The issue is probably Core. It's centralizing. There should be many popular node implementations.
James OB spitting the hard truths on twitter. I agree with everything he says here: https://x.com/jamesob/status/1857049961235403101
No idea how this will turn out. For most questions it's not that hard to get an idea of what the future will bring based on incentive analysis and historical precedent. I don't know of a good historical analog for the successful transition from benevolent dictator to anarcho-consensus. If anyone out there does know of one it could be *very* fruitful to share with the right people. Something to keep in the back of your mind. Where my bitcoin anthropologists at?
My gut tells me that we will continue to thrash on the soft fork question with no meaningful progress until the scaling pains or some other problem grows to crisis level. This is a very clear historical precedent. After the crisis is widely acknowledged we'll select a new hierarchy of leaders to push through an engineering fix.
This will probably bring about another schism in the bitcoin community. I suspect we'll need to fork off Saylor and Blackrock. Honestly, good riddance. These fat cats were never cypherpunks and they never will be. But with them goes the promise of unlimited price appreciation with no effort. You'll still need to work a day job to grow your savings.
I think the pleb ossifiers are directionally correct, but I think their specific concerns are misplaced. They are right about proceeding cautiously. We're only gonna get one chance to displace fiat. We can't afford to fuck this up. But I think they take this anxiety towards change and put it all on the risks they can see and understand and take direct action on: soft fork activation attempts from within the bitcoin community. It's a whole lot easier to stall something than to push it through and acting in any capacity makes you feel accomplished, even if you are pulling in the wrong direction.
In my opinion, premature ossification is a much bigger problem. The risk is that bitcoin achieves far less than it's full potential. We get there through inaction. There is a massive pot in the center of this poker table. We need to ante up to stay in the game. We need to take measured, intelligent risks to win. But it's not a poker game, we're actually betting on the course of human civilization.
Personally, I think attempting any soft fork under the current level of mining pool centralization is an unacceptable risk. One reason I decided to work on this problem.
Another reason is that mining software has been long neglected. Historically, the smartest and most ideologically motivated devs are attracted to other layers of the stack. I know because I feel this pull as well. This is why we've got free, private, decentralized, self-hostable, cypherpunk software for nodes and wallets, but not for mining.
Things are turning around thanks to the hard work of a small group of engineers. Sv2 is complete and SRI is nearly production ready. I am extremely grateful to be in a position to stand on the shoulders of these giants. But I don't see a compelling story for small miners to run this software and build their own block templates. It's not enough to get hobbyists on board; in order to win this fight we need to make decentralized mining more profitable than centralized alternatives.
Braidpool is another very promising project. I hope they succeed. A decentralized monolithic mining pool would be metal as fuck but I fear that even with Braidpool we're still stuck in the local maximum of monolithic mining pools.
I smell opportunity. I see a path to fix big problems in bitcoin by focusing on the neglected layers of the stack. I think we can decentralize mining with new ideas and architectures that bring unstoppable cypherpunk energy to the space. If successful, the fruits of this labor will unleash a tsunami of innovation and human productivity. I'm ready to go surfing. 🏄
My secret hope is that solving a big existential risk that everyone can see and understand might be the psychological victory we need to head off a greater conflict in the future. If we get most of the pleb ossifiers on board, we won't need to fork off the fat cats. With overwhelming consensus they will see the shift in the weather and get with the program to protect their vast wealth. Sun Tzu said to always leave a path for your opponent to retreat. I think this wisdom is directly applicable here.
We can show the ossifiers that they were right to be cautious, vanquish an existential threat to bitcoin, and, if we act skillfully and with emotional intelligence, we can redirect their attention to the next looming crisis: scaling trustless self custody to every human being.
I don't know how to do this. It's a social influence game, which is an area I have always intentionally avoided. This is a nascent idea. I'm still turning it over in my head. Maybe some of y'all have a good perspective to share or can take this idea and run with it. Let me know what you think.
Discussion
That isn't a core problem. The market will determine it.
I don't disagree with you. I'm merely informing others that alternative implementations would be useful. The market is not efficient in complex fields.
More node implementations would be nice but it does nothing to scale on-chain usage.
I think a wider range of implementations would open up avenues for getting changes added to consensus layer. If others become the majority over core, and those ones adopt feature flagging, consensus might change to support. It removes Core review as the ultimate gatekeeper to changes.
> those ones adopt feature flagging, consensus might change to support
I think you put the cart before the horse. You can't adopt features requiring consensus changes before you get everyone to adopt the consensus change. It doesn't matter how many node implementations there are.
That's not accurate. Even today many people run nodes supporting different features. Even before the softforks, nodes upgraded to support the features if the softfork might be activated. The problem right now is that Core reviewers are the gatekeepers for consensus implementation.
For a somewhat related example, take Bitcoin Knots. It is essentially a filter that still respects the most proof-of-work chain whatever that might be. The point of it is that if a majority of people adopted it, miners would be forced to update their consensus rules for fear of not having their blocks propagated.
Let's say a worthy alternative implementation gained majority popularity. Would consensus proposal review and reference implementation remain on Core?
How many consensus changes have you coded up? Are you waiting for bitcoin core reviewers to approve them?
Are you admitting that Core review is the crux, or are you suggesting my opinion is invalid if I haven't proposed a change to Core?
I heard there was intent to separate consensus rules from Core node implementation, maybe via a library? This would at least be a step in the right direction.
Just clarifying your level of knowledge of the process. You speak like you have experience. I assume based on your answer that you don't have any first hand knowledge.
I have not proposed a consensus change but I pay close attention to this stuff. Core review is not the blocker. No consensus change has achieved enough momentum to warrant core maintainer code review.
Maybe you assumed my use of the word "gatekeeper" meant that I was suggesting Core has been actively blocking proposals. As James OB pointed out, there is no direction. Or maybe I've misread or forgotten the point of his article.
Yeah that's how I read it. Apologies if I misunderstood you.
Regarding soft forks, the biggest blocker is that the community is fractured. No consensus change will get through until we achieve unity or near unity of opinion on both the necessity and end-goal of a soft fork. I am coming to the opinion that the only way to get there is from a crisis. This is just how anarchy works.
It is unclear to me if we can ever go back to the era of experts pushing through non-essential upgrades. That door may have closed forever. Only time will tell.
So maybe the best thing we can do is keep iterating on proposals that address a specific problem and have them ready to go (or as close as we can manage) in the event that a crisis develops.
I see two developing crises:
- quantum computing
- scaling on-chain throughput
These are not independent concerns. Currently, the best quantum signatures are 100x larger than elliptic curve signatures. So "solving" this crisis will make our scaling problems exponentially worse.
There are a few early stage proposals for quantum resistant signature algorithms but the vast majority of the work remains to be done.
As for scaling, the existing covenant proposals haven't been shown to increase on-chain throughput, which is necessary for self-custody. The layered approach taken by lightning, ecash, ark, et al scales off-chain transaction throughput but at the end of the day users need to withdraw their sats on-chain to cold storage. This will be the bottleneck.
Don't misunderstand me. I'm not saying scaling transaction throughput is not important. If we don't improve bitcoin's usability then we won't have scaling problems because it will stay niche forever.
But when it comes to consensus changes, self-custody is the thing we need to scale. This is how people protect themselves from debasement. If we don't address this problem then what are we even doing here? This was a great point raised by James OB and it has changed my thinking.
I will be applying this new filter when I consider soft fork proposals in the future: does it allow significantly more people to own an on-chain UTXO?
I've made some assumptions in this line of reasoning. It's always good to question your assumptions. There are the open questions in my mind:
Do existing covenants proposals increase on-chain throughput? I wasn't very rigorous in claiming that they don't. It is a loosely held belief.
Can bitcoiners protect themselves from debasement without owning a UTXO? Is a federated model like Liquid or Fedimint good enough? Is a trustless bridge to another chain possible? If it is possible, does it do anything to reduce fees?