NO answer. Perfect system.

Dwinblood. Can you find one flaw in this system and present that to me?

Sort:  

You pointed out a flaw. If people flag it then it is going to reach out to however many Jurors the system is set up to reach out to. Depending upon how that notification is received that can be a nuisance. IF they do not respond within a certain time the flag passes.

However, if bots or something are designed to down vote a lot (such as hivewatcher, steemcleaner type stuff) then that could trigger a lot of these things in mass which would suck and could cause problems.

As to solutions to that. Certain accounts could be flagged to be excluded from the process by something similar to how witnesses are elected. You'd want it to be an election process so that it remained something that is decentralized and such accounts could be removed from exclusion if they lose the support of the hive community.

Such accounts could be used without triggering the process.

Yet that wouldn't stop people from writing other bots to do this. You could monitor how long it has been since an account submitted a flag/down vote and have a cool down period.

There are numerous ways things can be approached.

Other potential flaws. If it is randomly selecting jurors sometimes those jurors may be bots or alternate accounts of someone.

That is unavoidable I'd think but I also think the odds of it happening and causing problems are low. It could potentially happen but if it was responded to by a bot unless they took steps to make it seem human (randomize response time) then you could potentially catch and discard some of those.

A lot of these types of problems have been solved in networking. The same principles can be carried over here.

Perfect system? No.

Better system? Definitely possible. Testing would potentially reveal other issues.

There are a lot of flaws. I can think of several more scenarios where it might run into problems. My goal is not to be a pessimistic dickhead, nor will I suggest you're shooting it down for pointing out flaws you see. Best to place all the flaws on the table first before finding solutions or you just end up running back and forth between the fridge and kitchen.

I think people would opt out. To get a good feel for it, take a poll first. However, in order to obtain accurate results, every single possible pro and con imaginable would have to be laid out. Can't just offer them some lopsided viewpoint and point at Mr. Victim over there. Besides, to even get that idea moving you'll need a metric shit-ton worth of support from this community. I can't promote it with all these flaws I see. No chance.

It turns a flag into free advertising. It creates a scenario where someone who's not qualified to make a call, has to make a call. How good are you at spotting art fraud? What if it's a false flag on art fraud. Will you know? So if the majority is wrong, then what? Then what? The system in place now is streamlined and removing these fraudsters is simple. Why add so many layers of complexity?

logiczombie offered a 'solution' with some kind of "form" stating each party writes some kind of description before it's sent out to 1000 people. Overlooked the part about how now the one getting flagged will simply lie (they're always in denial even with evidence provided), or they simply don't write anything and the operation hangs. So that's not going work. Can't just throw shit at the wall and see what sticks.

I don't believe for one second anyone involved here in support of that idea will even attempt building it. So that's a flaw as well. It's all talk. Prove me wrong.

And if you're willing to risk distributing illegal obscene images, or deny the existence of that risk, or do not understand the consequences of that risk, or just feel like saying it can't happen because happy thoughts; maybe just put the idea to bed. That is by far the biggest flaw. Even if that proposed system placed instances of that foul shit in front of potentially 10 accounts, that is ten too many. Just think about what kind of stuff gets flagged. Rainbows and unicorns?

Loading...