Algorand team's Response to Yongge Wang's "Another Look at ALGORAND" & Mauro Conti,Ankit Gangwal,Michele Todero's paper
- 2019.5.11 Yongge Wang - Another Look at ALGORAND
- “In this paper, we identify several design flaws of the ALGORAND protocol”
- 2019.1 Mauro Conti,Ankit Gangwal,Michele Todero - Blockchain Trilemma Solver Algorand has Dilemma over Undecidable Messages
[5:20 AM] Patrick B | nerd: Next - where can I find details on the current explicit consensus mechanism? The original papers are quite old and people still quote it as if it’s correct. It talks about 1 MINUTE block times. Clearly, much has changed. There should either be an updated paper, or a new paper with the current mechanism This should be updated while I’m here, someone asked in another forum about this old paper: https://arxiv.org/pdf/1905.04463.pdf Where can I find details on why this isn’t a problem for Algorand at the moment
1 [6:15 AM] Fabrice | Algorand Foundation: @Patrick B | nerd To my knowledge, https://github.com/algorandfoundation/specs/blob/master/overview/Algorand_v1_spec-2.pdf is the most up to date description of the consensus. GitHub algorandfoundation/specs Algorand Specifications. Contribute to algorandfoundation/specs development by creating an account on GitHub.
[7:27 AM] Patrick B | nerd: Has Algorand or the foundation addressed the claim in that paper ? Given that a single-block is agreed on in step 1 and voted on in step 2 - with an effective 2-stage commit I don’t see how there could be a ‘fork’. [10:18 AM] Fabrice | Algorand Foundation: Algorand has a formal proof that it cannot fork (under the required assumption of enough honest parties): https://medium.com/algorand/formal-verification-of-the-algorand-protocol-bbde5a52b830
[11:26 PM] Nuno: https://www.researchgate.net/publication/330725913_Blockchain_Trilemma_Solver_Algorand_has_Dilemma_over_Undecidable_Messages [11:27 PM] Nuno: Comments, is this valid/up to date? December 27, 2020 [12:19 AM] AndrewGardner: [3:58 AM] Fabrice | Algorand Foundation: @Nuno To my knowledge, these attacks do not apply to the Algorand protocol as implemented. As mentioned in the paper, the authors did not test their attack on the actual protocol but on a simulator they constructed. (edited)
1 [3:59 AM] Nuno: Thanks fabrice I had this impression specially when I read that they actually did not test.