299 Members
For discussing on-chain and off-chain governance of the Tezos ecosystem. See http://bit.ly/2rkSiUY for room info. Personal attacks and threats are not allowed. Be kind, nice and thoughtful. : )4 Servers

Load older messages

Timestamp Message
25 Jan 2019
12:57:39@et4te:matrix.orget4teTendermint is an easier transition, has been proven
12:58:40@et4te:matrix.orget4teAvalanche has 'issues' re security proofs
12:59:04@freelance_tulip:matrix.orggovernancy ^ what et4te said :)
13:01:27@et4te:matrix.orget4teThe basic issue with implementing Avalanche in Tezos is that it requires that nodes have an unbiased membership view - this requires a gossip protocol that is resistant to intrusion and can't affect nodes views of the network
13:02:49@et4te:matrix.orget4teTendermint on the other hand doesn't really care about gossip, just that it can achieve partial synchrony and that 2f + 1 nodes are not malicious when this happens
13:08:26@blindripper_85:matrix.orgBlindripperWhy do we even need another consensus model? What's bad with the actual implementation?

Why do we even need another consensus model? What's bad with the actual implementation?

it's not that the existing implementation is terrible. It's that BFT algorithms are safer and you get fast finality, which is critical for building things on top

13:16:47@blindripper_85:matrix.orgBlindripperMhhm. don't understand how it can reach finality faster if i need "a few hundred endorsers" instead of 32...
13:21:25@freelance_tulip:matrix.orggovernancydifferent algorithm: as Tezos consensus is now, it takes a number of blocks to be considered (probabilistically) final. The idea with Tendermint is that you take the larger set of bakers and thus set a higher threshold to approve each block...
13:23:06@freelance_tulip:matrix.orggovernancyyou get finality after each block, whereas in current Tezos, you get a sense of finality after a number of blocks (based on the chance that the chain will reorganize)
15:57:20@freelance_tulip:matrix.orggovernancy I should have also pointed out that the reason this higher threshold is possible w/ tendermint is because of the different synchronicity model Blindripper
15:57:45@freelance_tulip:matrix.orggovernancyit's not simply that the number of endorsers go up
28 Jan 2019
12:58:27@latte_jed:matrix.orglatte_jed joined the room.
21:30:21@somejack:matrix.orgsomejack joined the room.
30 Jan 2019
08:14:14@latte_jed:matrix.orglatte_jed set a profile picture.
4 Feb 2019
02:07:43@ezmat:matrix.orgezmat Blindripper: that would be a great question to ask on TSE ;) are we allowed to talk about tendermint/BFT on it then :) ?
14:54:50@felixlts:matrix.orgFelixLts | Chorus One changed their display name from felixlts to FelixLts | Chorus One.
9 Feb 2019
14:09:11@jmz101:matrix.orgjmz_xtzantipodes changed their display name from jmz101 to jmz_xtzantipodes.
14:13:48@jmz101:matrix.orgjmz_xtzantipodesRedacted or Malformed Event
just one example, dev A submits a valuable proposal that includes entirely new work but it has some flaws and is not accepted, dev B fixes the flaws and resubmits, who should get paid via the on-chain invoice?

cousinit: One criticism of on-chain governance is that it might incentivise 'package deals' which bundle popular and unpopular items in a single upgrade, which will gather use and legitimacy before they can be upgraded (or even forked) away from.

Making protocol proposals that stripped out the offending elements must be a relatively simple and profitable way to use a protocol development team. The realistic threat of this will cause protocol developers to not risk losing their work to the second-movers, and avoid mixing contentious or malicious protocol elements with popular ones.

14:20:50@jmz101:matrix.orgjmz_xtzantipodesInadvertent flaws are harder though, although the same principle applies, the threat of a buggy proposal will incentivise more rigorous testing; whether this is enough incentive to the proposer while keeping development costs feasible (tied ultimately to available market cap) is an open question
14:22:51@jmz101:matrix.orgjmz_xtzantipodesI believe we're only scratching the surface of protocol developer game theory and economics, there is a lot more to discover (and I look forward to it excitedly)
14:32:30@jmz101:matrix.orgjmz_xtzantipodes I also appreciated the discussion by 3rw4n , at the face of it some futarchic benefit might be available in a relatively riskless way (at least no need to subsidise market-makers). would benefit from better TGE auction mechanisms though, maybe descending-price
11 Feb 2019
04:35:30@altomedium:matrix.orgaltomedium joined the room.
12 Feb 2019
11:02:22@romarq:matrix.orgromarq joined the room.
14:58:32@mastequila:matrix.orgmastequila joined the room.
18 Feb 2019
20:56:41@freelance_tulip:matrix.orggovernancy changed their display name from goverNancy to governancy.

There are no newer messages yet.

Back to Room List