RE: OPEN LETTER TO STEEMIT INC., THE WITNESSES, AND THE WHALES

You are viewing a single comment's thread from:

OPEN LETTER TO STEEMIT INC., THE WITNESSES, AND THE WHALES

in abuse •  7 years ago 

None of what you just typed makes any sense. Voting happens directly on the blockchain. You don't have to explain to me the difference between the Steem blockchain and the various and numerous interfaces one could could use to access the blockchain.

Are you aware that it's possible to run some of these interfaces directly on your own machine? I have a blog post where I describe how to run a local version of Busy.

You're right, I don't like anything that attempts to centralize the system. If we are going to do that, then why even have a blockchain at all?

Steem isn't Steemit, or Busy, or Zappl. Each of the interfaces I am aware of requires a password, and could just as easily require 2FA.

Yes and it's the same password for all three sites. You log in using the same keys.

Authors get paid when people like you upvote their post.
If you enjoyed what you read here, create your account today and start earning FREE STEEM!
Sort Order:  

So, how does the blockchain refuse to allow votes that don't come through those interfaces?

Either the blockchain is open to voting with access that doesn't require the password, or it can also require additional information, such as 2FA.

That's pretty obvious, and you're a smart person. I'd be astounded if you hadn't grasped that without my saying so.

Edit: also, it no more is centralizing than requiring a password. Decentralization has nothing to do with it.

You have no clue what you are talking about. You really don't.

Ok, how would 2FA work on "the blockchain"? Keep in mind that the blockchain isn't a computer, it's just blocks of data. It would be the witness nodes (like the one I run) that would do 2FA.

How would this work? Actually try to think about it.

  • Person A casts a vote. It is signed with their private posting key.
  • The vote goes into a mem pool.
  • The vote is picked up by a witness node, but oops, it hasn't been 2FAd yet.
  • The witness sends a code to Person A's phone
  • Person A ...sends another message to the blockchain with the code ??
  • The witness would verify the 2FA..this would probably be minutes later, after many more blocks have gone by.

Even if this was possible, it's impractical as hell at best. Would you have to do it for every vote?

An email based 2FA would be even slower.

To add 2FA the routine that verifies the posting key can then call a routine that sends a text with a key to the phone linked to the account the posting key is valid for. Only when that additional key is verified does the vote then proceed to the mempool, and subsequently get written into a block.

The same mechanism that verifies the posting key can also require the key returned by text.

It's just data and software, and not even unique or especially complex code. Hell, Yahoo can manage it. I'm sure that you guys can as well. Either the password works, or it doesn't. Either the 2FA works, or it doesn't. If either doesn't, the vote never goes into the mempool and never gets written into a block. It's expensive, slow, and annoying, but no more bots.

The technical feasibility of 2FA isn't really the issue. ROI is.

You don't like getting rid of bots. I get that.

There isn't enough support for getting rid of bots to worry about it. The whales want the ROI. Stinc and you guys serve the whales, and if you don't you get the boot, or Steemit dies cuz the whales take their ball and go play another game. The minnows wanna win the lottery, which bidbots and whale updoots are.

There's no need to dance around the issue by claiming it's impossible, or centralizing (as if stake-weighted witness votes doesn't already centralize the ability to control the blockchain). The PITA it would be is probably enough to discourage folks from considering it.

We can skip the pretense that it's impossible, or even hard, technically.