Bitcoin Forum

This is an article by Mike Hearn against the proposed RBF (Replace by Fee) "feature", which is the ability to double-spend a transaction by sending a higher fee the second time, which would replace the first transaction by the fact of the higher fee on the second one. This would enable guaranteed 0-confirmation double-spends to be executed. Bad idea. This post explains the full ramifications of this if it were to be implemented:

The Article:
https://medium.com/@octskyward/replace-by-fee-43edd9a1dd6d (https://medium.com/@octskyward/replace-by-fee-43edd9a1dd6d)

Reddit Post with comments:
http://www.reddit.com/r/Bitcoin/comments/30lxo4/replace_by_fee_a_counter_argument_by_mike_hearn/ (http://www.reddit.com/r/Bitcoin/comments/30lxo4/replace_by_fee_a_counter_argument_by_mike_hearn/)


Help spread the awareness so foolish "features" aren't implemented into Bitcoin. It's in all of our best interests.

Peter Todd is the supporter and promoter of RBF by the way.

What is the rationale being the Replace by Fee proposal?

What will be the purpose of such feature and what problem does it solve?

I can't really think of any positive benefits. Especially none that outweigh the huge drawbacks.

Is this implemented on Viacoin (his coin)?

Good question. I'm not sure.

Just in case others are wondering, you are referring to Peter Todd, the chief developer of Viacoin (http://www.coindesk.com/peter-todd-joins-viacoin-development-team-chief-scientist/).

Well a "double spend" scenario will render the currency worthless and nobody would support it.

I actually thought the first transaction is reversed / cancelled by the second transaction, when you decide to increase the fee.

Merchants will also pull out, if this feature is introduced with flaws... The whole idea is not to have "double spending"  ??? ???

Lol i doubt theyd open up double spends arbritarily... It would just allow you to somehow null the first tx that may be taking too lomg by sending another with more fees so it goes thru faster.. Course all edge cases would be tested but really that is a pretty tiny usecase if thats all they are trying to get.. Not worth the risk if true

First, let's be more precise. It would enable guaranteed 0-confirmation double spends. Note that 0-confirmation double spends are already possible, but they are currently not guaranteed to work.

One benefit of replace-by-fee is that if a transaction is not included in any blocks because the amount of fee paid is too low, the amount can be increased.

When reading Mike's argument against RBF, keep these points in mind:

• Accepting transactions with 0 confirmations will always be risky, and the risk cannot be reduced without relying on a trusted third party.
• There is no number of confirmations that guarantees that a transaction cannot be double spent.
• Replace-by-fee is inevitable because it is accepted by the Bitcoin protocol and it can increase mining revenue. You can complain all you want about how it is a bad idea, but it won't make a difference.

I agree it is very bad idea, but how well works when you create new transaction using the unconfirmed output and add higher fee, lets see 2x normal fee. Does it help confirm both transactions quickly, any real test data?

Anyway there is missing easy way to do this in Bitcoin core unless you use coin control of course.

Bitcoin needs transactional reversibility as much as a person needing another hole in their head.

Dear Peter Todd,

Please implement RBF in VIACoin first and see how that goes.

This is one of the worst ideas I have heard of. There are plenty of useful features omitted from Bitcoin because of a slight chance they are not 100% safe. There is a massive chance Replace by Fee is not 100% safe and this proposal needs burying fast.

Replace by fee makes double spending easier, no one will support this change. The child pays for parent proposal is good, can help with speed up feeless transactions.

Not sure if you're sarcastic, but it is actually a good idea.  It could help via even...

Every miner can implement "replace-by-fee" and/or "child-pays-for-parent" on his node.

stop talk about 0-confirmation transaction.
they don't exist when you have more than 6000 nodes to check the ledger in real time ...

http://imagizer.imageshack.us/a/img538/7127/5Ub4Yp.jpg


use the fees about financial transaction to solve any problem on earth ...  ::)

Peter Todds contributions are exclusively useful for scammers and other criminals.

Thanks for the clarification. This is true. I updated the OP.