It would be nice if there was an option to download non-best blockchains and prioritise (even replacing existing mempool txns) anything confirmed in them. This would make confirmed transactions somewhat stronger.
Miners should prioritise transactions in forked chains #6574
issue luke-jr opened this issue on August 19, 2015-
luke-jr commented at 5:46 PM on August 19, 2015: member
-
casey commented at 2:54 PM on August 20, 2015: contributor
What's the rationale for this feature? So that if a user sees that a transaction is confirmed in a block that is then orphaned, it's likely to get reconfirmed quickly?
-
TheBlueMatt commented at 11:05 PM on August 20, 2015: member
I wouldnt consider it "stronger" unless its also rational to prioritise those transactions, which it is not.
-
luke-jr commented at 2:57 AM on August 21, 2015: member
Fraudulent double-spending clearly harms the Bitcoin economy, so I would say miners have a light incentive to make an attempt to prevent it when practical. (Non-fraudulent double-spending does not, AFAIK, have any case harmed by this behaviour.)
There is a subtle risk to this behaviour I should document: if a transaction is the cause of the stale chain being non-best, we would not want to re-mine that transaction. I don't know if there's a good way to determine and prevent this.
-
TheBlueMatt commented at 7:24 PM on August 21, 2015: member
I would have no problem with this as long as the transactions you are replacing have lower fee rates, and if you never prioritize transactions above others which have a higher fee rate (making the decision to do so irrational). I really dont think we should be adding new behavior which is not in a rational miner's best interest, we should be moving the other way (ie eventual RBF, etc).
- laanwj added the label Mining on Aug 24, 2015
-
Sjors commented at 6:49 PM on March 16, 2018: member
Doesn't the fact that transactions are mostly ordered by fee provide enough likeliness that a replacement block has mostly the same transactions?
- MarcoFalke added the label Brainstorming on Mar 17, 2018
-
MarcoFalke commented at 6:10 PM on March 17, 2018: member
Discussion should probably move to the mailing list
- MarcoFalke closed this on Mar 17, 2018
- DrahtBot locked this on Sep 8, 2021