Luke dashjr on Twitter “Community came to consensus on BIP8. These devs are IGNORING that and pushing their own agenda instead. It is an attack on Bitcoin, not a good thing.”

View Reddit by cy9h3r9u11kView Source


Leave a Reply

Your email address will not be published. Required fields are marked *

GIPHY App Key not set. Please check settings


  1. There was original;y general support for one approach but Luke (with help) deadlocked it by insisting on optional components like mandatory signaling (so miners that don’t update immediately get forked off) and lockinontimeout=true (which would potentially the network consensus split rather than activation safely failing and being ready for another retry if 90% hashpower didn’t deploy it). Other people both disagree on how safe those options are and disagree that the “activation landscape” justifies using them even if they were to agree that they were not *very* risky.

    It has been a long time since a consensus rule has been activated and everyone imposes their own speculations, hopes, and fears on the process… as a result there is a lot of subjectivity and not a lot of objectivity on exactly which time-frames, thresholds, and other trade-offs are justified for an activation. Lots of opinions, few facts. With everything so subjective it’s no surprise that some people found lots of details to argue about and their arguments were nearly irreconcilable because they came from different premises.

    To replace the deadlocked proposal speedy trial was proposed which basically avoids any hard tradeoffs in activation by just trying a simple fast activation and learning from how it goes. It replaces subjectivity by doing something that will either work or fail fast. If it works, Great– and if it doesn’t work it doesn’t harm anything, we all learn from how it goes and have objective data to inform next steps, and the delay isn’t that substantial.

    Luke supports speedy trial, AFAICT, but is unhappy with some obscure technical details in the implementation: It has a timeout is based on a time (like most prior soft-forks) rather than on block-height (like the BIP8 based proposals he deadlocked). There are extremely minor pluses and minuses for time vs height, deep in the technical weeds. Timestamps are easier to implement for one of the testsnets and and map better to human dates/times while heights are easier to review how they interact in an implementation (a point somewhat mitigated by the fact that timestamps are the longstanding tool…)… minutia like that. Either can be made to work. The implementer settled on using time, as far as I can tell support for going with that among people who are paying attention is overwhelming.

    Once speedy trial is done (either it expires or activates taproot) none of this activation minutia has any lasting effect on the system– the code will likely even be removed. The activation logic is just temporary scaffolding.

    As far as I can tell, 99% of the people that are even aware of taproot don’t care about the activation method (beyond it working and being relatively safe)l, of the 1% that do 99% have no particularly strong opinions about the time vs height stuff– it’s shed-painting minutia. The claim of ‘community’ consensus comes from mixture of [misreading people who were supportive]( of all options (or the prior proposal that deadlocked) as being in favor of Luke’s proposal, and mistaking his own filter bubble of a couple people as being more representative then the many times the number of people speaking up in discussion.

    In any case, a bit frustrating but I think ultimately not a big deal. Someones always gotta climb the Reichstag dressed as Spider-Man, and I guess this time it’s Luke’s turn. On one hand it’s good that people care enough to argue out details– for months after months after months–, but it’s also sad that Luke has become so aggressive that he’s fighting to the death (well not quite, but characterizing this as an attack???) over fairly minor details— both because of the personal cost in exhaustion but also because it blows political capital that could be better conserved for issues that actually matter.

    There is a lot of great stuff left to do. Hopefully people won’t be so fed up with little details being blown out of proportion that they lose the will to do it, as that would be a true loss on a much greater scale than some activation technical details that no one will remember 5 years from now.

  2. What’s he talking about? There wasn’t consensus so we’re trying speedy trial and giving the miners 3 months to prepare and signal support. If there isn’t 90% consensus then it wont activate. If it doesn’t activate then we activate it with a user activated soft fork (bip8 lot=true). I don’t get it, Luke didn’t get what he wants so he calls it an attack?

    Edit: I do support bip8 lot=true though.

  3. I hope this isn’t another dev that’s going to rage-quit over small details.. Remember how Andresen disgraced himself..

    During a decentralized decision making process, many will wind up not getting *exactly* what they wanted. That’s just the way it is.

  4. From what i gathered, please correct me if i’m wrong:

    Height-based is easier to implement and review.

    Time-based is harder to implement and review but not hard enough to be an issue for some.

    Since this is bitcoin, why not land on the side of less friction? I think i’m with Luke-Jr here.



What do you think?

The best NFT & DeFi project token sale, TimeCoin (TMCN)

The best NFT & DeFi project token sale, TimeCoin (TMCN)

The World Is Watching as Coinbase Goes Public