RE: Fake STEEM, SBD, witness, market, condencer and so on: Let's test steem for real from now on!

You are viewing a single comment's thread from:

Fake STEEM, SBD, witness, market, condencer and so on: Let's test steem for real from now on!

in steemit •  6 years ago 

This is a good idea, but reseting the blockchain state has some issues:

  • How fSteem/fSBD is distributed. If we go stright to HF21, it will be hard for people (including Steemit Inc.) to get tokens at first. One solution would be for the fake network should just give a bunch of coins to a faucet service
  • Depending on how fSteem/fSBD is distributed, accounts could too many transactions (and therefore fill up blocks), which would prevent others from transactions.
  • Test-whales could potentially elect witnesses that never produce blocks (by accident), and effectivly put the network in a unrecoverable state. Unless the elected witness start producing, which might not always be possible.
  • Hardfork block numbers are hard coded. This would result in fake-HF20 not happening until about 2 years after fake-net launches. Hardfork block numbers will need to be reduced a lot (maybe give each hardfork a day so we can test transitions between hardforks)
    • Although, allowing interactions with old forks might cause issues: certain things in the chain code rely on certain events on-chain happening/not-happening. In some cases hardfork checks are removed because a certain thing never happened on the mainnet. For example, at one hardfork, several accounts got a balance increase, for free. (because their Steem got stolen).

BTW replay attacks are protected against already. Chain IDs prevent them.

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:  

@smitop, excellent food for though! You are tackling two very important issues:

  1. Security and reliability of fake-steem. Most of your concerns here would be solved with good planing. For example, making a closed launch and going public later. A window of 3 to 7 days before opening the gates seems reasonable to me.
  2. Hardcoded stuff. There are some reasons to hardcode things and fake-steem should be hardcoded as well. It means that the source code should be ready to skip certain things if it is running on fake-steem. Again, this is just a matter of start doing it.