Re: [Bitcoin-development] Ninki Wallet view on blocksize debate

[Stephen Morse <stephencalebmorse@gmail.com>]

[-- Attachment #1: Type: text/plain, Size: 2145 bytes --]

Ben,

How does your wallet calculate the fee that should be paid to miners? Do
they automatically adjust when transactions take a long time to be
confirmed? And how does it respond when transactions are not mined
successfully, such as when blocks are full?

I strongly urge Gavin to withdraw from this standoff and work with the
> bitcoin core devs via the existing and successful bip process.
>

The BIP process has not resulted in any hard forks, so this is a little
different. While I don't like M&G's proposed solution of convincing miners
and services to switch to Bitcoin-XT, I recognize that it is done out of a
sense of urgency. These types of changes take a long time to roll out, and
we should start them before it is too late.

This whole debate comes down to: what is more risky, a consensus hard fork
or letting bitcoin exceed its imposed capacity limits? The former could
result in many services not being compatible and even loss of funds. The
latter could result in software failures, instability, and inability to
transact: essentially, what bitcoin is supposed to be good at. Both are
dangerous and could result in a significant loss of public confidence.

Something needs to be done, that's for sure. In the short term, I think we
need to do one of two things:

   1. All miners and wallet developers need to upgrade to support
   first-safe RBF, to allow for double spending one's own transactions when
   they lack sufficient fees to merit confirmations. Wallets also need to
   randomly request transactions from blocks to see what kind of fees are
   being paid to get confirmations, so that fees can be paid dynamically
   instead of with hard-coded values.
   2. We can implement either Gavin's or Jeff Garzik's proposal to change
   the consensus parameters around the block size limit.

So Ben, if really don't think that going with #2 is the right way to go
(even though everyone agrees that we will need to increase the block size
limit eventually anyway, why not now?), then I hope you start to work hard
on implementing #1 so that your wallet software can handle hitting capacity
limits gracefully.

Best,
Stephen

[-- Attachment #2: Type: text/html, Size: 2729 bytes --]