[bitcoin-dev] Speedy Trial

["Russell O'Connor" <roconnor@blockstream.com>]

[-- Attachment #1: Type: text/plain, Size: 2374 bytes --]

On Thu., Mar. 10, 2022, 08:04 Jorge Timón via bitcoin-dev, <
bitcoin-dev@lists.linuxfoundation.org> wrote:

>
>
> You're right, we shouldn't get personal. We shouldn't ignore feedback from
> me, mark friedenbach or luke just because of who it comes from.
>

For goodness sake Jorge, enough with the persecution complex.

As the person who initially proposed the Speedy Trial deployment design, I
can say it was designed to take in account those concerns raised by luke-jr
and the "no-miner-veto" faction.  I also listened to the
"devs-do-not-decide" faction and the "no-divegent-consensus-rules" faction
and their concerns.

The "no-miner-veto" concerns are, to an extent, addressed by the short
timeline of Speedy Trial.  No more waiting 2 years on the miners dragging
their feet.  If ST fails to active then we are back where we started with
at most a few weeks lost.  And those weeks aren't really lost if they would
have been wasted away anyways trying to find broad consensus on another
deployment mechanism.

I get that you don't like the design of Speedy Trial.  You may even object
that it fails to really address your concerns by leaving open how to follow
up a failed Speedy Trial deployment.  But regardless of how you feel, I
believe I did meaningfully address the those miner-veto concerns and other
people agree with me.

If you are so concerned about listening to legitimate criticism, maybe you
can design a new deployment mechanism that addresses the concerns of the
"devs-do-not-decide" faction and the "no-divegent-consensus-rules"
faction.  Or do you feel that their concerns are illegitimate?  Maybe, by
sheer coincidence, all people you disagree with have illegitimate concerns
while only your concerns are legitimate.

A major contender to the Speedy Trial design at the time was to mandate
eventual forced signalling, championed by luke-jr.  It turns out that, at
the time of that proposal, a large amount of hash power simply did not have
the firmware required to support signalling.  That activation proposal
never got broad consensus, and rightly so, because in retrospect we see
that the design might have risked knocking a significant fraction of mining
power offline if it had been deployed.  Imagine if the firmware couldn't be
quickly updated or imagine if the problem had been hardware related.

[-- Attachment #2: Type: text/html, Size: 3274 bytes --]