You are viewing a single comment's thread from:

RE: Haejin has gone for downvoting rampage

in #abuse5 years ago

If self votes weren't allowed, the abusers would just use different accounts to post and upvote.

That's what I actually said in a different expression.

Why shouldn't I make suggestions? Do you think you speak for the entire usership here if you claim that my proposals don't work? And to what proposals do you actually refer?

Can't remember that we ever talked about this topic. I don't understand what arguments you're talking about.

Sort:  

It's also not possible to restrict people to only have one account. Trying it is pointless, even centralized platforms like Facebook fail trying.
It doesn't work because it's impossible, not because I say so. You can of course ignore reality and propose it, it's just pointless. And blaming Steemit and the witnesses to have it designed wrongly while it could easily be abolished is hilarious to anyone who understands that.

I hear this argument for the first time that multiple accounts are impossible to avoid. As long as I read something to the contrary somewhere else, you can't tell me different sources outside of Steemit that prove it, it's exactly your word I'm supposed to believe. Especially since, should that be true, the quote I used from Pi is therefore a false announcement by the operators of Pi?

Apart from that, my statement regarding the simple change referred to the self-voting function.

Then I ask you concretely: How high exactly would the effort be to abolish the self-vote function? What would have to be technically changed?

Questioning the design you perceive as blame?

Blame as in it's their decision that self-voting is possible, and if they wanted they could change it. They just can't.

For the self-voting function, we're running in circles there. That would only make sense if you could guarantee that abusers cannot have multiple accounts.

The quote from Pi is definitely a false announcement. You don't have to take my word for it, just think a bit how that should work.
Every possible option has flaws which render it useless. IP addresses aren't static. Device identifiers can be changed, cookies deleted, and most people have multiple devices anyway. Phone numbers are easy to obtain. ID documents don't prove you're the one sending it in.

Online banks use post ident, where your identity is verified in person by the mail man, or video ident, where a person checks that you're really the one with the document. Those are basically the only two more or less secure ways. And even there fraud is happening. Accounts can change owner once they're verified - with a bank account that'd mean trouble for the one who opened it. With a blockchain account - who cares?
Besides, it's not a good idea to store ID documents on a public blockchain ;)

There's the old meme "On the internet, nobody knows you're a dog". It means you can completely make up your identity. And you're not restricted to one. Trying to enforce it will only affect the technically ingenuous.

Yes, the mechanism between the verification of a person and an account is possible, although with some difficulties. I'm still skeptical about the initial design. From my perspective it must have been clear to the developers and programmers, because if I code the technical possibilities, I'll know what problems I'll encounter. Abuse in the digital world isn't new.

At Steemit, there are no particular hurdles to registering multiple accounts. I see a difference between "completely impossible to get 100% protection" and nevertheless efforts in strong protection to at least keep away the first abusive intentions. Of course, anyone who really puts criminal energy into it and wants to hack a system at all costs will, it seems, find ways to cross these borders. But I don't necessarily have to leave the window wide open. Please correct me, if I am wrong in this but I have the impression that it's quite easy to start many accounts here.

To fight the abusers and then one day it would be finished, I may even understand. But for how long does one want to fight and put up energy with all necessarry excecutive actions when the design can't be changed for more security?

This actually turns people down.

"Yes, the mechanism between the verification of a person and an account is possible, although with some difficulties."

I explained to you why it is not. You can keep assuring yourself that it is, but that doesn't change reality.
Also, steemit != steem. But I won't bother to try explaining the differences between backend and frontend to you now, as you proved that you don't care about explanations by ignoring basically everything I said in the last comment.

Enjoy wasting your time with thinking about "solutions" without understanding the basics of the technology. I'm out.

It is not clear to me why you refer to your professional competence in this conversation, but then avoid concrete critical questions and do not name reasonable sources that support what you say.

My question to you was what exact technical processes are needed to deactivate the self-vote mechanism? You didn't answer that.

So the difference between absolute protection and relative protection doesn't mean anything to you? What additional security barriers exactly did Steemit set up over the last two years to prevent multiple accounts when setting up accounts? Not answered.

What else are you talking about? What "solutions"? You mean, my post on "the buttons"? Man, that is an experiment referring to debates I observed here. It's another form of a democratic voting method in systems. Why are you taking personal what I say?

When you debate about castles in the sky, no democratic process will ever help in building them.

I answered that it's not possible. You didn't like that answer, so you dismissed it. Any measures Steemit could implement are on the interface side. The ones they have (phone number, email, manual check) are set up because it costs them to create accounts. Anyone with a bit of steem can create as many accounts as they need in seconds. There is no technical way to prevent that.

There is no such thing as relative security in software development. Either a system is secure, or it isn't.

As a blockchain developer who has worked with this technology for more than 6 years, I am your reasonable source. If you dismiss this, I am wasting my time.