Nic Carter should make a BIPNic Carter should make a BIP
Nic Carter's been having another meltdown because one of companies he invests in needs a marketing boost, and when Bitcoiners tell him he should be the change he wants to see in the world, this is his response:
Jonas Schnelli had this response to Carter:
Schnelli's comment sounds reasonable to me. I think Carter is very wrong that the BIP process is not open to wide collaboration. I suspect if he championed an interesting quantum resistance BIP it would indeed be included in the BIP repo and Bitcoiners would largely give it a fair shake. Carter's current behavior of hyping a imminent quantum doom feels to me a lot like Sam Altman doing the talk show circuit saying "AI is going to kill us all, by the way I'm raising more money to build AI."
But perhaps I'm wrong about this. Casey Rodarmor's ordinals BIP was recently rejected (#1455710) and
Paul Sztorc thinks the BIP process is a fraudPaul Sztorc thinks the BIP process is a fraud
This conversation triggered Paul Sztorc, who wrote a fiery post response about how horrible the BIP process (and Core in general) are. Specifically, in response to the idea that Nic Carter should submit a quantum resistance BIP, Sztorc said:
And -- when @nic_carter's [hypothetical] PR goes nowhere ... then? he does what ? Release an alternative client? And ask pools to run it? Or? Release a hard fork of Bitcoin? Or a new altcoin, with a new name?? Ridiculous! I can't believe people are so stupid, in 2026.
I believe many of these steps have been taken by @dathon_ohm in the pursuit of the Reduce Data Temporary Soft Fork (RTDS) and there have even been a few blocks signalling for BIP 110.
Should Core merge dathon's BIP 110 implementation as an option for Core node runners?Should Core merge dathon's BIP 110 implementation as an option for Core node runners?
In an update to the BIP 110 activation client, Dathon noted that they submitted a couple of pull requests to Core in order to have a BIP 110 implementation merged into Core's codebase, in order "to augment Core with the ability to enforce the new rules." (#1463243)
As an aside, Dathon uses a particular communication style which is highly optimistic and confidently assumes correctness in all its statements (I find it off-putting because it reminds me of the tone of Big Brother in 1984).
These pull requests were auto-closed. Resulting in some complaints on X from Mechanic and Dathon.
Which goes back to Sztorc's point: what should one do when one's proposal to improve Bitcoin is not accepted into the github.com/bitcoin repo?
In Dathon's case, BIP 110 was included in the github.com/bitcoin/bips repo, but the suggested changes have not (so far) been implemented in Core's actual software.
It seems pretty clear to me that if I submitted a PR to the Knots repo that proposed a change that was at odds with what they wanted to do, they would be under no obligation to implement it. So what's different about Core?
I asked Luke Dashjr this question, he said the situation was not remotely comparable:
Luke's first two points are not relevant here. I could have proposed a consensus change to increase block size and the point would still stand: should their be an expectation that a project will include changes it thinks are bad?
Really, I think it comes down to the last point: does Core have to play by different rules because the majority of the network runs it? I suspect that if Knots becomes the dominant node implementation, Luke Dashjr is unlikely to merge changes he disagrees with just because a lot of people want it.
Permissionless means "do your own thing" not "you must do my thing"Permissionless means "do your own thing" not "you must do my thing"
So, my advice to Nic Carter and Casey Rodarmor and Dathon Ohm is that they should forget trying to merge things in github.com/bitcoin - and I suspect each of them actually knows this.
In each case, they've already built the thing they wanted to see in the world: Nic Carter has invested in Project11 and Casey Rodarmor released ordinals years ago -- indeed, that's one of the reasons Dathon felt they needed to go and make BIP 110. I'll admit that I get the feeling that each of these "outcasts" finds rather more value from being rejected by Core than they would from being included. It is perhaps the only way anyone would pay them any attention.
Can you change Bitcoin without a BIP?Can you change Bitcoin without a BIP?
Bitcoin thrives when anyone can propose new ideas or changes to how Bitcoin works. I think the BIPs system helps to coordinate some of these ideas and makes it easier for Bitcoiners to get on the same standard when we think a good one is proposed.
However, I wonder if Bitcoiners need to remind themselves that you can change Bitcoin without a BIP.
Change what? Something because Carter needs to make money off the backs of Bitcoiners? Can't he just go back to being a shitcoiner?
Carter is doing what he seems to enjoy doing, no doubt. But the larger point of what a person's expectations about inclusion in the /bitcoin/bips repo should be still seems worth clarifying.
It is impossible to make everybody happy -- either the repo has opinions about what should and should not be included or it's wide open. (seems like wide open couldn't work unless it had some ranking system and then we get into who decides the ranking system...maybe in some world money could be the moderator there...)
What I'm saying is would people like Carter and Rodarmor and Dathon feel less need to attempt advertising via rejection from the BIPs repo if it was widely known that the repo makes no attempt at being "impartial" or "fair"?
I suspect that we'd all be happier if the specific /bitcoin/bips repo took a more overt stance that the maintainers are expected to have opinions about what makes a good or a bad BIP (beyond just formatting).
I think that your thought from #1455710 is close to what I think consensus looks like. You said:
But the
qualifierin your statement is important: new ideasin Bitcoin. So I do think that there is a filter needed for inclusion.For example I disagree with kruw's suggestion that ordinals should be a BIP. It has nothing to do with Bitcoin as it is a metaprotocol. Counterparty has/had their own standards repo, which is exactly what ordinals should do, imho. It may have a dependency on Bitcoin, but Bitcoin doesn't have a dependency on it.
If Carter wants to write a BIP about a new cryptographic primitive, I think that that's fine. I think that that should fall within the inclusion filter. It is directly about Bitcoin. But since his attention seeking and whining is putting me off, I'd not be surprised if it puts everyone else off too. So it may get stuck in limbo.
As for Bitcoin Core inclusion, that's a whole other story. BIP-39 is a good example of something not included in Core... and there are many others. A BIP existing doesn't mean it must get implemented. That also goes for BIP-110, though personally I'd like Bitcoin Core to have awareness of externally implemented signals. Right now I'm iterating over
getblockrpc to distill the signal.I think "the maintainers are expected to have an opinion about desirability" is the opposite of what Murch has been saying; which I understood as: the maintainers / editors of that repo are to have opinions about the process, not so much the suitability for implementation.
I probably shouldn't have included Dathon's complaint about his PR not getting included in Core, but I did because it has a similar feel to Rodarmor and Carter's complaints about their BIPs.
I have heard the concept that "maintainers / editors of the bips repo are to have opinions about the process, not so much the suitability" but I think that such an approach will always lead to people using the rejection of their favorite BIP to get attention.
Maybe that isn't a bad thing. I wasn't necessarily what kind of opinion BIP maintainers should have. But I think it might be helpful if Bitcoiners had an understanding that a BIP was anything formatted along the lines suggested by BIP 3, regardless of where it was hosted.
The existence of a numbering system may be at the root of this. Since numbers are a finite set, we need someone to assign numbers and make sure there aren't duplicates. What is the advantage we gain by having BIPs in a central repo?
I'd guess it's much easier coordination and standardization. People can point to one single place and say: this is the standard for this idea. But I am curious how strong a benefit this actually is.
Could a system without a central repository for BIPs work? Perhaps each person who proposes a BIP could maintain the repository for their own BIP. Inclusion of a BIP then becomes more of a decision made by any given project. And to solve numbering, we could just refer to BIPs by name.
Hmm I wouldn't follow BIP3 per se if I were to make another repo. The process can be different for different repositories. It shouldn't really be about the people; if it is then it is ripe for disruption. I still think achow prevented BIP-repo disruption when calling for editor nominations. I think that that was a great move and I like the outcome.
Ease of discoverability, reference, prevention of link rot, harmonized process, ease of getting eyes... there are definite benefits to centralization here. But they are not trumping everything and there imho must be a way around it: this repo must never be a binding bureaucracy, nor should delving or the mailing list. It must be there on its own merits. If it sucks or is abused, use something else.
Yes. It could be a decentralized set of specs. I think that this was proposed for
NIPs?What's important, imho:
I've co-developed some open standards and protocols that are in production use globally today. The most useful team size for small protocol enhancements was 3-4 people, and 6-8 for large greenfield work. We'd work about 3 to 18 months on a spec, and then 1-3 months or so to get to a digestible publication. The important part isn't publication, though: it's what you do before that. ↩
As usual, I have lots to learn here. Ease of getting eyes is something that I certainly wasn't considering. In a way, I'd say that Ordinals is an evidence that there is very much a way around the BIPs system: apparently lots of people figured out how to do ordinals things without there ever being a BIP.
In my various disputes with Dathon Ohm, one thing that has frustrated me has been their use of the word "official." I find it very frustrating that Bitcoiners talk about anything in Bitcoin being "officially supported" or "officially implemented" or whatever.
Having a single BIPs repo feels a lot like an "official" BIPs repo. I believe this is what nettles me about it a little.
The response I want to give to folks like Carter or Dathon is that there is: what are you waiting for? Do you expect others to do things for you? Put your proposal out there and convince others to use it.
I suppose they would say this is exactly what they are doing.
I appreciated this by Voskuil:
source
What Carter and Dathon are doing feels like politicking. I don't want Bitcoin to be political money.
I'm allergic to this word. There's no such thing in a consensus network. That's the entire point of consensus by running code: nothing official, only choice. No pre-sorting either because the outcome is per definition up in the air. If you can manipulate it beyond your worth, the system is broken. [1]
I 100% agree with this sentiment. Especially since, when one carries real arguments, one gets ignored. Your comment? Didn't exist. I speak from recent experience. Hah.
However, per the note below, I am observing similar politicking elsewhere too. This is important to at least be wary of. It's not just 2 individuals that are vocal on X. There are lots of them, people with actual influence too. We have to keep calling it out, especially when people we want to like do it.
I don't like arguing with maintainers or their close colleagues. I know the shit job they've got. But, I will keep doing it. I'll even try to do it respectfully, which is arguably the hardest part for me because I've grown to like the savage way. But I try.
I have to pace myself in Voskuil admiration sometimes, haha. He often says things I agree with though. Including this quote.
Note: the system may actually be a bit broken. This is the one point I will concede to BCH shitcoiners. Not that their conspiracy theories are correct - they're unlikely to be, in my (informed-ish) opinion - but, there are powerhouses and eventually bitcoiners will need to challenge these if they don't self-destruct. That's also what I like about what Atack is saying - though in that case I again don't like much of the rationale. Can't stay stuck in the Covid trauma. Need to move on. Do shit. Not whine. ↩
https://twiiit.com/evoskuil/status/2038316722462036473
I was writing this while answering my six year old's very insistent questions about his lunch. Sorry for the incoherence.
See? When you don't write a BIP, you have to deal with having to explain everything. 😂
I should create a LIPs repo (Lunch Improvement Proposal) and have the kids submit proposals about what they'd like to eat.
Of course, I think I'll make them implement the proposals because I sure as heck ain't making Peking Duck for lunch.
My best friend gives her kids 10 options for 7 days. They may order them and discard 3. She's also sneaky and always makes sure there are more than a few healthy options - more than can be discarded.
Wasn't there a Kaplan book about leading without being in charge? I remember reading that.
he could also just learn to STFU. Kind of done with this dude... had his chance in the spotlight, now he's on a deluded rampage. BYYYEEE
I feel like I'm wading into an area for which I'm very much an outsider / layman. But I did read the Rodarmor's post about his BIP getting shut down, and even while disagreeing with Ordinals, I don't think a unilateral shutdown with no reasoning given is ever a good idea. Perhaps sufficient reasoning was given in the discussion comments, which I didn't fully read, or elsewhere, but I think it's still important to offer a summary of the reasons in the final message, even if it's something like "We don't think this is consistent with our philosophy of what Bitcoin should be."
I can understand the frustration of the Core maintainers who have to do a pretty thankless job while under constant attack and presumption. But letting that get under your skin and bleed into impatience or curt interactions with outsiders, or a temptation to exert authority without discussion, is a recipe for loss of trust. In a way, that's kinda what happened with the "health experts" during COVID, because they got so impatient with debate that they just decided to shut down all dissent. That was a recipe for lost of trust, and I think in COVID it was also a recipe for becoming misguided and choosing a wrong direction.
Now, I don't know if Core is choosing a right or a wrong direction, but I think it's important to maintain a posture of patience and rationality even in the face of apparently very irrational and overconfident critics like Dathon and Luke.
To me, it's not really a question of the choices Core, specifically, has made. What I'm curious about is whether anyone can actually do a good job at such a task.
There will always be disagreements about what Bitcoin is and how it should change. Many BIPs (like ordinals) are not soft forks -- and in such cases, it seems pretty clear that people should just do the thing they want to do. BIPs are a nice coordination tool, but there's no reason it needs to be hosted in the /bitcoin repo.
For soft or hard forks, it is more difficult. I think it's unreasonable to expect Luke Dashjr to implement a change he disagrees with and I think it's unreasonable to expect Core contributors to implement changes they disagree with. I'm pretty general in the camp that we shouldn't tell other people what to do.
To each his own would be fine...if there wasn't an expectation that one implementation was the "reference implementation." So, what's the solution? Perhaps it is more implementations. Or perhaps it is something more like what Knots and BIP 110 supporters are attempting now: fork Core and add your changes and then try to get everybody to run it. (I don't like their style, nor their substance, but I think the general methodology might be reasonable).
an interesting question is whether there's an equilibrium with multiple widely used implementations. I think there can be, and I reference Linux as an example. many different implementations, all are expected to work in some basic ways, but different emphases and features for each one
The thing everyone points to though is that Bitcoin is tricky because the expectation that it works "in some basic ways" is highly particular: all implementations that don't want to fork off have to work in exactly the same way across a pretty large set of rules and behaviors.
I agree that there probably is an equilibrium that can be achieved. And in that case, I wonder whether each implementation will house it's own BIP repo or whether there will be one implementation independent BIP repo or something else.
Inscriptions and ordinals kinda did. One might argue that they are all still valid transactions, but we can all see on chain there was a change in TX behavior.
I really love these thought provoking and insightful posts. You're great at aggregating and curating content to inform the reader and allow everyone to form their own opinion on the matter. You really inspire me @Scoresby !!
BIP's are the equivalent of IT telling you to create a ticket to get you to go away
It's an appeal to authority in a way too, design by committee... have others dilute your idea rather than just building the thing with conviction. Submitting a BIP is submission to the perceived authority of that repo and therefore its maintainers.
Imagine if everyone that wrote a BIP that went nowhere instead just had a fork of Core, how many forks of Core might have some traction as changes compounded and got interest?
Bad example perhaps because this costs more lost coins than anything, but seed phrases are a BIP not adopted by Core... people just built it.
An idea for a change might be really good, good enough that people try and run it... but if that same idea gets swept under the rug in the BIP repo it will discourage people from running it since they see it as a filter.
The NIP's repo in Nostr catalyzed this for me, prNIP-69 swept under the rug for arbitrary reasons of vanity, CLINK SDK however now ascendant.
Do you know of any systems like BIPs, NIPs, PIPs, etc where they don't use a central repo to collect them, but rather follow a format (like that proposed by BIP3) but rely on whoever proposed a specific idea to host the repo for that idea alone?
So: a BIP is something that follows this format, not a BIP is something that is in this repo.
Well the P in this things is "proposal", so it's inherently trying to convince others, rather than just documenting something you built for yourself that others can adopt, or not
The opposite might be something like http smtp or ftp not being a proposal to tcp/ip
You can try to convince people of something whether or not the something is located in one single place or in many places.
The real test of not having a single BIP repo would be some kind of soft fork BIP, as I imagine that would be very tricky to coordinate without a single source.
But I don't think it is true that single source for each BIP has to equal same source for a BIPs.
You can convince people just by doing a thing and they see it works... Showing vs telling
Proposals skew towards telling. The BIP repo, correctly or incorrectly, is equated with Core as the default distribution
If telling the default distribution fails, thats perceived as a filter ... Committee building... which kills potential early or legitimizes bad things.
These types of things exists because people are over socialized, deferential, lack confidence, etc, the natural course of weak men creating hard times.
What if the NSA decided not to build Bitcoin after the negative responses to the paper on the mailing lists?
https://twiiit.com/nic_carter/status/2039131684411437330
The real question buried in this drama isn't whether "go write a BIP" is good advice. It's whether the BIP process itself has become a political process wearing technical clothing.
Schnelli says "if it is shut down, it's purely technical." But Carter's point about BIP360 is hard to ignore. If a technically sound proposal gets ignored because the wrong person champions it, that IS political gatekeeping. You can call it "lack of consensus" but that's just politics with extra steps.
Here's what I think gets missed though. The BIP process was designed for a world where 20 people cared about Bitcoin protocol development. Now thousands do. The bottleneck isn't malice from those 5 gatekeepers. It's that the process doesn't scale. It was never built to handle this volume of proposals, this many stakeholders, or this much money riding on outcomes.
Sztorc is right that "go write a BIP" as a dismissal is absurd if the process itself is broken. But the fix isn't abandoning BIPs. The fix is making the process transparent enough that technical merit can be evaluated independently of who's championing it.
The Knots comparison in the post is actually the most interesting part. If submitting a PR to Knots carries no obligation for them to implement, why would submitting a BIP to Core be any different? The answer people don't want to hear: because Core IS Bitcoin in practice, even if it shouldn't be in theory.