Some of you are saying that this BUIP may be premature, and that we could explore other options. You might be right. While this BUIP fully solves the problem of miners overwhelming other miner's AD and issuing huge blocks, and I believe that it is a "good" solution, there may be other "better" options. This BUIP was issued under time pressure as a counter to BUIP038 which I think is not deployable from a miner's perspective and doesn't fully solve the "huge block issuance" problem.
But note that "AD" is not a "consensus parameter", so we can pretty easily change it later if a better option appears -- a "good" solution now is good enough.
On the other hand, a simple solution is to make miners aware of the attack so they use higher AD values. No coding required. As per sickpig's analysis (posted in the BUIP038 discussion), the cost of the attack gets VERY expensive very quickly.
So tldr; please vote this BUIP "yes", or vote BOTH 038 and this BUIP "no", but don't vote for 038.
@freetrader, the C++ code has not yet been written...
But note that "AD" is not a "consensus parameter", so we can pretty easily change it later if a better option appears -- a "good" solution now is good enough.
On the other hand, a simple solution is to make miners aware of the attack so they use higher AD values. No coding required. As per sickpig's analysis (posted in the BUIP038 discussion), the cost of the attack gets VERY expensive very quickly.
So tldr; please vote this BUIP "yes", or vote BOTH 038 and this BUIP "no", but don't vote for 038.
@freetrader, the C++ code has not yet been written...