inamerrata

I’ve been trying to come up with a good way of thinking about what to prioritise in Bitcoin work for a little while now — there’s so much interesting stuff going around, all of it Good For Bitcoin, that you need some way to figure out which bits are more important or urgent than others. One way to think about it is “what will we make the price go up?”, another is “how do we beat all the altcoins?”, but both of those seem a bit limited in scope. Maybe an alternative is to think about it backwards: if Bitcoin gets better, more people will want to be Bitcoiners; so what would it take to make more people Bitcoiners? That sort of question is a pretty common one in sales/marketing, and they tend to use “sales funnels” for analysing it — before becoming a customer, people have to hear about a product, be interested in it, and find it for sale somewhere, and you get some attrition at each step; reducing the attrition at any step (without making it worse at any other) then increases your sales and your numbers go up.

One way of looking at that might be to consider the normal sorts of things Bitcoiners do: they buy some Bitcoin, setup their own wallet to have control over their funds, run a full node, and maybe eventually start giving some input into Bitcoin’s development (whether that be in the form of code, discussion, investment or making bets over twitter). The problem with thinking about things that way is that while there are some clear incentives for the first steps (Bitcoin’s increasing in value so a good investment or at least better than earning negative rates; self-custody reduces the risk of some company running off with all the coins you thought were yours), there’s a breakdown after that: having a hardware wallet under your mattress is cheap and easy, but running a full node constantly is an ongoing cost and maintenance burden, and what’s the actual direct benefit to you? If you look at the numbers, those steps are something like 8B to 160M (2%) to 4M (2.5%) to 50k (1.25%) to maybe 900 (1.8%), but there’s no obvious levers to use to increase either the 2.5% or 1.25% figures, so that approach doesn’t seem that useful.

A different way of looking at it might be to first break out people who regularly transact with their Bitcoin balance, rather than just buying and holding. The idea being that this covers traders who actively manage their Bitcoin investment, merchants who sell products for Bitcoin, people who get paid in Bitcoin, and so on. I’ve got no idea what a valid number for this is — BitPay claims to be “Trusted by thousands of businesses — worldwide” which makes it sound like the number probably isn’t in the millions, so I’ve picked a quarter of a million. Going from “actively transacting” to “self-custody” is a different step than self-custody for “buying-and-holding” — don’t think of installing a mobile wallet or buying a hardware wallet, but rather as using software like btcpay or lightning rather than hosted solutions like bitpay or travelbybit. I’ve picked 15k as the number there, based on the number of lightning nodes reported by 1ml.com, and rounded up a bit.

The nice thing about that approach is that the incentives at each stage are a fair bit clearer. You maintain a Bitcoin balance if it works as a store of value and fits into your investment strategy. You go from just holding a Bitcoin balance to actively transacting with it if spending Bitcoin is less of a pain than spending from your bank account — which makes it pretty clear why that step has a 99.85% attrition rate and what to do about it. Likewise, you go from transacting in general to self-custody when you decide that the costs of using a Bitcoin bank outweigh the benefits — risk of loss of funds or censorship, KYC frustrations, privacy concerns versus ease of setup and someone else taking care of ongoing maintenance. Having that option is hopefully a good incentive for businesses (and regulators) to keep those risks, frustrations and concerns relatively rare for everyone that doesn’t self-custody as well. Going from actively using Bitcoin to helping it develop is still a big step, but it’s also a fairly natural one (or so it seems to me). I think those levels also fit fairly well with business models: getting people into Bitcoin in the first place is financial education/advice and exchange services; actively transacting is banking and merchant services; self-custody is hardware wallets, and things like btcpay and lightning nodes; even consensus participation has been monetized by the likes of bitfinex’s chain-split tokens. (A nice thing about this approach is that self-custody for people actively transacting, generally implies running a node for technical reasons, and at that point the costs of running a node are a much smaller deal: you’re getting regular benefits from your regular transactions, so the small regular costs of running a full node are much easier to justify)

One way to view those levels might be as “pre-coiners”, “store-of-value”, “method-of-payment”, “self-sovereign” and “decentralised” — with each level implicitly depending on the previous levels. You can’t pay for things with money that nobody values; there’s no point being in control of money that no one will accept or that’s not worth anything; there’s not point having decentralised money if it can be stolen from you, etc. There’s some circularity too though: there’s no point storing value if you can’t eventually transfer it, and a significant part of the value proposition of Bitcoin for store of value or method of payment is that you can control your own funds and that there isn’t a central group able to inflate the money supply, confiscate funds or block transactions.

What does that mean for priorities? I think there’s a few general principles you can draw from the above:

• From an industry-growth point-of-view, increasing the percentages for the top two levels and maintaining the percentages for the bottom two seems like a good focus: getting a billion people owning Bitcoin, and hundred of millions transacting using it, even with “only” 12M (6% of 200M) people running their own full nodes (due to self-hosting their lightning balance), and 750k (6% of 12M) people actively paying attention to how Bitcoin works and evolves seems like it could work out.
• This approach has “store of value” as a foundation that the other properties of Bitcoin rely on — if that makes sense, it probably means messing with the “store of value” features of Bitcoin is a really risky idea. Instead, it’s probably more important to work on things that reinforce the existing foundations, than neat new ideas to change them.
• The “having Bitcoin” to “transacting with Bitcoin” step is the one that needs the most work — probably in a million areas: not just all the things on the todo list for lightning, but UX stuff, and working with regulators to avoid knee-jerk money-laundering concerns, or with tax agencies to reduce the reporting burden due to Bitcoin valuation changes, to deploying point-of-sale systems, and whatever else.
• If we do manage to get lots more people holding Bitcoin, and/or lots more people transacting with it, then maintaining the percentages of people doing self-custody or contributing in general will be hard, and require a lot of effort.

So for me (with an open source developer’s perspective), I think that adds up to:

• Number one priority is keeping Bitcoin working technically — trying to avoid bugs, resist potential attacks (both ones we already know about, and those people have yet to come up with), stay backwards compatible, do clean upgrades. Things to work on here include monitoring, tests, code analysis, code reviews, etc. This also means keeping development of bitcoin itself relatively slow, since all these things take time and effort.
• Number two priority is, I think, lightning: it seems the best approach for payments, both for people who want to do self-custody, and as the underlying payments mechanism for Bitcoin custodians to use when their customers instruct them to make a payment. There’s a lot of work to be done there: routing, reliability, spam/attack-resistance, privacy, wallet integration, etc. Other payments related things (like btcpay) are also probably pretty high impact.
• After that, I think being prepared for growth is the next thing: finding ways of doing things more efficiently (eg, batching, consolidation), coping dynamically with changes to the system (eg, fee estimation), developing standards to make it easy to interoperate with new entrants to the ecosystem (eg, psbt, miniscript), and having good explanations of how Bitcoin works and why it works that way to newcomers (podcasts, books, academic papers, etc).

And more particularly, I think that means that I want to prioritise stability over new features (so work on analysis and reviews and tests and no rushing the taproot soft-fork), and as far as new features go, I’m more interested in ones that can provide boosts to lightning or payments in general (so taproot and ANYPREVOUT stay high on my list), but growth and interoperability are still important (so I don’t have to ignore cool things like CTV fortunately).