A new “consensus”: The Tangle Multiverse [Part 4]

Overview of the status quo

  1. Calculating the cumulative weight is expensive and doesn’t scale very well.
  2. Honest transactions have a chance to be left behind even though they are not even attaching to a conflicting transactions (i.e. because they were unlucky and didn’t gain enough weight).
  3. The random walk is gameable and opens up all kinds of attack vectors (like parasite chain attacks, blow ball attacks and so on).

The new voting scheme

1. Tracking opinions of transactions using realities

TX4, TX5, Tx7 and TX8 are honest but belong to the future cone of the CONFLICT and inherit its “opinion”

2. Polymorphic transactions

Setting the branch liked switch to 0, allows us to reference transactions without having to “inherit” their opinion.
TX4, TX7 and TX8 are able to reference the disliked transactions without having to inherit their opinion.

3. Local modifier based initial opinion

4. Tip Selection Algorithm

  • We maintain two lists of tips: strongly liked and partially liked.
  • We select the trunk uniformly at random from the list of strongly liked tips.
  • We choose the branch also from the list of strongly liked tips.
  • We set the branch like switch to 1 — the transaction strongly likes both of the referenced transactions.
  • We choose the branch from the list of partially liked tips.
  • We set the branch like switch to 0— the transaction strongly likes the trunk and partially likes the branch.
  • We remove the referenced transactions from their corresponding tips list and add the new transactions to our list of strongly liked tips.
  • We add the new transaction to our list of partially liked tips.

5. The Future Liked Cone (FLC)

  • solely out of “strongly liked” references or
  • solely out of “strongly liked” references with a single “partially liked” reference at its end
Since none of the transactions are conflicting, all of the transactions “strongly like” both of their referenced transactions (solid line).
Nodes can pick up honest transactions behind disliked double spends by setting the branch like switch to 0.

6. Cumulative weight (based on mana)

The FLC of tx1 contains statements of more nodes (and potentially more mana) than the FLC of tx2.

7. Numbered transactions and changing opinions

8. Consensus rule — the heaviest reality wins

9. Faster decisions — collaborative transaction issuing

Summary

--

--

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store