Quote from: matthewh3 on January 17, 2016, 01:33:50 PM

Can anyone tell the major benefits of this update apart from what I understand is to help induce a free market for transaction fees. Won't it break the ability to accept zero confirmation transactions?

Nope. Those transactions are opt-in, meaning that if the transaction doesn't opt-in to use RBF, then it can't be Replaced later.

Quote from: matthewh3 on January 17, 2016, 01:33:50 PM

As there are currently methods to do so with great certainty that a zero confirmation transaction will go through the next block.

Not really. The currently consist of CPFP, waiting for the transaction to disappear, or attempting an RBF transaction without really having RBF enabled anywhere. All of them take a while and are rather difficult to do.

Quote from: matthewh3 on January 17, 2016, 01:33:50 PM

And also doesn't this break the core ethos of bitcoin being a pure push system only.

How does this break that?

Quote from: matthewh3 on January 17, 2016, 01:33:50 PM

I've read the RBF transactions update is being merged into bitcoin core at the next update.

Already merged in.

Quote from: matthewh3 on January 17, 2016, 01:33:50 PM

Isn't that a little hasty for such a drastic change to the protocol to be brought in so quickly.

It isn't a protocol change, it is a node policy change. This change is something that only nodes will decide whether they want to accept the transaction into their mempool or not. It has nothing to do with consensus or protocol.

Quote from: matthewh3 on January 17, 2016, 01:33:50 PM

After it took us a good several months to agree on re-increasing the blocksize.

That is because this is a node policy change, not consensus or protocol. Consensus changes and protocol like increasing the block size limit require a fork, hard fork in the case of block size and soft fork in others like OP_CLTV. Forks require consensus and agreement since those can fork the blockchain. Node policy changes don't since it is specific to individual nodes, not the entire network nor the blockchain. Since this is just a node policy change, nothing is being affected on the network, just whether nodes want to accept or reject RBF transactions.

If you think that this change was too hasty, I'm here to tell you it is not. The subject of RBF has been discussed for years already. Peter Todd has had the code implemented in a fork for a while and this code has been tested and tested and discussed for ages. In fact, the idea of replacing transactions is not original. It was originally implemented by Satoshi but removed due to DoS concerns, which have been fixed by requiring that the fee be raised. Hence Replace-By-Fee.