You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
This repository has been archived by the owner on Oct 4, 2019. It is now read-only.
An extra reward to the winning miner for including uncles as part of the block, in the form of an extra 1/32 (0.125ETC) per uncle included, up to a maximum of two (2) uncles
Usage of the singular form and the placement of commas suggests that it's a single reward for all of the uncles included. Therefore any rounding down of the possibly non-integer result should be done for the whole reward.
This code however, calculates a reward per uncle individually and sums them up (in practice - multiplies). This may lead to accumulation of rounding errors.
The code is perfectly fine if we interpret the above definition to mean that these are separate rewards per each uncle, which have already been subjected to rounding down. However, we at Mantis team, interpreted it as explained above, and the difference may lead a network split in further eras.
All that being said, the ECIP-1017 definitely requires some clarification.
The text was updated successfully, but these errors were encountered:
rtkaczyk
changed the title
ECIP-1017 monetary policy: possible interpretation
ECIP-1017 monetary policy: possible misinterpretation
Sep 19, 2017
Agreed that this wording could use clarification. My interpretation had emphasized the in the form of ... per uncle clause, which moves reward to a more abstract idea, eg
"For earning first place, Joseph received a reward of a blue ribbon and a knitted hat."
As stated in the ECIP-1017:
Usage of the singular form and the placement of commas suggests that it's a single reward for all of the uncles included. Therefore any rounding down of the possibly non-integer result should be done for the whole reward.
This code however, calculates a reward per uncle individually and sums them up (in practice - multiplies). This may lead to accumulation of rounding errors.
The code is perfectly fine if we interpret the above definition to mean that these are separate rewards per each uncle, which have already been subjected to rounding down. However, we at Mantis team, interpreted it as explained above, and the difference may lead a network split in further eras.
All that being said, the ECIP-1017 definitely requires some clarification.
The text was updated successfully, but these errors were encountered: