[bitcoin-dev] Modern Soft Fork Activation

["Yosef" <asix@disroot.org>]

tl;dr How about 80% ?

The fallback to BIP-8 makes sense, but it's not a graceful one and we absolutely prefer BIP-9 to succeed. A failure to reach 95% readiness signalling means 2.5 years delay, 3.5 years in total, not yet counting.

95% can prove difficult to achieve. Some % of negligent miners that forget to upgrade is expected. Completing that to 5% is not too difficult for a small malicious minority trying to delay the activation. This is the issue Matt's goal #5 aims to prevent, and while the fallback to BIP-8 helps, BIP-9’s 95% requirement makes it worse by allowing quite a neglected minority to force a dramatic delay. Also note how in such case it would have been better to skip BIP-9 altogether and maybe save 1.5 years.

Matt mentions the 95% requirement under goal #3 "Don't (needlessly) lose hashpower to un-upgraded miners". If that is the trade-off, I'd say 2.5 years delay is worse than a momentary loss of hashrate. The protocol is quiet resilient to hashrate fluctuations and, as others mentioned, at that point miners don’t just signal, but lose coins if they don't upgrade, so the hashpower loss is expected to shortly correct. This also means goal #4 is not really effected.

On Sat Jan 11 14:42:07 UTC 2020, Anthony Towns wrote:

> For me, the focus there is on Matt's first point: "avoid activating [or merging, or even proposing] in the face of significant, reasonable, and directed objection"

I agree, and believe the outreach and review process around taproot are maybe the best we had so far. Notably goal #1 should be mostly satisfied already at merge time, so risking 3.5 years delay after that, seems excessive.

BIP-91 “Reduced threshold Segwit MASF” was deployed by miners specifically to reduce the 95% requirement down to 80% during the segwit drama. While hopefully taproot won’t produce any such excitements, the events around segwit activation and the weird “hash wars” meme that later followed – might encourage some to try similar games again.

The difference between `5% minus apathetic-miners` and `20% minus apathetic-miners` is dramatic and can make such attempts an order of magnitude more difficult.

The taproot process is looking great so far, I feel it will be a mistake to put it on a route that can easily extend to so many years. I suggest keeping Matt’s proposal as is but decrease BIP-9’s 95% threshold down to around 80%.

Yosef