My understanding is PoW and PoS systems provide some kind of basis of introducing some kind of…
Subramaniam (Subbu) Mani

Thanks Mani. I will be publishing my whitepaper soon enough to answer in detail. Briefly, in simple terms,

  1. PoW and PoS are themselves ‘gameable’. PoW through concentration of hashpower and PoS through concentration of wealth. The current implementations of trust in PoW and PoS systems are very much susceptible to different types of attacks — technological, economic etc.
  2. The aim of TrustChain is to have a trustworthy and democratic blockchain where anyone without complex machinery or extreme wealth can participate.
  3. 1 TrustToken is allocated to a validator on one successful vote for verifying one successful transaction in the blockchain. This is similar to a mining reward, except that TrustTokens are not transferable and indeed not usable in any way!! It is used by the system only as a count.
  4. Validators are nodes which participate by verifying (voting/mining) transactions in a blockchain. This is a very blockchain specific definition and is more than doing sanity checks on certain transaction parameters. Validators in TrustChain and in PoS systems are similar to miners in PoW.
  5. The basis for choosing S/M/J is to assign trust levels based on previous participation in successful transactions and certain other parameters, which will improve security against different types of attacks. It also allows for the onboarding of validators who may not be ‘immediately trusted’.
  6. The basic principle in Byzantine Fault Tolerance is that if more than 33% of the nodes are ‘bad’, then consensus is not guaranteed. So any blockchain (and TrustChain also) MUST assume at most 33% of actors may be malicious.
  7. TrustChain selects a small number of actors using verifiable random algorithms into a committee (very small representative subset of the whole population) based on the accumulated TrustTokens and the tier in which they belong. This turns out to have the same optimal tolerance as that of the generalized BFT (max 33% bad), so we are as good as can be, but not more or less!

This is only a brief and simplified explanation. Rest in whitepaper!