@Dawson and @madMAx43v3r together softly control 2/3 of netspace (8 for nossd, 12 for GH, out of 31 we assume)
If these two were to modify their node to not accept blocks containing pre-farm transactions other than to a burn address and their farmers to not produce such blocks, then this would become the longest blockchain.
I believe the pre-farm is at least twice the size that it should be, so taking out just two of the pre-farm addresses could be an idea. Chia inc. could accept this proposal by burning the two chosen addresses.
2 Likes
Let me see, so do we need Chia anymore?
From that chia doc about majority attack / first table (Attacks and Countermeasures | Chia Documentation), if the attacker controls 2/3 of space, it needs just 1 timelord that runs at half of the speed of the fastest to create a new and looks like a stable fork (my read of that section, may not be correct).
Looks like Max and NoSSD may have a secret / desperate buyer right now.
1 Like
April already… ( 20 chrs )
3 Likes
Time flies doesn’t it
1 Like
Forks have already tried this (Chives comes to mind, and somehow they still have a little trading volume! https://www.coingecko.com/en/coins/chives-coin). If it were just a matter of dumping the team and dumping the pre-mine, the market would have already spoken 
.
It would be a different type of fork, one that would create a dishonest fork (chia’s terminology) on the chia’s blockchain (check that linked doc I provided), I think forcing every single node to follow that dishonest timelord (run by NoSSD / GH). In such case, not sure if chia would be able to resolve this problem with software updates. (Maybe that would boil down to some kind of timelord war (timelord speed).)
However, at this moment, as @Bones stated, this is just April based rambling.
Although, this forced chia to desperately start working on CHIP22. This work will be relevant only if GH and NoSSD will adopt it (not sure if there are any incentives for them to do it, though).
Therefore, as mentioned, maybe the best / only option for chia right now is to buy either NoSSD or GH and open up the code.
1 Like
My (slightly humurous) proposal would produce valid blocks that in the end the loyal nodes would have to accept. This is because with 2/3 netspace the rebel nodes produce the chain with the highest weight. No timelord changes are needed
Edit: I admit that I don’t really understand it, just in it for the cash
1 Like
Take a look at that chia doc I linked above, the first section/ first table. The very first row states that with no timelords (TL) the majority attack is not possible, even if the attacker controls the whole netspace. With 1 super-fast TL (2x speed of the fastest TL), 33.3% of netspace is enough to take over the network. Both NoSSD and GH have already crossed that threshold. Not sure if what NoSSD has counts as controlling that netspace, though, but assuming that they have such control, they can focus on TL right now to make it happen.
As stated before, that is how I read that section and I have not much if any blockchain domain expertise, so it is hard for me to relate that doc to other parts.
Just to better clarify it. My understanding of “netspace” with respect to that doc is that it is the effective disk space (in normalized K values, i.e., k32 = 1 unit, k33 = 2 units, …, or as others want to call it TiBe-ffs). Therefore, it doesn’t matter how many nodes / farmers the rogue site controls (for the sake of the argument it could be just 1 node and 1 farmer to have a majority attack described in that section).
1 Like