Welcome back to the Bit by Bit podcast.
In today's episode, I speak with Jameson Lopp.
Jameson created and manages Lopp.net,
one of the best resources online for people who are interested in Bitcoin.
Jameson is the CASA co-founder and CTO.
CASA's aim is to make security on Bitcoin more accessible to the non-technical.
CASA provides Bitcoin inheritance, multi-sig security, and 24-hour support.
Bitcoin can be complicated. CASA is there to make it easy for you.
At the time of recording, the world's gone mad.
People are on lockdown, money printers around the world are going brrrr,
and new laws are being passed to take away your rights.
I speak today with Jameson about what I think is the most important technology
and our last chance for freedom.
That technology, obviously, is Bitcoin.
So let's jump straight into it.
Jameson, like I said before, thank you very much for taking the call.
And yeah, if I can start with finding out what is it that led you to Bitcoin?
How did you find it and how has it changed things for you?
I'll probably never remember the specifics.
I know I heard about it a few times on various geek websites
and dismissed it just as many times until eventually some article came along.
I'm pretty sure it was on Slashdot, but that piqued my interest enough
that I read the white paper and once I did that, my computer science interest
really got kicked off.
I realized that it was solving an interesting problem
that I'd never really thought about before.
And then as I continued to think more about that from other perspectives
like the philosophical and libertarian side of things,
just looking at how money operates because I never thought about that concept before,
it made sense to me that money should be this open project
where anyone who wants to can collaborate and try to make it a better thing.
So just the entire concept of money as an open source project
was what I also found to be interesting just from the standpoint of,
well, maybe that will end up being more robust, more fair,
or at the very least less likely to be wielded as a tool for certain special interests.
So I got interested from a variety of different perspectives
and started playing around with it, created some side projects,
and then a few years later decided that I might as well go full-time.
So when was it switched in your mind? Did you very quickly think,
oh, I just want to work in this space?
Or was it a gradual, you got sucked down the rabbit hole
and at one point you just thought, no, I have to be just focusing on this?
No, it took several years as I learned more and more about the operation.
And let's see, I forked the Bitcoin Core repository in 2014 to create Statoshi,
which was essentially Bitcoin Core plus a bunch of instrumentation and metrics
because I was just trying to better understand it,
trying to bring more transparency to how the nodes themselves were operating,
trying to help other developers understand and optimize various aspects of node operation.
And I actually applied for a grant from the Bitcoin Foundation,
which I never heard back from them on.
And it wasn't until after I had been working in the industry for many years
that I actually heard that I had won that grant internally,
but then the Bitcoin Foundation fell apart and ended up not paying it out.
So I would have ended up working probably more on Bitcoin stuff even earlier
if they had paid out that grant for me to work more on that project.
But it all worked out in early 2015, a lot of venture capital was starting to come into the space.
I saw more job openings and that's when I decided I was just going to start applying to any position
that I thought looked interesting.
So you've done a few things before CASA, now that you are your CTO of CASA now.
What is it that drove you to the way that the business is set up?
It seems to me from reading through lock.net, some of the things you've written on Twitter,
basically security comes first in your mind.
It seems that that is number one.
Yeah, and that's because of the nature of Bitcoin.
It is very easy to shoot yourself in the foot.
But it's very easy to permanently lose either through an attacker or more likely your own negligence
and really lock yourself out of your own wealth.
And there's a million different things that can go wrong.
It's a very tricky maze of obstacles to navigate.
And that's why if you talk to anyone who's been in the space for a number of years,
they most likely have a story of something that they screwed up that resulted in them losing money.
And so this is something obviously I learned a lot more about once I started going full time
and working at BitGo doing enterprise, multi-sig, Bitcoin security.
And even though that was a very well architected setup, there were still things that could go wrong,
usually the result of bad decisions by users.
I also personally realized that I was spending an inordinate amount of time,
basically a weekend every year, refreshing my own cold storage setup.
And that was very convoluted and I had my own inheritance that I set up,
which I was not completely confident would actually get executed correctly if I got hit by a truck.
And so I just realized that if it's this complicated for me and I'm supposed to be one of the experts,
then for most people, they're not going to go anywhere near the links that I went to.
And I suspect that that would result in one of two things.
Either they're going to have a more fragile setup where they're more likely to lose their private keys,
or they're just going to throw up their hands and leave all their money on the exchange or at some other custodian.
And of course, that's against the ethos of what we're building here.
It also is creating more systemic risk in the system if a lot of the money is held by a small number of entities.
I have to say, I've only been involved for a few years now.
And for me, coming into someone non-technical, it's really scary doing this stuff.
Every time I'd send a transaction or be thinking about how I do my cold storage
and thinking through what's the safest way I can do it and where do I put my seed phrase and all these things.
And definitely for normal people, you're right.
I think it puts a lot of people off and scares a lot of people.
Yeah, I mean, an easy way to actually show the complexity is we have about 40 pages.
That's 40 pages of documentation on our website. That's on docs.keys.casa,
which goes through all of our own decisions and threat modeling when we were building our product.
And if someone reads all of that and internalizes it all, anyone is more than welcome to build their own solution.
I think that one of the reasons we're able to provide a lot of value
is simply due to the amount of man hours and thought that we've put into this
and all the experience that we have seen of various mistakes people have made over the years.
And really the idea is to build software where, yes, it is security first,
but it's also usability as a component of security.
Because if the software itself is not guiding the user to do things the secure way,
to follow the best practices, then even if you have a product that is basically impossible
for an external attacker to penetrate and get those keys,
you're probably still leaving a number of different possible branches down various decision paths
that end up in the user shooting themselves in the foot.
So I always hear that's sort of what happens more than anything.
It's not the wrench attack. It is more just people making a stupid mistake and losing their coins.
So I guess that's your mission to stop that happening.
There's just so many possible things that could go wrong.
It's kind of like insurance, right?
Almost nobody is going to spend a ton of time thinking about everything that could go wrong in their life.
Rather, you buy an insurance policy that covers all types of different things.
And this is the same type of mindset that I guess we're kind of approaching as a business of
we're helping people avoid all of that overhead of thinking through and planning out every possible contingency.
And we're kind of wrapping up all of that planning into one package.
And as long as you just follow the instructions on the screen, you'll be good to go.
And you have this, it's like a key rotation thing. Is that right?
You sort of have keys across a certain amount of multi-sig basically, but making it more user friendly.
Because that's something I keep thinking about. I still haven't gone to multi-sig.
And I just get comfortable with something. And then once I'm comfortable, done enough research, I change.
But from what I understand, you're sort of making that process a lot easier for people to go into multi-sig.
Right. And this is another one of those things where if you start trying to explain multi-sig
and the benefits of this protocol feature to most people, their eyes are just going to glaze over and they'll get really bored.
At a much higher level though, what we're doing is we're systematically eliminating any single point of failure.
And that's another thing that we cover a lot in the documentation that I was talking about,
where if there's any single point along the line where a catastrophe happens, it doesn't really matter what.
It doesn't matter if we're talking like a physical flood or fire versus an actual thief coming in
or even your maid accidentally throwing away your seed phrase or something.
If there's anywhere in your system where a single action, whether by you, someone else or an act of nature or just bad luck,
if that can cascade into a complete catastrophe for your whole system, then you've got a major problem.
And so by using multi-sig, by requiring multiple different keys in order to sign off on a transaction,
by using the hardware devices that are already available on the market that are made by specialists,
we're a software company, not really a hardware company, and by geographically distributing these hardware devices,
we can create an extremely robust solution that is resistant to all types of different loss
and also provides a sufficient level of resiliency where even if the user screws something up,
even if you lose or destroy one of your devices or even two of your devices, we have a level of resilience in the model
so that humans can still make mistakes and still recover from them.
And so you mentioned the key rotation stuff.
One of the other novel things that we've done, I don't think anyone else has at this point,
is we actually got rid of the need for you to store seed phrases.
And that was because, once again, it just made the security model so complex,
and it was very difficult to reason through how a user might store their seed phrases
and basically trying to enforce best practices around all that.
And so once you get into a multi-sig setup, it's actually possible for you to not have to keep track of that
because we built in this key rotation functionality where if you lose a device
or otherwise have a device become inaccessible, then you just go into the software and tap on, you know,
replace device, and we walk you through a very simple wizard of go buy a new device, plug it in,
and we then help you rotate out that particular device so that you once again reconstitute your wallet back to the full secure state.
That's cool. I think the security, like I said before, is one of the biggest things when I speak to people about it,
and they just go, ah, it's too complicated.
But when we're talking about, well, the way you just described it, it sounds pretty simple.
I can imagine most people can get on board and understand that, and it's got a nice flow and it's, you know,
you feel like if you make a mistake, it's not all gone. And I think that's the people that scares them.
And now with the way, with the last couple of weeks and the current uncertainty around the world,
it sounds like more and more people, it feels to me like more and more people might start waking up and going,
do you know what, it'd be quite nice to have money that I control and no one else can fuck with.
And, you know, it's going to become, I feel, more obvious.
So having these services for people jumping in, it just stops, like you said, people having it on Coinbase or some exchange, which is not what we want.
So, yeah, the Black Swan events, I think make it more clear that you do need to be prepared for edge cases. And it's kind of a weird situation of
trying to convince someone that, you know, they basically need to buy insurance for events that are unlikely to happen.
But if they do happen, could be catastrophic. And that is why I think, you know, when these events start happening,
at least some people will start to open their eyes. But I still suspect that probably the vast majority of people
will go through this event and still, you know, not be prepared for similar types of things.
Because they'll probably have survivor bias of like, oh, I made it through fine with the other one.
Anything else that happens, I'll also be fine.
Yeah, it's been really weird, especially like all this stuff that we chat about on Bitcoin, Twitter,
and I speak to people who are interested in this stuff. You know, we talk about the scenarios that are happening around the world literally today.
We've been talking about them for ages. And now it's happening and it just, it feels really weird.
It's sort of like everyone's been saying, oh, it's coming. But this meltdown is just crazy.
And then how people don't wake up to it just, I find it amazing. I find it amazing that people don't see the value
or most people don't see the value in having something that is completely controlled by you.
And no one can mess with, you know, I've never come across anything like that before.
Even property and there's nothing you really own because the government has their hand in it.
And this, it just seems, it seems...
It's kind of a new asset, right? It's like you said, and I and many other people have said,
if you're incapable of actually defending any given asset against attackers, do you really own it?
And once we take assets out of the physical world and bring them into the digital realm,
that's the really cool thing about cryptography is that it now gives the average person
an inordinate level of defensibility at very low cost if you can use it correctly.
And so that's what we're trying to do is trying to leverage the hardware and the software that's already out there.
The best practices that people have learned the hard way over the past decade of using Bitcoin
and wrap it up in a very user friendly package so that people can attain the level of self sovereignty
that Bitcoin has been promising all along, but has had a really, really high learning curve to actually get there.
So if you talk through for anyone who's, I mean, most people who listen to this show already interested in Bitcoin,
for people who are newer to it, if you talk through like say today,
someone goes and they buy themselves a couple of thousand pounds worth of Bitcoin and they then say,
OK, I want to actually take control of this and I want to make sure that this is mine
and no matter what happens, if we have an apocalypse, whatever happens, I control this. What do they do?
If they were to come say to CASA, what does that flow look like?
They've got their coins on an exchange. What do they do? What would you suggest?
Yeah. So, you know, the first step is to actually take control of your private keys.
And so that means setting up a wallet that is generating private keys that only you know.
And, you know, there's a million different ways you can do that.
You can set up wallets on desktops or mobile devices where it's just the software keeping track of that
and storing the private keys on a hard drive.
That's only really recommended for small amounts of money because, you know,
keeping those keys on a full fledged operating system that could have a ton of different possible security issues.
The next step beyond that is moving your private keys to a hardware device
like a Trezor, Ledger, Coldcard, and so on, where now those keys are no longer on an internet connected computer.
They only exist on this hardware that never actually has internet access.
You know, you can do that for, you know, $50 to $100 basically depending on which device that you buy.
Sorry to interrupt, do you have a favorite?
I always had Ledger, then I went to Trezor, and now I'm looking at Coldcards.
Do you have a preferred hardware wallet?
Well, this is an interesting question because they all have their own pros and cons.
And actually we covered those in our most recent blog post where we announced Coldcard support.
We were talking about why we believe Coldcard is a great complement to Trezor and Ledger as part of a full package.
But, you know, the good thing about Trezor is that it's fully open source.
One of the downsides is that there are some known physical exploits where a sophisticated physical attacker could extract the keys off the device.
Ledger, on the other hand, does not have any known physical vulnerabilities.
However, it's not fully open source. But then Coldcard is actually both.
It's fully open source hardware and software and there are known physical vulnerabilities.
However, one downside I would say with Coldcard, because of how new it is and the fact that it's using PSPT,
which is not yet supported by a lot of other wallets, I would say Coldcard if you're using that as a single device,
that's still more of an advanced user type of setup.
It also gives you really cool air gap signing flexibility, which Trezor and Ledger don't support, to my knowledge.
I think you could get a Ledger Nano X, which has Bluetooth only, but then you have to go down that whole rabbit hole.
So, you know, suffice to say all of these devices are designed differently.
They have pros and cons, but the really cool thing from a security standpoint is when you start using them in combination
as part of a multi-seg setup, the security properties are actually additive rather than subtractive.
So, while any one given device could potentially be attacked in a certain way,
you know that all of the devices are not going to be attacked in the same exact way because they don't have the same vulnerabilities.
You're also getting rid of things like supply chain attacks simply because they're coming from different manufacturers.
And so, there's one of the many reasons why having a more diverse setup and distributed key set, we believe, offers the absolute highest level of security.
That makes sense. And so, for people, if they're to set things up that way, I mean, it's been described to me before
that if you were to try and break the encryption or trying to hack or get one of these private keys,
that basically it would be something like you'd have to choose two grains of sand on the planet
and someone would have to guess which ones they were out of all of them, some sort of crazy number like that.
Does having it put onto multi-seg mean that it's even more complicated and even less likely?
I'd have to do some math, you know, but I think suffice to say even if it is, you know, from a cryptographic standpoint,
even harder to brute force because multi-seg is a more complex script.
You know, we're already talking so astronomically high from a computing standpoint that, you know,
you're basically talking about like needing quantum computing to even make it feasible and even then Bitcoin scripts
are quantum computing resistant if you haven't spent from them, which is another reason why address reuse is a bad idea,
not just from a privacy standpoint, but also potentially quantum computing standpoint.
But really like when we're thinking about the most likely types of attacks and losses, you know,
brute forcing is already pretty much out of the question.
The next level down is just like general hacking.
And so you get rid of that by using dedicated hardware devices.
And then, of course, there's all the different loss vectors of just something goes wrong and the device gets lost or destroyed.
And at the end of the day, you know, if you get a really well designed setup,
then the only thing that you should really be having to worry about is some sort of phishing attack
or even, you know, malware that is like making you think that you're sending your transaction to one address
and it's actually going to another.
And the way that that's another good reason why we use different dedicated hardware devices is that
even if you somehow got malware on your phone when you were initializing a spend out of your CASA wallet,
you then have to go and confirm the value and the recipient addresses on multiple diverse hardware and software.
So that even if something happens to your device or to our software, none of those are single points of failure.
Yeah, that makes sense.
The other thing that I had a question with multi-sig.
Say, for example, you were trying to control, you know, maybe you'd bought a little bit of Bitcoin from one exchange
and then some was bought in cash and then some is sent from another wallet
and you didn't want to have those coins all in one wallet because then it's more trackable.
If you were to have, say, a multi-sig setup, is there a way of setting it up
so you could have one multi-sig setup, but you could have, say, 20 different addresses
and there's no way that those touch each other? Does that make sense?
Yeah, yeah. So this is basically a sub-accounts feature and I don't know how deep you want to go into protocol level stuff,
but basically like the way that a wallet creates the addresses is a standard for address derivation
and you just have these different numerical paths and you can increment the paths.
It's part of the standard where if you increment one part of the path,
that's considered to be like a sub-account under the main account.
So it's certainly doable and it's a lot easier, I think, to do it that way as opposed to the other way,
which is you have one main account and then you're still depositing everything in there,
but then you're having to manually control each individual UTXO. That gets really messy.
It's like coin control basically.
Yeah, so coin control. I know Bitcoin Core and Electrum both offer coin control.
At CASA, we do have coin control at the address level, not at the UTXO level
because it basically ends up being the same thing from a privacy standpoint,
but still it's annoying to deal with and I don't recommend it.
It's definitely easier to use these sub-accounts.
Okay, and so if you make these sub-accounts, then those coins don't effectively touch each other
and there's no way of anyone saying, okay, well, this person has...
We can tell that this person has these two addresses and that these coins are from here
and these coins are from there and they can link it. It's kept separately.
Yeah, and I mean, that's the tricky thing about on-chain privacy in general,
is if at any point in the future, those funds get commingled and you send both of them somewhere
or essentially you end up spending those UTXOs at the same time
or I mean, there's a million other heuristics that companies like Chain Analysis use
to try to come up with at least a percentage probability
that certain funds are owned by the same entity.
Privacy in general is very difficult, but on-chain privacy even more so
because this is a permanent history and one single mistake can potentially link
any number of different funds together.
And it's there forever, right? That's the problem.
Yeah, they're dirty bastards, these Chain Analysis. They drive me crazy.
It's so frustrating because that's always the one thing with me.
I just almost get scared to do anything because I think, well, I don't understand these systems
well enough to know if I'm doing something wrong.
And then you think, oh, well, I've made one mistake, like you said, and it's there forever
and all the things you've tried to do to keep your privacy, then you've lost it from one mistake.
Do you have high hopes for things like Whirlpool and Lightning Network
and all these sort of things eventually overcoming these problems?
I think second layer solutions will have much better chance of offering stronger privacy
simply because you're no longer broadcasting what you're doing to the entire world.
The main problem with mixing in its current form is that even if you do manage to mix your funds
with a sufficiently large other set of participants, and I would argue that currently
they're probably not sufficiently large enough, not enough people are doing it,
you have this problem where it's highly obvious.
You're creating, once again, heuristics around these coin joins are so obvious
that we've already seen several different services start to ban users that were mixing their coins.
And it's because the service doesn't want to take the risk of them potentially accepting funds
that are used for some activity that they can't track.
That's ridiculous because I don't think that there's anything even similarly analogous to that
in traditional finance. If you deposit a bunch of funds at a bank,
they can't trace where it came from more than one hop, for example.
But on the blockchain, you can trace it all the way back to wherever.
You get into this really weird situation of quote unquote taint analysis.
And then the question is like, how many hops does the money have to go
before you're no longer culpable for whatever it may be?
Yeah, it's such a weird, I don't know.
To me, it's just crazy that it's even something that should be considered.
To me, I think, well, it's digital cash.
And you don't do it with cash. If I go down to the shop and I buy a paper,
they give me the change back and then I can't get stopped and said,
oh, where did you get that change and where was it five hops before?
No one cares. It doesn't make any sense.
At least in America, we do have the unfortunate civil asset forfeiture stuff
where cops can basically say, oh, you're carrying too much money on you.
And we're going to assume that that's drug money because why else would you have
10 or $20,000 in cash on you unless you have illegal activities?
Yeah. Yeah, it's crazy. It's absolutely crazy.
Do you hold out high hopes that this stuff, obviously for someone like yourself
and people who understand these systems well, I'm sure that these things are doable.
But for the normal people who just want some privacy, do you think that
in the next, say, five years, we're going to see large improvements
or is it all things that I've had a lot of talk about that needs to be changes
on the protocol level to make it so that, say, coin joins aren't so obvious?
And then is it Adam Gibson who's doing this pay join or something?
Yeah. Yeah. Yeah. So Adam Gibson is one of the maintainers of join market.
He's also did a present. He's done a number of presentations on pay join
on the VR thing, wasn't it? Yeah. Yeah. Yeah. And that's awesome.
Yeah. I think the most interesting thing in terms of what we traditionally
consider to be a coin join, because I would say pay join is kind of a different
kind of coin join than we normally think of.
You know, normally we think of large group of people coming together
and just mixing all their money together. I think the most exciting thing
on that front is going to be signature aggregation that will reduce
the fingerprint on chain of these coin joins that are happening.
Okay. So they will look the same as any other transaction basically?
That's definitely the goal. Will it happen in the next five years?
I think so. I mean, we should be getting Schnorr signatures
hopefully in the next year or two. I mean, the code is already written.
The spec is up. All that really needs to be figured out is deployment.
And of course, that tends to sometimes end up being the most fun
controversial part, though I'm not aware of any strong arguments
against signature aggregation. What you end up having more controversy over
when we're talking privacy is super strong privacy that's more experimental,
you know, stuff like ZK Snarks, where you're actually completely hiding
the values of the transaction and where it's going.
And the main issue with that is that it bumps up against another fundamental
property of Bitcoin, which is the auditability of the total money supply.
And so while privacy is, I think, very highly ranked in terms of desirability,
I think that auditability of the total money supply will always end up
trumping privacy. Yeah, that makes sense.
And say it's obviously this would have been thought of, but I just thought I'd ask.
Say you had like a way of changing, you know, how you have like liquid
and you sort of you have Bitcoin, which then is turned into a liquid token
or I don't know how you describe it exactly, but then that's basically one for one.
Would there not be a way of doing a similar sort of thing with Bitcoin
and having like a set up where you have, I don't know, ring signatures or something
you had like it almost transformed into a Monero-ish type coin
that was just the same one Bitcoin one.
Well, yeah, I mean, you use that. So, you know, you can create a side chain.
The main problem with side chains that I see is that no one has figured out
and deployed a trustless pegging mechanism.
There is the drive chain concept from Paul Stork that I think is interesting,
but also seems controversial mainly due to the power that it gives to miners
to potentially misbehave.
Is this the thing it takes six months it's got to be locked up for?
Yes, it's something like that and it's complicated.
It's been a while since I've looked at it.
Unfortunately, it seems to be fairly controversial though.
I do think that they will eventually launch a side chain or a drive chain of some sort.
As far as I know, they're still working on it.
But, you know, even liquid, you know, you don't need to create a new side chain.
You can use liquid if you peg into liquid or you can do an atomic swap basically into liquid,
at least theoretically.
I know I think some have been done, but I don't think it's extremely user-friendly at this stage.
Then on liquid, they do have the confidential transactions enabled,
which obscures the addresses in the transaction.
I think the values are still in the clear, but the addresses are not.
I've actually, I think I've kind of brought this up a few times before and never really got an answer.
I'm not aware of any reason why we couldn't set up mixing pools on liquid
that are a major improvement in comparison to mixing on Bitcoin.
I'm not aware of anyone who's actually gone down the path of trying to write a CoinJoin software
that does confidential transactions or, you know, if there's any hidden gotchas there
of why you wouldn't be able to do that.
It's also, you know, in general, perhaps what will end up making more sense than doing on-chain CoinJoin
is we'll be doing maybe, you know, a series of atomic swaps and you can do those on-chain
or you could do them with Lightning.
The whole idea there being that, you know, you break the linkability by simply going to a different blockchain.
So there is no real link.
Right. Yeah, I see.
But, you know, I think those are just still mostly theoretical concepts where no one has actually tried to build something.
You know, there are projects that are working on atomic swaps, but as far as I'm aware,
like there's not any extremely user-friendly ones, but I also haven't really had time to play with many of them.
I know we had an interesting presentation at MIT Bitcoin Expo from Arwen and they've figured out
they've made an improvement upon the traditional atomic swap operations,
which basically leverages some like Ethereum smart contracting to store these pre-images for hashes.
So you're essentially doing something that's similar to, you know,
a hash-timelock contract except you're basically operating it across two different blockchains.
Hmm. Well, that's interesting.
I guess, like you say, there's all these sort of things will happen on second layers and it just takes time.
I guess it feels like it takes time when you're interested in this stuff and you're just chomping at the bit.
But yeah, these things take time and it will come.
But do you have some sort of plans within CASA?
You have the nodes and am I right in saying you're not supplying or selling those anymore?
Right. And actually tomorrow, so maybe even before this gets published,
we're going to be publishing a how to build your own basically do-it-yourself CASA node guide
since we have stopped manufacturing and shipping them.
But, you know, we still want anyone who's interested to be able to use that on their own.
And it's something that we will probably revisit at some time in the future.
Just for now, we're going to be focusing on our original mission with helping people, you know, secure their wealth.
It's, I think, a much higher value proposition,
whereas the node stuff is much more cutting edge and it's interesting,
but people aren't putting their life savings into, you know, a lightning node.
No, and it's more important that you have everything safe.
I guess that's step one, isn't it?
And, you know, making sure that you feel comfortable and things secure.
That's interesting you're going to actually do that because I'm ordering a Raspberry Pi and a few bits and pieces.
And I was looking at my node where I was really leading with this question was
if we have technologies like lightning and the usability gets better and better,
can you imagine that given time you do have a sort of a node at home and everything just runs through there
and you can store your data and you can have your, you know, lightning on there
and you can have everything sort of, I don't know how I'm phrasing this question,
but basically that you have everything controlled by yourself
and that given time we can have less and less of our data sort of sat in silos and controlled by other third parties.
Is that realistic? Is it sort of a dream to say what we take back control of more and more of our data?
Well, yeah, I mean, this is actually a cycle.
This has happened before and it may happen again.
And, you know, in the very early days of computing, all the computing actually occurred on mainframes
and you would have a lightweight terminal that you would connect to that mainframe
and request, you know, your various data and computing operations from.
And then we had the advent of the personal computer where, you know, we grew by orders of magnitude in computing power
so that you could actually afford to have an immense level of computing power in a machine that's, you know, in your home.
And so at that point, all the data, all the processing went mostly away from the mainframes
and became more decentralized in people's homes.
And then in the 2000s era, you know, as the internet started happening, the pendulum started swinging the other way
where we would still have our personal computing devices, but a lot of the data was getting sucked up and stored
and computed mostly for marketing purposes by the quote unquote cloud.
And now, even today, you know, cloud computing is able to operate at such a cost effective scale
that it's in many cases, it's cheaper for you just to buy a subscription to like a Dropbox or a Google Drive
than it is for you to actually buy several hard drives that you then manage yourself and, you know, create redundancy for.
And so it's quite possible that we see the pendulum start to swing back yet again,
but I don't think that it will happen just in the sense that we go from doing cloud for everything
back to doing everything on a device at home.
Rather, I think what needs to happen is I think the cloud needs to migrate to these, you know, many computer at home devices,
by which I mean, you know, the reason that all of these enterprise cloud computing stuff has become so popular
is because it abstracts away all of the boring IT administration stuff, you know,
just dealing with redundancy and hardware failure and other natural disasters.
You know, real cloud computing companies mirror entire infrastructure across different continents at that level.
And like, you know, almost no one is going to be able to have the resources to go to that extreme.
And so it makes sense because you're buying a lot of security against those type of losses.
Of course, you're losing a lot of security in other ways by handing all of your data over to a third party.
Now, if we can get the best of both worlds, if we can start to build decentralized clouds
where your little mini node at home is actually participating in a peer-to-peer decentralized storage network
for all types of different things, that's how I think we might be able to get the best of both worlds.
And, you know, we learned a lot about that at Caso, shipping out all these nodes over the past year or two
in that, you know, there's still a lot of just IT administration headaches of deploying devices into unknown networks.
You know, people had all kinds of crazy network setups with different ISPs.
And, you know, you need to be able to abstract a lot of that away to really make a plug-and-play device
that also has all of the resiliency of cloud computing.
Hmm. That would be so cool if that's doable.
When I was looking at the MyNode stuff, I was looking and it had sort of, you know, a little thing with VPN on there
and then it had Tor and it had all the other bits and then it had your wallet and it had Lightning.
And I was just looking at that dashboard, you'd call it.
And I was looking at that and I was just thinking, how cool would this be if you had your photographs or your documents
and, you know, everything that you'd have on a computer.
And like you were describing, you have it as a system where it is some sort of decentralized cloud computer
and you control everything on there, but you can still use it from wherever you are.
Because I think that's the thing. It's like you don't walk around with hard drives in your backpack or whatever.
You want to be able to jump onto a phone and still be able to use these things.
But if that was possible, I can't see why people wouldn't switch across to it.
Because I think people in their gut feel things are wrong.
I think maybe they don't know, you know, with things like Facebook and everyone having your data.
And I think people feel uncomfortable. They just don't know that there could be another way.
And it's become so easy now with, like you say, that they've made it so easy to store your data with them
that it kind of has to be as easy or close to as easy for them to switch.
Yeah, I mean, it really comes down to convenience.
And I think part of the problem here is if we're talking about building open systems,
then often convenience gets left by the wayside.
I mean, usually these open source systems are designed by engineers
who are going to be prioritizing things like security and scalability.
But there's usually a big gap on the usability and convenience side.
And so that will be an interesting question to see how that progresses.
I mean, you can kind of look at similar things with Linux, for example.
I mean, I've been a Linux user for 20 years.
And the experience has immensely improved, especially over the past 10 years or so.
And you could probably have an entire show or thesis about why that's the case.
But I can only hope that we'll see similar things with this particular set of industry.
Maybe it just takes time, but it feels like that's the way things would naturally slide.
I feel like enough people are sick to death the way that things are.
And if enough people want it, presumably it will eventually come along.
Well, there's some sort of crowd theory around all of that,
that you only really need maybe 5% to 10% of people to go through a higher friction experience
and be early adopters into technology that is sufficiently superior.
So I think that the vast majority of people simply don't have the time to think about these things.
I'm sure they would care if they had time to sit down and think through them.
But people are worried about other things.
Horrors and stock market crashes.
People like us have the privilege that we're in a position that we can think about
some of these more esoteric and theoretical edge case black swan type of problems.
I think the average person is just worried about their next paycheck and insurance and food on the table
and just living their lives.
Yeah, for sure. That's definitely true.
And then like you say, maybe if the 10% go through that friction, then it changes for them as well.
That would be nice.
So have you got anything that you're particularly excited about?
Any developments within Bitcoin or Lightning that are particularly exciting to you at the moment?
The biggest thing that I think is going to happen in the next year or so is Taproot, Mass, Schnorr.
That's going to significantly expand, I think, the usability of Bitcoin scripts,
both from a scalability perspective and a privacy perspective.
We're going to see much more complex Bitcoin-based applications coming out,
though the keynote that I just gave the other week at the MIT Bitcoin Expo
is kind of a warning to the developers who are going to be working on that.
There's going to be a lot of increased complexity under the hood,
and we need to think about how we build these systems so that they are protected against disaster scenarios
and without going too deep into the weeds of the technical stuff.
Basically, it comes back to what we were talking about earlier, getting rid of seed phrases,
because even just giving somebody 24 words and telling them to keep it safe
is just the tip of an iceberg of a whole mountain of IT and security knowledge
that I think most people are going to need to avoid
if they want to actually be able to operate safely and securely in these systems.
We need to continue to work on usability so that the average person can feel confident
in taking control of their digital assets.
The flip side of this equation and the reason why it is a double-edged sword,
it gives you power, but it also gives you the responsibility to wield that power,
because if you use it incorrectly, if you make a mistake, there is no one else with the power to save you.
Yeah, sorry for your loss.
This is what I don't particularly understand. With Schnorr, from my basic understanding,
Schnorr would replace ECDSA, is that correct?
If it was to replace that, is it just a security that it helps with or does it have other impacts?
Does it change other things, switching them?
Yeah, I believe it's actually like a different set of public keys and, of course, the signatures.
It's not a new method for cryptographic signing,
but it was under, I think, a patent for 20 years or so.
Is it going to change things?
I think one thing I haven't actually even looked into is will existing hardware devices, for example,
be able to easily support this?
We do know that we're going to be able to roll it out in a backwards compatible fashion,
so it's not like people are going to have to start using it.
That'll be similar to the way SegWit was rolled out.
It can be very tricky to have a wallet that you're maintaining that then has a major change like that.
Basically, each wallet developer is going to have to figure out how they want to do their own sort of migration
of different UTXOs, different scripts that they're creating.
At the really low cryptographic level, even I don't really go down that low
into understanding the actual constructions of the cryptography for a lot of things.
Does it change anything more than security then?
Or does it give you more ability to do more things changing to Schnorr?
Or is that just one part of many changes that are looking to Taproot and Schnorr and all of these?
They get bundled into one software update or whatever you would describe it as.
I think one of the reasons why Schnorr was preferred is that it does have provable security.
This is more of a theoretical thing and it's beyond my understanding.
I'm not the only one.
It also has some other properties to it that I think it's some of the mathematical properties
under the hood that make it possible for you to do the signature aggregation.
This is basically a summing of a group of public keys.
Obviously, the protocol developers can tend to explain it better than I can,
but this property allows you to have a group of public keys that essentially sum to a certain number
and you know that no other possible set of public keys could have that same property.
The linear nature of the math then allows you to have the signatures basically be aggregated
so that you don't have to show every single signature to someone to prove that it's been signed,
whether you can show the sum of all the signatures and the one who's viewing it can have that same assurance.
There's some mathematical proof around that that gives you that assurance.
That makes sense because I hear a lot of these things that they're trying to get them done
and that people are arguing over certain things.
I hear these Schnorr and Taproot and things and a lot of this just goes well over my head.
It's interesting to know why some of these things are going ahead,
but that does sound like a very worthwhile thing to do.
These are the things that more experienced protocol developers spend years looking into
and you're trying to save a few bytes here and there,
just trying to keep the cost of operating the system as low as possible
while continuing to expand the functionality for end-users like us.
Well, I'm very grateful that they do.
I'm very grateful.
Jameson, like I said, I don't want to take too much of your time.
I know it's late here.
What time are you there?
Are you eight hours behind us?
Meraka.
I wanted to just finish up on asking you, obviously from reading through lock.net
and hearing you on other podcasts,
it seems that above all really you care about freedom and privacy.
I just wanted to hear from you.
Do you feel confident and what's your outlook over the next, say, 10, 15 years?
Do you feel that Bitcoin gives us some hope to turn things around?
It feels to me like we're on a tipping point.
Sometimes I feel positive and sometimes I feel negative about how things are going to play out.
My fear is 1984, George Orwell.
Sometimes it feels like that's coming.
But you obviously spend all day every day fighting a good fight with this stuff.
How do you feel it's going to play out?
I'm confident not for any particular technical reason,
but rather because of the original reason why I got into it all,
which is creating an open collaborative project
where anyone who wants to can help contribute and coordinate.
The properties of a project that is architected in such a way mean
that you can essentially create this Hydra
so that there is no single point of failure in the project.
There are certainly plenty of large actors and participants
that play various roles in the system,
but it's sufficiently spread out and sufficiently large enough at this point,
I think, that it's even resistant to nation-state attacks.
The way that I tend to put it is that Bitcoin can't die until we all agree that it's dead
and then we all agree to give up trying to work on it and improve it.
Just from that philosophical standpoint,
that's why I think that we've already gone past the point of no return.
It's very hard for me to envision a scenario
where everyone around the entire world decides to give up on it.
We've even seen nation-states like India and China crack down on it a lot.
Sure, there's a loss of confidence,
but never to the point that the people like myself,
who have basically dedicated our career to working on it,
and many people who have volunteered a lot of their free time
and their skills to working on it,
I think it would create a disaster or require a disaster of proportions
that I can't even really imagine to sufficiently disillusion all of those people.
It does feel like that, and I'm not technical or of any use really,
but it's something I will never give up on as long as it's possible
because I believe that this is our best chance of freedom.
I think it's such an amazing opportunity, such an amazing system.
I think most people who are involved in this,
it seems that they all care so deeply.
I think you're right. It's for people to give up.
I don't know what it would take.
Hopefully, we'll never find out.
Hopefully not.
Well, yeah, like I say, thank you so much for joining me.
James really enjoyed chatting,
and it's been really useful to go through some of these things
that go well above my head.
I'm hoping to get one of these VR headsets soon,
so maybe we'll meet in virtual reality at one of these meetups
or something like that.
Great. I'll see you in cyberspace.
All right. Fantastic. Yeah, thanks again.
I'm going to link lop.net and CASA in the show notes.
Anything else you want to finish on
or anything you want me to add in?
No, all good.
Okay, fantastic. All right.
Thanks a lot. Hopefully, we'll speak in cyberspace.
Thanks. Bye. Cheers.