You are viewing a single comment's thread from:

RE: Steembay (a bot) under attack by a wannabe AI

in #flags6 years ago (edited)

I am sure that you would. I'll try to sum it up.

I am opposed to most organized rule setting processes. Morals can't be "cought" pressing them into words or rules. The world simply isn't black or white, but all over grey.
The second you draw ANY line, people will claim the most outragous acts to be "legal". People can't anymore see the difference between the two (very different) things legal and moral. Wiggeling around or chosing the very line as a standard is then seen as "normal". The same line of thought leads to the inability of differenciate between doing something "lawful" and doing something "right".

In my opinion (and I know, not many share this opinion) things should BE right and consistent, not only "feel" right.

The problems I see in countering selfvoting, vote selling and circle jerks as much as I hate them are the following:

  1. From a game theory perspective the fastest growth + rise of steem value can be obtained by ONLY voting for others, no bidbots, no gilds, no flags. As there is all of that, you can try fighting it, making it worse (with a chance of about 99% is my estimation) or you can show/offer other incentives. Turning around your system could be much more effective ( I have not much thought about that, but monitoring selfvoting like your system does it is a very good starting point. If you inform people about the harm selfvoting (circle jerking and vote buying) does and incentivize “stopping it” you raise awareness, you will need much less funds and there is no value nullified (there is a very small bot voting for posts that did NOT use bidbots… I love this one)

  2. From a mathematical point of view, countering selfvotes with flags is as much slowing down the overall growth as selfvoting itself. Therefore it doubles the harm, at least short term, probably also midterm. In the longterm the potential loss is exponential. If at any time there are NO 100% selfvoters, and as your system grows and becomes more "intelligent", you will have to go for the 99%, the 98%, the 50% until where? You will have to go for the hidden selfvoters such as this one https://steemit.com/@kurosawa/comments
    (not trying to point fingers, just an example) then for the big systems such as the diverse bidbot armys and finally such as haejin. If I calculate the overall "nullified" rewards, I get dizzy.

  3. From a philosophical point of view, I will try to make an example:
    If you ask 1000 people if they believe that selfvoting is a bad thing, you will get a large majority agreeing. If you ask the same people if it should be punished / countered almost the same amount will agree as well.
    Now you ask 1000 different people if it is good that there is free service that normally costs a fee, you will get a similarly large majority agreeing. If you asked the same people again if it is OK that they selfvote, again almost the same amount of people will agree.
    Something I really don’t know is what would happen if you asked both sets of questions to the same people… Probably it will depend upon the order of questioning.
    The point about this is, that there is no real line to work with, only a vague feeling about what is right and wrong strongly depending on the circumstances, while most of the time the real issue is ignored or worse simply unknown (or sacrificed against better knowledge in favor of short term gains.)
    Again all scientific research points to incentivizing for a certain behavior instead of punishing for the wrong behavior (flags unfortunately are largely perceived and used as punishment)

Finally you can ask how can I be so much against selfvoting while running a service that is 100% selfvoting. A very valid question…
This is “ME”
ME.png
I vote about once a day for myself and exactly twice a day for my service “steembay”. I think this is quite reasonable. (and "optimal selfvoting" would look a lot different)

In general my preferred “weapons” in fighting against stuff I don’t like is reasoned ostracism or incentivizing the preferred behavior. For what I know from history, read in countless books and experienced personally: every other way is doomed to fail from the beginning at least in the long run.

The "real" problems about the Steem blockchain in my very personal and thus subjective opinion are the actual distribution and linear rewards (and ofc. unreasonable cashouts but this is a totally different story). The reason for the failing of n2 was because of the distribution and bidbots/lucrative selfvoting are directly caused by linear rewards. Actually I'd rather invite a million businesses all selfvoting while storing 1000$ in the system than risking scaring one of them away.

Sort:  
Loading...

We're not asking you to contribute to @sadkitten.

  1. How is that free bot sustainable in anyway? You're basically financing the bloat of the chain by funding spammers. RIDICULOUS

  2. No we're not going to 50% ANYTIME SOON, - RIDICULOUS. People can opt-out their delegation whenever they think it's getting counter productive.

Now you ask 1000 different people if it is good that there is free service that normally costs a fee, you will get a similarly large majority agreeing.

I hope some people will let you know this service is RIDICULOUS and cannot last.

Well... I was sure most will not even come close understanding what I am aiming at.

We (steembay) are sustainable because of donations and the upvotes people give us on each auction and our daily reports. Our service is so "ridiculous" that it relocated more SBD in the last 7 months than all minnows together. Most selfvotes serve to organize the overview of the auction process. If I wanted to use Steem as a piggy bank the system would look MUCH different and would be not effected by sadkitten anytime soon.

Cherrypicking from my dystopian forebodings (50%) doesn't change much about possible outcomes of an arbitrary (or by an algorithm) set line. I simply don't like that for reasons I tried to explain. It is a well researched fact that people tend to approach to those lines as much as it is possible. Sadkitten has the potential to become such a "rulesetter". Let's pick any number. 96% ...
If enough people are involved and invested in your project this number will become the number to tend to. (you can see this development e.g. with minimal hourly wages, taxes, even abortion just to name a few). People will see that line and "feel" that it is socially accepted to selfvote at 96%. Seen from this perspective your service can have the contrary effect than expected.