Atack speaking tourAtack speaking tour
This is another interesting interview with Jon Atack. He gave a short speech in January at PlanB in El Salvador expressing concerns with the culture and organizational dynamics in Bitcoin Core (#1447597) and then was on Bob Burnett's show (#1456592) discussing the same topic in more detail.
This interview by Knut Svanholm elaborates on many of the points Atack brought up in PlanB. In many ways it has more to think about than Bob Burnett's interview, despite Svanholm's more emotional approach.
The interview is definitely worth listening to. Here are a few highlights:The interview is definitely worth listening to. Here are a few highlights:
Grant-funded Bitcoin development
One of the most interesting things that is covered in the interview is how grant-funding in Bitcoin development works. It is clear that Atack has not found Brink and Chaincode's style very forthright, at times feeling like those organizations were interested in activist funding that intended to "corporatize" Bitcoin Core. On the other hand, Atack speaks positively about his experience with grant-makers like OpenSats and HRF.
Chaincode leadership
It is clear that Atack did not see eye-to-eye with leadership at Chaincode Labs. Some of this may be explained as Atack and Chaincode leadership having very different beliefs about what makes a good project. Atack supports a loose organization, with geographic and
Atack discusses in detail his development experience in Bitcoin as well as experiences he has had with specific pull requests or comments. The case he presents is that over the last five or six years, the organization of Bitcoin Core has moved away from solely making decisions based on the merit of arguments toward making development decisions based on the identity or relationships of the developers involved.
Changes to Bitcoin Core organization culture
Atack notes that previously, most of the development discussion in Bitcoin Core took place during weekly public irc meetingsbut that now most of the important conversations take place in private working groups or in lunch meetings.
Atack presents a case that Bitcoin Core is very centralized around two US offices at the moment and that this is a negative thing for Core because it reduces the number and kind of people who are able to contribute to the project.
ConclusionConclusion
I am not a developer and I've never contributed to an open source project, so there is some level at which it is fair to say that none of these internal politics are any of my business. But I don't think Atack is bringing them up in order to gossip or make drama.
Human groups will always have internal politics. I don't see why open source development or Bitcoin development should be any different. However, if we have a single project that is by far the dominant implementation of Bitcoin, the side effects of such internal politics are hard to escape.
Competition is healthy. It is unlikely that new or other implementations of Bitcoin will be free of the internal politics Atack identifies in these interviews -- although they may be concerned with different values. There is a hope, however, that if there are a number of healthy and actively developed projects, such projects would be improved by the pressure from their competitors.
But it's not like we can just sit around and expect other people to do it for us. It means putting time and effort into educating yourself about how Bitcoin works and what other implementations are doing.
So, if you aren't running a node, give it a shot.
If you have been running one kind of node, consider experimenting with another kind.
If you are capable, consider contributing reviews to some of the other node implementations.
I'm glad Atack is getting out there more. The circumstances are a bummer.
If we are committed to devs being funded by nonprofits, which it seems like no one is yet willing to question, we need to figure out how to fight the tradeoffs that nonprofits introduce.
For example: I suspect BIP 110 has gained more steam than it deserves because Luke, even if his aim is wrong, is running a business and his incentives are obvious and relatable to everyone. Luke ultimately butters his bread by doing things many people want.[1] What are the incentives of folks surviving off of grants? By all appearances, and best case, they butter their bread by doing things some tiny federation wants. With near-zero respectable bitcoin companies entering the ring to fight Luke that aren't themselves in some way tied up in a nonprofit, it looks like the aristocrats against the people. The optics are terrible even if the reality is not (and the reality might be terrible too idk).
Bitcoin protocol developers deserve to get paid for what they do and the market would fail to pay them without nonprofits, but: money is dumb by itself, dumb money is poisonous, and there's a lot of money flowing through bitcoin nonprofits lately. If we add that nonprofits, absent saints managing them, tend toward politicization, corruption, and status games - and that nonprofits are generally weapons of influence - we might be curing one disease with a variety of poison.
Most people respond to this with well achtually he has VC money. I wish it were easy to communicate how this differs from a grant (or how it differs from having a job or how a grant even differs from having a job). The best I can do: (1) running a business is one of the hardest possible things for people to do (if you disagree with that, why aren't you doing it?). (2) If Luke fails to do what many people want, he's taken years off his life only to pay an enormous opportunity cost, and he will have failed to reciprocate the generosity of customers/VCs and stand in a kind of social/moral debt for the rest of his life. ↩
I'm not sure it's the right solution, but I've heard it suggested that folks build a redhat-like bitcoin implementation. It adds incentive diversity in addition to implementation diversity.
To be clear, I'm not against nonprofits generically, but with all the qualified ideological folks getting grants, who is left to build for-profit bitcoin-involved ventures? Shitcoiners and masochists afaict.
Why do you think a bounty system never really materialized? Bitcoin is internet money, it seems like such an obvious way to do things. But all the bounty systems I remember never got much attention.
Bounties work best when the work is well scoped and easy to verify. A lot of work, like code review, isn't well scoped nor easy to verify. Also The Nature of the Firm.
I didn't spend more than a day cooking SN's repo bounty system, but I spent long enough to conclude it's at least very non-trivial to create a system that produces results comparable to a tenured team.
the non profits may be dumb but bitcoin holders aren't dumb.
eapecially with consensus, if nonprofits steer bitcoin in a direction that fails to maximize value, forks ensue and the shitfork dies a shameful deah to become a pitiful altcoin.
under the hood it's pure economic darwinism, no matter how the developers are paid.
I agree that bitcoin holders are smart, and we have these nice check and balances. Whether we are on a fork or not though, the solutions to developer funding are the same, and unless we figure that out, history will keep repeating itself.
The tension between grant-funded development and core decentralization is the ultimate stress test for Bitcoin's social layer. Jon Atack’s point on the shift from public IRC meetings to private groups is a red flag we can't ignore.
If the development culture becomes a silo, we risk the very 'corporatization' we're fighting against. Open source isn't just about the code; it's about the transparency of the decision-making process. Signal > Noise.