in

Andreas Antonopoulos is on the Fence with the Bitcoin BIP 119 Controversy & Thinks A Speedy Trial Activation is Entirely Inappropriate For Now

Andreas Antonopoulos is on the Fence with the Bitcoin BIP 119 Controversy & Thinks A Speedy Trial Activation is Entirely Inappropriate For Now

Andreas Antonopoulos is on the Fence with the Bitcoin BIP 119 Controversy & Thinks A Speedy Trial Activation is Entirely Inappropriate For Now from Bitcoin




View Reddit by Fiach_DubhView Source

Leave a Reply

Your email address will not be published.

GIPHY App Key not set. Please check settings

23 Comments

  1. I read over BIP 119. My initial thoughts are that this does sound useful. But those are just initial thoughts. I’m a dad, my parents are aging, and I have a very busy job.

    I don’t have 24×7 to think about bitcoin or what’s good for it. As such, I won’t upgrade my node to signal support until I’ve had more time to think about this.

    In general, I think it’s better if upgrades take years. If there’s no emergency, we incur risk by having it be ‘the norm’ for changes to go through quickly once a small group of intensely online people have decided amongst themselves that it’s not a risk.

  2. If you have looked into it at least a little bit, Jeremy Rubin has come off extremely bad in any interaction I’ve seen him in. He has kept brushing off concerns about consensus, and putting the burdeon of proof on people trying to keep the protocol simple, instead of taking the burdeon onto him, to show us how these features are really useful.

    The entire “soft signal” list, where he included people that didn’t say they would signal for this change to be done right now, through a speedy trial, but only showed interest in the BIP in general, has been extremely disingenuous, and even with the disclosure that these “soft signals” aren’t binding, he’s trying to make it look like everyone has reached consensus already, and there are some contrarians that are trying to slow down progress.

    This narrative he is painting is clearly far away from the truth, and I understand him wanting his changes to make it to Core, but he needs to take a look at the way the people developing SIGHASH_ANYPREVOUT have handled the same situation, in a more mature way.

    I see no reason we wouldn’t do a user activated soft fork instead of a speedy trial, and even though I don’t think the proposal would pass either of them, the risk is there.

    The 2 main problems I see as maybe possible are:

    1. Covenants could create whitelists and blacklists, and even though Rubin has assured us this change doesn’t allow for recursive covenants, I’m not sure it’s impossible to create covenants that propagate into the future, let’s say 100 million blocks, which by all means would be as good as forever. This could be disastruous for the fungibility of Bitcoin, because it would enforce KYC-like regulation at the protocol level. More research on this that clearly demonstrates this is impossible should happen.

    2. Enabling smart contracts on L1 is not something I see as an advantage. We’ve already seen how L1 SCs look on L1 — you either have a decentralized system with huge fees, or a centralized system that fails every couple of months. The main L1 for smart contracts is already looking into moving the execution of most smart contracts on higher layers.

    I can’t argue against the fact that the concept is cool, and it seems like it enables some pretty interesting features, but I think the simplicity and reliability of the protocol is more important than that.

    I, and most of the community, would have pretty much no problem with this BIP if it wasn’t for Jeremy trying to constantly push for it to be implemented by a miner activated speedy trial, while the community has not reached consensus yet. It’s a hard fork waiting to happen, and I hope we won’t have to repeat the whole 2017 blocksize wars type of deal.

    The process of adding features to Bitcoin should be slow, and difficult. This ossification is not an issue, it’s a feature and an ideal we should strive if we wish for Bitcoin to be a core global protocol of value transfer in the coming decades, and in perpetuity.

  3. These guys think that Bitcoin is a fucking mobile app, that it needs to be updated with more and more features every month. That’s the classical symptom of [featuritis](https://en.wikiquote.org/wiki/Featuritis), they are forcing Bitcoin to be a victim of their disease.

    Or maybe these are the symptoms of classical developer narcissism, they want to put in their resume that they are the authors of a bitcoin OP code.

    If they want useless features, do it off-chain. They can easily keep piling up features off-chain.

  4. This BIP does not belong to bitcoin. Possibly improve defense in depth or improve privacy/fungibility. Smart contracts in bitcoin does not make sense, we need to protect our currency from governments – to protect often poor, underbanked, or enslaved people. This is not helping. Thanks Andreas.

  5. Where can we vote on the disapproval of BIP 119.It has not satisfactorily passed consensus from all important players in the space.

    Not sure why this cant be done in the second layers and what is the fucking hurry?

  6. The more we ignore Andreas the better we as a community will be. Do we really care what people who promote scams to enrich themselves think?

    edit: I like how this was positive for a while then a bunch of shitcoiner noobies show up and are like “what did you say about my Andreas?!”

What do you think?

Bitcoin just touched “Accumulate”

Cryptos to Surpass Traditional Investments in 10 Years, Survey Finds