Welcome to The Blockchain VC, a podcast about crypto and the digital assets ecosystem.
My name is Tomer Federman, and I'm the managing partner at Federman Capital.
We invest in the most promising blockchain startups across the globe.
I have more than 15 years of experience in tech, and before starting the fund, I was
on the product side at Facebook, where I led product strategy and global growth of some
of Facebook's major ad products.
Previously, I also lived in Silicon Valley for a few years, where I attended Stanford
Business School.
You can find me on Twitter at Tomer Federman.
Before we begin, please note that this podcast is for informational purposes only, and all
the opinions expressed on this show, either by guests or me, do not reflect the opinions
of Federman Capital.
Nothing on the Blockchain VC podcast represents an investment or financial advice.
Please do your own research.
Also, if you liked this episode of The Blockchain VC and want to help us bring more awareness
to the space, I'd really appreciate it if you can rate, review, and subscribe to our
podcast.
This only takes a few seconds and helps us get the word out.
Okay, let's do this.
So really excited to welcome to the show today, Jameson Lopp, the CTO of CASA.
Jameson, thanks a lot for coming on the show.
No problem.
Glad to be here.
So maybe to get started, Jameson, you're one of the, I think, most well-known people in
the space.
We'd love to hear more about your background and how you became interested in crypto and
blockchain to begin with.
So I got involved in Bitcoin a number of years ago.
I was just interested in it mainly from a computer science standpoint.
I heard about it a number of times over the years and was always dismissing it because
I figured it was some system that was going to get hacked and everybody was going to lose
their money.
But after it kept coming back over and over, I finally realized that it wasn't dying off
and that I should probably look into it.
So once I actually read the white paper and I realized that it was doing some interesting
things on the computer science side, it also piqued my interest from a sort of philosophical
viewpoint because I figured that if money is this abstract concept that kind of belongs
to humanity, then it makes sense that the definition and development of money ought
to be an open project where anyone who cares can contribute to it.
Right.
And was it immediately obvious to you when you started looking into it that this thing
could be so transformative?
I definitely thought it was an interesting development.
Once I read the white paper, it wasn't obvious that it was going to work incredibly well,
which is why I went ahead and actually got some Bitcoin so I could actually play around
with it and see it work.
It was once I actually started using it and was able to exchange it for other goods and
services and realized that this was a new type of payment system that could not really
be censored.
That's when I really started to see the value in it.
Got it.
Makes sense.
And when was that?
When did you start?
When did you read the white paper and became more interested in this paper?
That was back in 2012 and I spent a few years just as sort of a hobbyist talking about it
and reading the forums and discussing with people.
And then after I had been interested in it for three years or so, that's when I decided
that I wanted to go ahead and work in the space full time.
A bunch of venture capital had been coming into the space and that's when I started
working for BitGo doing basically infrastructure engineering for their enterprise multi-sig
security wallet.
And I had been doing some other projects just on the side, open source stuff, trying to
better understand how Bitcoin worked.
But that's when I went sort of head first and was really living and breathing Bitcoin
24 seven ever since then.
One thing I was curious about is, you know, obviously with BitGo and more recently now
with Casa seems like you're pretty focused on solving custody and key management and
how people protect their crypto assets and Bitcoin in particular.
Why is that?
Like is that something that you think is particularly interesting in the space?
Obviously there's a lot going on.
Why did you decide to focus on that segment of the market?
You know, it's not necessarily the most interesting thing going on, but I think that it is a fundamental
issue that is still not fully solved.
Basically I think that the easy solution where people could just sort of throw up their hands
and give up is to just give your private keys to a trusted third party and allow them to
custody your funds, but I find that to be very unpalatable because then we're just recreating
the banking system, you know, with a new layer of infrastructure underneath it.
So the promise that Bitcoin allows you to be your own bank, in my opinion, is not yet
fully fulfilled because while it is technically possible to be your own bank and it always
has been, the level of like sophistication and the high learning curve required in order
to do that well has prevented a lot of people from doing it or a lot of people who have
done it have ended up shooting themselves in the foot and losing access to their private
keys.
So I think that this is a fundamental unsolved problem in Bitcoin and that, you know, we're
never going to realize the dream of a fully distributed financial system if it's not easy
for the average person to hold their own funds securely.
Yeah, absolutely.
I mean, I've actually been hearing more and more recently, I think a lot of people in
the space think that, in a way, custody is a solved problem at this point, right?
Because you have all these companies, you know, whether it's Coinbase or Fidelity are
now entering the space or Bitgo, you know, Kingdom Trust and so forth, who are offering
a regulated custody solution, oftentimes now also with an insurance plan.
But I completely hear you, what you're saying is that's not really why we got into this
space to begin with.
Yeah, I mean, it's, to me, it's incredibly boring and a cop out and it recreates a lot
of the systemic risk where, sure, you can make arguments that, you know, highly respected
and secure companies with lots of experts are now custodying private keys for people.
But if you're concentrating the holdings of large portions of the system in a small
number of actors, it doesn't really matter how secure they are against your average threats.
You're still creating honeypots for either sophisticated attackers or for, if you want
to go to the extreme, nation state attackers.
And, you know, in this system, you know, it was kind of born and developed out of almost
a level of paranoia and adversarial thinking.
But we do want to build a system that is robust against any conceivable attack, even if it's
highly unlikely.
Yeah, yeah.
And is that also what prompted you to more recently make the move from BitGo to CASA?
Yeah, so the three years that I spent at BitGo, I was working on non-custodial multi-sig solutions
where BitGo would have one out of three keys, but we were never actually fully custodying
anyone's funds.
As I was leaving, BitGo was starting to spool up a custody solution, which they're now also
offering, which is, you know, a completely different system, actually.
But I switched over to CASA.
It was a very small pivot for me on the technical side because it's also a multi-sig, non-custodial
type of solution.
The main difference is that BitGo is specifically targeting enterprise customers, and at CASA,
we're targeting individuals.
So I felt that this was actually a personal problem that I had been dealing with the entire
time I had been in Bitcoin, where I spend one to two days every year basically refreshing
my cold storage setup, making sure that everything's up to date and is still usable.
And if it takes me that long to do it, then I could very easily see how the vast majority
of people would not bother to go to that extreme.
And so I wanted to build a solution that was as robust against both attacks and accidental
losses as the system that I had already set up for myself, but that was much more easy
to use and that the average person could easily follow along with.
So there were various options out there, guides like the Glacier protocol, but that is even
more complicated than my own solution that I had set up.
So we managed to build a very user-friendly solution that is easy to use because it's
just a mobile app interface, but on the back end, we're leveraging a lot of the security
features from hardware devices that are already out there on the market and making it very
easy for people to visualize and update basically the health of their own vault setup with these
multiple different hardware devices that are geographically distributed.
Interesting.
I mean, I think you can also make the case that solving for the end users or I guess
the long tail of users is arguably a more difficult problem to solve as opposed to solving
custody for large institutions who may have more resources behind them and are more sophisticated
in nature.
Yes.
And it's hard to sell security.
So BitGo, I think, made the right decision early on in targeting other businesses because
these exchanges and payment processors that are probably custodying millions if not tens
of millions of dollars worth of Bitcoin, they have a lot to lose and it's not actually
their money.
So they have an incentive to pay in order to have better security and not lose other
people's money.
Whereas when you're talking about the individual, the average individual does not really think
that way.
We're not really used to paying very much for security.
You might have an alarm on your house or something, but in general, we're more used to paying
for insurance, which will reimburse us after the fact.
We don't really think about paying for security to prevent the bad things from happening in
the first place.
Right.
Right.
So I guess how for people who are listening to this episode and are not familiar with
Casa, what is Casa?
Sure.
So at a very high level, our motto is to help improve personal sovereignty.
That's a very broad mantra for us to sort of operate under.
But the way that we're starting out approaching that is by helping people manage their private
keys and helping people manage their full nodes and basically acting as fully validating
entities on the Bitcoin network, because that also gives you the best security posture.
So we have a couple of different products that we've released over the past few years,
the first of which was our Keymaster product, and that is the Multisig Self-Custody Vault
solution.
We've got both a three-of-five multisig and a two-of-three multisig, several different
tiers based upon the value that you are trying to protect and how much effort you are willing
to go to, how distributed you want your keys to be, and how redundant against various loss
and failure scenarios.
And in fact, we just released about a 50-page PDF that is our Wealth Security Protocol that
goes into all of the details of the decisions that we made around these different multisig
tiers and the attacks and mitigations for those attacks that each of our products helps
protect you against.
Now the other thing that we released about a year ago was our Casa node, which is this
plug and play Bitcoin and Lightning node.
It's basically a Raspberry Pi platform with a lot of custom software that we've developed.
And the idea is that you just plug it into your network and your power, and it's very
easy to set up and deposit some Bitcoin onto, which will then be input into the Lightning
node to create channels and allow you to send and receive payments on Lightning.
Originally, this was only really accessible through your web browser when you were on
your home network for security reasons.
And then over the past few months, what we've done is create a few other pieces of software
that make it possible for you to access your node from anywhere in the world securely over
Tor.
Those things are the Casa browser extension, and then more recently, the Sats app, which
is another mobile app, which is very easy for you to connect to your node just by scanning
a QR code and inputting your password.
So that now you're able to send Lightning payments from anywhere globally to other people
on the Lightning network or other people who have Sats app usernames.
And you can do that all over Tor while still actually using your own node that is running
back at your home or wherever you put it.
So kind of what you can see is we're creating this ecosystem of products where we are helping
users help themselves to be sovereign currently from a financial standpoint.
But over the long term, we want to help people be sovereign from many different standpoints,
including having full control over their own data, full control over their own digital
identity, and really any other sort of decentralized and distributed protocols that come out that
help empower people to take back the power from the trusted third parties and basically
run their own services in a way that is censorship resistant.
Right.
There's a lot there.
So I'd like to try to unpack it a bit.
Maybe starting actually with one of the things you mentioned towards the end.
I read through your wealth security protocol that you released recently.
One of the things I really like about it and appreciate about it is, like you said, you
basically walk people through your reasoning for making a bunch of decisions there.
Why did you think it's important to share that?
Because I think it's quite unique.
And I personally really appreciated that.
But I don't see many companies who release a 50-page PDF explaining why they made a bunch
of different design, usability, security decisions.
Curious what prompted that.
Well, it's actually a natural result of the conversations that we've ended up having over
the past couple of years.
What we've found is that a lot of our high end users who have millions of dollars worth
of Bitcoin that they want to keep secure and don't want to trust to third parties, these
tend to be more sophisticated users.
And they, on the whole, do not want to just blindly trust CASA that we have a good solution
because of my reputation or anyone else's reputation who is working at CASA.
We ended up having what we found to be a lot of very similar conversations with these new
users during the sales process.
These are the types of questions that would come up.
They would basically want to know what happens if X happens?
How do I recover from failure Z?
And it just became a lot more efficient for us to formalize all of the questions and answers
that we were seeing on a regular basis so that we could just give people this document
rather than piecemeal having one-off conversations with people.
And once again, it kind of goes back to what I said earlier, that it's very hard to sell
security to people.
This type of document helps people understand that we're not just claiming to be secure.
We're actually, we're showing our proof of work that we have worked through all of these
possible failure scenarios and we have built a solution that can mitigate against almost
all of them.
Now, the ultimate, I guess, problem with any self custody solution is that there always
will be a path to failure where because the user is in control, they can make enough mistakes
that they lose access to their funds.
And really what our job is at CASA is to try to build guide rails within our software,
within the processes that the actual wallet follows to try to prevent users from going
off on any of the paths, the sort of decision trees that can lead to failure.
But ultimately, we can't prevent someone from making enough mistakes if they're not following
our guidance, if they're not following the software and the warnings that we're displaying
to them.
But we try to make it as bulletproof as possible.
Right.
One of the other things I really like about your approach, I know you've mentioned previously
the emphasis that you put not just on security, but also on usability and trying to make it
as intuitive as possible for users to start using CASA without necessarily having deep
technical knowledge.
Yes.
I mean, the goal is really to make it as easy to use as any other mobile application.
So we have a great team of designers and they have made it very easy for you to visualize
your security basically all on one screen on the mobile app.
And then you can dig into the different key sets and different devices that are helping
secure your funds and you can perform health checks.
You can even perform rotations of devices.
If a device breaks, you just go buy another one off the shelf, plug it in and click through
the wizard that we have in our software there.
One of the unique things actually about CASA's Keymaster is that we have actually eliminated
the need for the user to keep the backup seed phrases stored.
And we actually found that this was very important to help better understand the security model
because when a user sets up a new hardware device like a Trezor, a Ledger, a cold card,
what have you, it gives you this 24 word seed phrase and says, write this down and keep
it safe.
And there's just an entire mountain of IT and security knowledge that is contained within
that sentence that gets completely glossed over.
And what we found is that when we were trying to reason about the security model, it became
impossible for us to reason about the security model if we had no idea what was going on
with these seed phrases.
And so we figure if the user is writing down the seed phrase and then could be doing absolutely
anything with it, they could be leaving it out in the open, they could put it somewhere
where someone else sees it and throws it away because they think it's trash.
It just became impossible for us to fully be confident in a multi-sig setup where you
have basically the exposed private keys floating around who knows where.
Right.
It defeats the purpose of the whole product.
Exactly.
And we basically said to ourselves, if everyone agrees that users should not be managing private
keys directly, I mean, that's why we have hardware devices, specialized hardware to
do that, then why is it okay for us to just give the seed phrase in an unencrypted format
to the user?
Because that is all of the private keys right there on a piece of paper and it didn't really
make sense to us.
And so when we figured out that we could actually get rid of the need to store the seed phrases
by having a more flexible system in the multi-sig setup where you can basically rotate out a
lost device, that I think was one of the major advancements that we made just in the security
model in general.
Absolutely.
I think that's really important.
Actually one of our thesis is, you know, keep the can't really scale much more before we
actually solve the private key issue, right?
Like writing down a seed phrase or a private key or whatever it is and, you know, storing
it at the safe or maybe not, it's not going to work.
Not just because it feels very clunky and, you know, prone to hacking, but also because
of what you said earlier and maybe even more likely, people just forget about it and won't
know where they left it and then they just lose access to their funds.
So on Keymaster specifically, so when you said earlier two to three and, you know, you
talked about three to five, basically what that means, got me if I'm wrong, in the three
to five case, you only need three out of the five keys that you get in order to unlock
access to your funds, but each one of the keys on their own is basically worthless.
And then you can rotate between them if you lost access to one of the keys.
Correct.
So in the three of five setup, you'll have one key that is on your mobile phone secured
by the secure element that's in the hardware on your Android or iOS device.
You will then have three hardware devices, you know, Trezor, Ledger, Cold Card, what
have you, and those will be in geographically separated locations.
And then the final key is held offline by CASA as a sort of extreme disaster recovery
scenario.
So in normal operation, if you lose your phone or you change phones or it breaks, or any
of your other three hardware devices get lost, stolen, or stop functioning for any reason,
then you can just go into the Keymaster app, click on that device and say, hey, I need
to replace this, go buy a new one, and we will walk you through the process of basically
sweeping the funds from the current set of keys and sending them to your new set of keys,
which will be the same four out of five, but only one new key set.
And so that gives users a level of flexibility where they can basically re-secure themselves
without ever having to call up CASA or bother support.
Now, if you lose two devices at the exact same time, then you get into a situation where
you have to go through the CASA assisted recovery process and will need to do some additional
authentication to make sure that you're not under duress and that, you know, nothing fishy
is going on.
And we set up some of the processes beforehand with the user of how long do we put a sort
of a freeze on this process.
It's usually three to five days, but it can be longer, whatever the user wants, before
CASA will actually sign off on a recovery transaction with our key that we're keeping
in cold storage.
And so in general, this three of five setup seems to be incredibly robust against all
of the factors that we outline in our paper.
And we have ended up releasing the two of three multi-sig, which is not as robust, but
it's simpler and we price it a lot lower where basically anyone who buys our gold level package
for $300 a year will get that two of three setup along with the plug and play node and
hardware device and probably a few other things.
So we also have like a Faraday bag for you to keep your hardware device in.
But the idea being there that, you know, we want to get people using our system and understanding
how user-friendly it is in the hopes that, you know, over the long term, as Bitcoin continues
to grow and become worth more, that people will find that they're very happy with their
current setup and that if they get to the point where they are securing a much larger
amount of value, then they may be interested in upgrading to the higher security, higher
key set.
Yeah, and how important is it to use different hardware wallets as part of this process?
You mentioned, you know, ledger, tracer, so forth.
Can I just use, you know, two or three different tracers or is your recommendation to diversify
that as well to reduce this?
Yeah, so the entire point of everything that we have set up is to eliminate single points
of failure.
We don't want CASA to be a single point of failure.
We don't want any of these hardware device makers to be a single point of failure.
We don't want Apple or Google or any of the phone makers to be a single point of failure.
So that is why we recommend using a diverse set of manufacturers basically to prevent
supply chain attacks because no one is actually going to go in and audit the hardware.
I mean, that's an insane proposition that requires a level of resources that pretty
much nobody is willing to go to the expenditure to do.
And so if you figure that, hey, I'm spreading out my keys both geographically and digitally,
you know, and across a number of different companies and different hardware and software,
then the likelihood that enough of those different companies have all colluded or have all been
exploited by the same actor and will be able to, you know, work in concert against me simultaneously,
it becomes incredibly low.
And that is really about the best guarantee that you can get in this space.
You know, there's no 100% guarantees.
It's all kind of about probability and thresholds for risk.
Right, right.
Makes sense.
And then you talked about the assisted recovery in the unlikely case that you lose access
to two devices pretty much at the same time.
Can you expand a bit more about how that works?
I mean, when I sign up, do I need to provide a lot of details about myself so that in the
unlikely event that I do need to go through the assisted recovery process, you can verify
that I am what I claim I am or how do you do that?
Yes, this is where it gets interesting because of our privacy protection, where we're very
keen on avoiding getting to the point where we do any sort of AML or KYC level identity
verification.
So it actually is customized for each user based upon what level of information they
are comfortable with giving.
So some people will give us a selfie and do a video call with us so that we can have that
level of detail.
But it is definitely not required that you give us your real name or even photo or anything
else like that.
If a customer is more privacy conscious and we have plenty who are, who basically give
us a throwaway like proton mail email address that they created specifically for this purpose
and they never interact with us using their real name, then we have other options where
basically we can use other forms of authentication like recovery questions.
We have a whole bank of questions that we've developed that are not your standard type
of questions.
They should be things that are not publicly known.
You can also do other types of photo authentication where you take a photo of some object that
is personal to you that you may have in your possession or in your house or whatever, but
no one else would know that that is something that you would use for authentication purposes.
So at the higher platinum and diamond tier levels where we have users that are securing
millions of dollars, we have what is almost like a white glove concierge level of service
where we are customizing a lot of things basically to suit each user and ultimately the final
tier of customization and where it gets really complicated and this is something that we're
still working on.
But basically the next extension of the CASA well security protocol is the CASA inheritance
protocol and that is another thing that a lot of users have talked to us about.
They're interested in having a setup where they can be confident that it's not only robust
against attack and robust against loss, but they also want to be confident that in the
case that they pass suddenly that their errors will be able to access the funds.
This is where it gets really customized and really complicated because everybody's personal
family situation is different.
People live in different jurisdictions that have different estate laws and so we've spent
pretty much most of this year working on the inheritance protocol as well trying to come
up with a general guide that is also customizable based upon each user's needs and we're currently
in the sort of beta testing process of having some customers go through that and give us
feedback.
Yeah, I can imagine that's a whole new challenge in itself, inheritance and how do you solve
for that?
So it sounds like the system that you're offering is very flexible and customized to each user
and their own needs.
Definitely.
One thing also I wanted to ask you about Keymaster is you talked about using your mobile device.
What about desktop support and do you think that mobile is more secure inherently than
desktop?
Yeah, so it can get complicated to evaluate the security of any given device.
In general, I would say that mobile devices are more secure than desktops and specifically
we actually have a big issue with a lot of services that are browser-based.
We've been researching how easy it is for malicious browser extensions to screw around
with web-based services.
But the problem with desktops is that they have a much larger attack surface.
It's much easier to get malware and basically exploits that will affect your desktop.
One of the most common ones that we've seen in the crypto space is actually the clipboard
hijacker malware that will actually swap out crypto addresses that you copy and paste for
addresses that belong to attackers.
While it is possible to get malware on your phone, just the nature of the fact that the
iOS and Android operating systems tend to sandbox the applications much better than
what happens on Windows, for example, it does give you a much stronger security posture.
In general, we find that developing for mobile platform is just easier to get a more consistent
user experience.
Got it.
And then what's behind your decision to focus on Bitcoin and so far not to expand to support
other tokens beyond that like Ethereum and so forth?
Yeah, that's actually mostly based upon my experience that I had at BitGo where the last
year that I worked at BitGo, we overhauled a lot of our infrastructure to make it more
generic and make it easier to add in support for all kinds of other tokens and crypto assets.
And personally, what I found, at least for me, was that we ended up adding a ton of complexity
by adding all of these other assets, and it's really hard to justify the return on investment.
Our infrastructure became an operational nightmare to manage.
I think I wrote a few blog posts and gave a few presentations about how much more challenging
I found Ethereum infrastructure to keep running compared to Bitcoin.
And when you go in and you actually look at the value that is being stored and transacted
on these other networks, I mean, it just falls off so quickly and there's such a long tail
of assets that while sure, you could argue that we're losing out on some potential revenue,
it's ultimately so much smaller relative to Bitcoin that the technical complexity that
you end up adding, it's hard for me to justify.
And when I look at how much progress we've made over the past couple of years, but still
how much more work I would like for us to get done just on Bitcoin, it kind of pains
me to think about how many things we would have to forgo adding support for on Bitcoin
if we started adding other assets.
That's really interesting.
I could see that pretty much for every other crypto asset, but in the Ethereum case, I
guess when you think about the ERC20 ecosystem and how many of these crypto projects are
built on top of Ethereum, that's a really interesting decision.
Well, there's also the security aspect when it comes to multisig.
And I also have spoken about this, but there's an inherent problem with Ethereum because
they don't have native multisig functionality.
You actually have to run a smart contract that has multisig functionality that is built
into the contract.
And this is something that BitGo spent, I think, a year and a half developing a multisig
smart contract and had it audited three or four times.
And basically every audit kept finding problems that could have been catastrophic.
And it seems to be a little bit better now.
I think a lot of the standardization has fallen around the Gnosis multisig smart contract.
I'm not sure if anyone else is using BitGo's multisig smart contract, but we saw multiple
catastrophic failures with smart contracts in Ethereum because it is, I believe, deceptively
easy to write smart contracts in Ethereum, but deceptively difficult and complicated
to write secure smart contracts.
So there were a few examples.
I believe Parity, for example, had a multisig smart contract that basically got attacked
and broken.
And forget how many tens of millions of dollars were pretty much frozen forever as a result
of one simple bug.
And that's just the kind of security nightmares that keep me up at night.
Yeah.
Can you talk, Jameson, a bit about your nodes?
And I guess the basic question is, why should an end user even run a node?
What's behind this?
Why is it important?
And why one of your very first products?
Yeah.
And I've written a number of articles about this because it's a common question.
People basically say, why should I run a node?
It's not paying me anything like I would get paid if I was mining, for example.
What is my incentive to run a node?
And the incentive is actually quite clear.
It is not a direct financial payment incentive, but rather it is a security posture incentive,
where if you are running a full node and full is short for fully validating, then you are
in the strongest security model that Bitcoin has to offer.
And that is the model that is often referred to as trustless, but it is at the very least
trust minimized because you are not trusting that anyone else on the network is being honest
to you because you are going out and talking to peers on the network, you're asking them
for data, and then you're validating the data that you receive.
And if you receive invalid data, then you basically block the peer that sent you invalid
data and you don't trust them to send you anything anymore.
So why is this important?
Well it kind of goes back to the whole decentralization and censorship resistance philosophy that
the network was founded upon, which is that if you're connecting to a trusted third party
to tell you the truth about your transactions and balances, then sure you might be a little
bit better off if you're still maintaining your private keys, but you're once again recreating
a sort of bank-like system where you're creating choke points where third parties could be
lying to you.
And why would we want to do that when we don't have to?
Now obviously for a lot of people running a node seems like too much work or requires
too much technical knowledge, but that's exactly why we created the CASA node is because we
believe that it should be as simple as plugging something into your router and your electrical
wall jack and then using other software to basically make use of the full node.
So I often describe a full node as it kind of acts as a sentinel on your behalf.
It is making sure that nobody is breaking the rules on the network.
And it's doing this 24-7-365 and it's basically acting as your own personal auditor.
And as long as you can keep that computer running, then you can be assured that no one
is trying to defraud you, at least at the protocol level.
And then what happens if for some reason I unplug the node?
Maybe there's even, you know, I lost electricity in my house or something.
Am I at any risk if temporarily I unplug the node?
Not really.
The biggest problem that comes into play with having your node go offline for an extended
period of time is if you're running on the Lightning network, you could run into problems
if your node is offline for more than several days at a time.
And this is where we start getting into the area of research around what is being called
watch towers, where you can have other nodes on the network essentially that are watching
out for transactions that are trying to defraud you and basically stopping those from stealing
your money if you go offline for a long period of time.
But you know, that's a bit more complicated to get into and watch towers are still in
the early days.
There is watch tower software available, but it's not yet widespread use.
Okay.
And what's next for Kaza?
So you see yourself, for instance, going beyond Raspberry Pi and like, what's your vision
moving forward?
Yes.
So we want to do a lot more than just Bitcoin and Lightning, like I said, which is going
to require faster, beefier hardware, though we're going to want to keep it all in a small
form factor.
You know, we don't want to have to ship out desktop sized computers to people.
But you know, this hardware is continuing to improve.
We're actually currently working on the next minor iteration in the hardware, which is
just going to swap it out for the Raspberry Pi 4, which is already much faster than the
three that the current Kasa node is based upon.
And in fact, it's fast enough that we will no longer have to ship the blockchain precinct,
which means you'll get the node and within one to two days of turning it on, it will
have fully verified the entire history of the Bitcoin blockchain there at your house
or wherever you put it.
But looking further out, you know, to next year and beyond, we are most likely going
to end up doing more customized hardware that is even more performant and hopefully will
get us even greater savings of manufacturing at scale.
This first device has been all off the shelf parts and hand assembled by us.
And if we can continue to keep ramping up production and actually getting to at least
some medium scale manufacturing level, then we can start reducing our costs a lot and
be able to do more customized stuff.
There's a lot of things that I would like to see, but I do hope that in a year or so
we'll be able to put out a more customized board that is also going to have more customized
case and a lot more processing power that we'll be able to run a variety of different
services.
We kind of anticipate having almost like an app store itself within the Kasa node of,
do you want to run a like a data file storage, you know, distributed Dropbox type of setup?
Or do you want to run your own BTC pay server for commerce purposes or who knows, you know,
some people have even expressed interest in running their own email server, though I know
from personal experience that that can be a nightmare.
But you know, these are all the type of things that we're going to continue experimenting
with.
That's fascinating.
Like you said, I think earlier, there's just so much ahead.
So really curious to see what you do next.
One more question on that just before we do want to ask you also a couple of questions
about the market more broadly.
But one more question about that, curious about your view on, you know, MPC technology
and threshold cryptography.
Have you looked into that, and do you see that as something that you're going to build
on top of in the future?
So you know, multi-party computation stuff.
Are you specifically referring to like the setups that are being used for various ZK
snarks?
Yeah, exactly.
Yeah.
So the whole area of zero knowledge stuff seems to be moving really quickly right now.
So it's definitely something that I'm keeping an eye on, though, you know, when it comes
to security, you often don't want to implement the freshest, most cutting edge stuff.
You actually want to implement the things that have been around for a number of years
and have been attacked and attacked and, you know, withstood a ton of attacks.
So this is the type of thing where, unfortunately, I have to say, you know, time will tell.
Because kind of like I said earlier, there are no real guarantees.
It's once again, one of those things where the longer something has withstood the test
of time, the more confident that we are that there are no lingering major exploits out
there.
Yeah.
Yeah, that makes sense.
So we are focusing on the tried and tested, as I think many proponents of Bitcoin argue.
Bitcoin has been around for over 10 years now and has been tested certainly to a certain
degree more than any other crypto asset out there.
So I can definitely see how that applies to security even more so.
So kind of shifting gears a bit, I do, as I said, want to ask you a couple of questions
about the market more broadly.
What's your view on the status of the Lightning Network and the pace of scaling of the network
so far?
Well, it's definitely grown a lot over the past year.
The kind of awkward thing is that as Lightning Network continues to be improved on the privacy
side, it actually becomes more and more difficult for us to visualize the growth of the network.
It's already impossible for us to actually know how many transactions happen on the network.
And it's becoming more and more difficult for us to even know how many nodes or how
many channels are on the network.
And it's actually by design.
So while it's hard to actually look at the network itself, we have to then kind of look
at second order of effects of how many services are hooking into the network, how many digital
different people are talking about using the network.
And that is something where I think there has been some progress made.
But one of the most important missing pieces that I think is going to have to happen for
Lightning Network to really grow and become more usable is we need to get the exchanges
onto the Lightning Network.
At this point, I'm only really aware of I think Bitstamp has said that they're experimenting
with it.
But I don't think any of the major exchanges have actually added it yet as an option.
And that's important just to, again, increase the volume and the number of transactions
that are happening on top of Lightning?
Well, and it's important from a liquidity standpoint.
I wrote an article, I think about three years ago now, where I actually showed some simulations
of sort of channel rebalancing and liquidity management, where I believe that it's incredibly
important that people should be able to get inbound liquidity to their node via paying
out of band in some other fashion, by which I mean, if you have a bunch of channels on
your Lightning node and they get imbalanced and they can potentially become unusable for
either sending or receiving, depending on which way the channel is imbalanced, then
you want to somehow rebalance that channel to avoid having to close it and reopen it
with on-chain transactions.
And if you can reach out to some other liquidity provider on the network, one example right
now would be BitRefill.
I think they do something like this with Thor and some of their other Lightning services.
But if you were able to just go to your exchange where you already have an account and basically
send them money through a traditional financial network, then they could send you the money
over Lightning to rebalance that.
And this is, I believe, going to be one of the greatest challenges over the long term
with the Lightning network.
It's not going to be so much the security or the privacy engineering, but I think it's
actually going to be the engineering of managing the liquidity and doing that in as automated
a fashion as possible.
One of the things that I've been saying quite a bit is that I don't believe that Lightning
network is going to be able to be a mainstream thing as long as the user has to understand
what channels are.
I kind of look into the future and when I try to envision what the optimal Lightning
wallet would look like, it would just be a Bitcoin wallet.
You wouldn't even necessarily know that it's using Lightning.
The software itself should be figuring out automatically whether to do something on chain
or off chain and how to manage all the channels under the hood and how to do channel rebalancing
and seek out liquidity providers and all of that stuff.
So that's where I think a lot of the coding and a lot of the research is actually going
to need to take place over the next few years.
Right.
All right.
What do you say to critics who claim that the hub and spoke model of Lightning could
actually generate quite a bit of centralization?
That is actually something that I addressed in that article three years ago, which seems
to be holding true, which is that we are going to see a kind of power law of liquidity providers
where you would expect that the large enterprises like the exchanges and payment processors
are going to have the largest presence on the Lightning network.
And then you'll have some medium to small other retail businesses that will also have
a decent amount of liquidity.
And then you'll have all the individuals who are just interested in making the occasional
payment, but they're not going to be receiving a ton of payments.
And so this will create a graph that to some people looks highly centralized.
But ultimately, the question is not, what does the network graph look like?
The real question is, is it censorship resistant?
And so if you had a hub and spoke model where there were only a handful of hubs and everyone
had no real choice but to connect to those hubs, then you would have lost the censorship
resistance there.
But in Lightning network, that's not the case.
You can connect to almost any other node on the network that is making itself publicly
available.
So you can always route around the hubs if they become bad actors.
And this is one of the things where I don't think it's possible for me to prove that it's
impossible for Lightning to turn into a hub and spoke network.
But we can continue looking at the network graphs and see pretty clearly that it is not
today.
And I find it highly unlikely that it will turn into that.
And I am actively fighting against that by making plug and play node hardware and software
so that the individuals can be a part of this network mesh on their own.
Right.
That makes sense.
And speaking of the market, curious about your views on, you know, we've recently seen
Bakkt launch, much anticipated, I think.
Any thoughts on that?
How important is that to the ecosystem, to the crypto ecosystem overall?
Maybe you see even a potential collaboration with CASA moving forward?
So far, we've generally stayed away from any institutional stuff because once you get to
a really large level and you start being regulated, we probably don't want to have anything to
do with that.
We want to remain a pure software provider.
But I don't know.
It's been so many years now with people talking about, you know, quote unquote, institutional
interests coming into Bitcoin that I don't even really worry about it or think about
it very much.
Personally, I don't want like institutions to get into Bitcoin super early because I
want the individuals to be able to get into Bitcoin super early and, you know, reap all
of the benefits of that themselves.
But you know, institutions are free to do as they please, at least within the confines
of their regulatory framework.
But it's not really something that I think is going to make or break Bitcoin itself.
And I think if you just kind of look at what I've been working on and what I focus on,
you know, I care more about the individuals.
Right, right.
So the end user, the individual who is learning about the space and wants to get involved.
So last question before we end, Jameson, your emphasis, I guess, on personal privacy has
been well documented.
The New York Times ran an article about it earlier this year.
One thing that really caught my eye was you mentioned that these days you're working remotely
and don't really join meetings in person.
And that's something that's really interesting to me because I always tend to think that
there's a lot of value, just for my own experience building products, there's a lot of value
to a team being at the same location.
You know, oftentimes it just fosters innovation and casual conversations can lead to new ideas.
I'm curious if you can share more about your experience working remotely and do you think
it is better than, you know, being in person in the same location with the team?
Or maybe there are multiple locations, but the idea of centralized locations, I guess,
for teams.
Right.
So I've been working remotely for almost four years.
I certainly don't miss commuting.
From a team building standpoint, there is something important about having this natural
background noise, like as people are collaborating and, you know, cross pollination, overhearing,
you know, people talking about things that you may have not learned about otherwise.
And actually something that we did at BICCO that I have also brought over to CASA is the
idea that you can actually kind of simulate or recreate some of that by having a persistent
video conference on in the background.
So, you know, as people may want to just chat about one small thing or another, they can
actually do it on the video conference and you still have the ability to overhear that.
And if there's something interesting where you want to get clarification or you want
to put your own input into, then you can actually do that over the video conference.
Or if it's just getting annoying, you can mute it, which is not necessarily an option
if you're actually in the office.
Right.
Interesting.
So that's something I haven't, I don't think I've heard before.
So basically like ongoing video conferencing rather than scheduling a video conference
for a specific meeting.
Exactly.
Interesting.
Yeah, because I find that scheduling meetings and all of that, that just, you know, it becomes
so formal that sometimes you just have these casual conversations where, yeah, you didn't
even necessarily think about some of these ideas before and then they just come up as
you're having lunch or, you know, talking between meetings.
But interesting.
Well, thanks a lot, Jameson, for coming on the show.
It really has been a pleasure talking with you.
Thanks for having me.
Thanks for listening.
If you like this episode of The Blockchain BC and want to help us bring more awareness
to the space, I'd really appreciate it if you can rate, review and subscribe to our
podcast.
It takes a few seconds and helps us get the world out.