{ "episodeId": "SLP42", "speakers": { "stephan": { "name": "Stephan Livera", "role": "host", "tag": "STEPHAN" }, "adam_back": { "name": "Adam Back", "role": "guest", "tag": "ADAM" } }, "segments": [ { "speaker": "stephan", "time": "00:02", "start": 1.5, "text": "You're listening to the Stephan Livera podcast focused on Bitcoin and Austrian economics. Listen in and learn alongside me as I interview some of the sharpest minds ranging from economists, software developers, investors, entrepreneurs, and writers. Hi everyone, welcome to the show. This is your host Stephan Livera, and today I have a very special guest. He is Adam Back. He has, he has a PhD in computer science. He is the CEO of Blockstream, one of the really important Bitcoin and blockchain companies out there at the moment. He is a cypherpunk and cryptographer, a privacy advocate, and he is also the inventor of Hashcash, which is cited in the Bitcoin white paper. So definitely a very important person. From our ecosystem. Welcome to the show, Adam, and thanks for taking the time." }, { "speaker": "adam_back", "time": "00:51", "start": 50.52, "text": "Yeah, thanks for inviting me on." }, { "speaker": "stephan", "time": "00:52", "start": 52.06, "text": "Yeah, so, I see you guys have been doing some really, impressive things, and, today we, we were gonna talk a little bit about Blockstream Satellite. So I think maybe it would be good if you could just offer the listeners some intro and background on the Blockstream Satellite service and product, and then after that we can then look to talk about what the new update is coming with Blockstream Satellite." }, { "speaker": "adam_back", "time": "01:19", "start": 78.85, "text": "Yeah. So, there are a number of reasons why it can be interesting to, receive Bitcoin data from, satellite. So one is, blockchains are, you know, the Bitcoin blockchain is a pretty interesting thing, but you do have to, have a single ledger. Like a fork in the ledger accidentally can be risky in that you would accept transactions that are on part of the network. And occasionally, countries will have, partial internet disconnections, so people in the country may be able to access local websites or- Now, use the internet locally, but unbundled to them, depending on what they're accessing, they may not be reaching the rest of the world, and this is called a net split in technical terms. It can happen for a number of reasons, including temporary configuration errors, on router, on BGP routers, or like physical disruption, like some, a fishing boat accidentally breaks an undersea cable or something like that. So With the satellite, you have a completely, independent redundant connection to the internet, so you're gonna notice if you're, you know, if there's a divergence between the data you're getting from the satellite and what you're seeing on the local internet. So that's a kind of, redundant connection which is good for reliability. another reason is cost. So in Western countries, you know, many people have internet connections and don't think much about the cost of that, but there are many countries in the world, in emerging markets, where the cost of a high-speed internet connection is, you know, perhaps as much as a average salary or so on. So, and you see that with Sites that try to catalog where Bitcoin full nodes are, and you see that some countries don't have many at all. so the satellite provides, it, it reduces the cost of participating in the Bitcoin network, you know, your ability to run a full node on the network. Because there's no charge for the transaction data and equipment you need to receive it, it's very low cost. So it's about a hundred dollars US dollars worth of parts. maybe less if you can reuse a satellite TV dish, and it uses, software defined radio, so small USB analog to digital converter and does most of the calculation, the signal processing in software on a laptop. so that's, that's another factor, and it's also more private. So even if you have a high-speed connection, you're advertising to the internet at large that you're running a full node. It's a peer-to-peer network, so it can look for your node and connect to it. And, you know, even if you're- Not, accepting inbound connections, other people on the network will see your outbound connection, your ISP can see that you're running Bitcoin. And so, receiving Bitcoin data via satellite has much better privacy. It's a passive, receive-only technology, so people can't tell that you're using Bitcoin." }, { "speaker": "adam_back", "time": "04:25", "start": 264.72, "text": "and I think it also helps with decentralization, so some robustness and availability. So there are situations where governments in countries undergoing political turmoil, where the government or military may disrupt or turn off internet or GSM phone service, and so that prevents people from, you know, coordinating, exchanging pro-private messages. Political commentary about what's going on and, being able to pay with, with, Bitcoin. So having redundant connection and connectivity can allow people to transact. reliably in, you know, any location and in any kind of network conditions." }, { "speaker": "stephan", "time": "05:14", "start": 314.24, "text": "Yeah, I love that explanation, and it really, to me, it strikes me as this is like a critical piece of Bitcoin infrastructure that Bitcoin potentially, it's on this pathway of potentially becoming the global money of the world, and if you want the global money of the world, everyone has to be able to access it. So the next thing I was keen to understand was how, can you talk? Talk us, to us a little bit on the structure of Blockstream Satellite. So from what I was reading, it looks like there are teleports and uplinks that transmit the blocks to the satellites, and then the satellites transmit the Bitcoin blocks to people who have set up the receiver. Could you talk a little bit about that setup?" }, { "speaker": "adam_back", "time": "05:56", "start": 355.67, "text": "Yeah, so, just today we added, well, it's, it's been online in testing mode, but we turned on additional coverage soon, so I can talk about that as well. So we have now four satellites, covering six coverage zones, and they're" }, { "speaker": "adam_back", "time": "06:20", "start": 379.87, "text": "there are redundant connections. So the, the teleport, let's just back up. A, a satellite is actually a fairly, simple piece of technology in one way of thinking, because it doesn't do much in the way of signal processing on the satellite. So generally speaking, it's broadcasting down through an array of antennas to cover land areas. Data that was sent to it in, in an analog form factor, so it just sort of turns it around and broadcasts it down again. So the- The intelligence, if you like, is in the uplink. So the uplink or teleport is a, a site with a redundant, reliable internet connection, power, some, signal processing to convert digital data into The right signal to send to the satellite, the right frequency range and encoding, and then that gets broadcast up to the satellite with a, with a, a dish, a larger dish. And then It's, the characteristics of it were tuned to support small dishes, so forty-five centimeter, dishes, which aren't sort of standard, satellite dish size for satellite TV to keep costs down so people could use, the basic equipment. Now, there's some redundancy with the teleports, so for example, over the US, there are two different satellites, and another form of redundancy is the, At the Western, teleport, it's able to see the Eastern satellite. So if the internet connection were to go down on the Western teleport, the Eastern teleport, it, it would be able to receive blockchain data via the other satellite. So there would be a higher latency because the data would go up to one satellite, down to the other base station, and up to the second satellite. So as we've added more, satellites and coverage zones, We've been able to improve the redundancy using the satellite connection itself, if, if an internet connection went down at a teleport." }, { "speaker": "stephan", "time": "08:22", "start": 501.62, "text": "Yeah, I, I'm really getting that sense of building, building out redundancy into the Bitcoin network and ensuring that there is greater global access and ease of access, privacy, and just maintaining, as you mentioned, or lowering the fork risk. So maybe it would now be good to touch on this new update that, that Blockstream, coming- Out with, which I understand is the satellite API. Can you tell us a little bit about that?" }, { "speaker": "adam_back", "time": "08:50", "start": 530.36, "text": "Right. So, one of the things that we found when we first launched the satellite service last year was that people were interested to broadcast other data on it, so Bitcoin related data or applications related to Bitcoin. And so we thought, okay, well, we can probably make an API to allow people to send application messages through the satellite. So how that works is We, on, on the teleports that are sending data to the satellite, we added additional software so that a user can connect to it using, a A REST kind of programming web interface and send a Lightning payment, so using Bitcoin Lightning, which supports fast and low cost micropayments using Bitcoin. A payment for the message cost. So it's quite low cost because it's just comparable to Bitcoin transaction data, and that data would be broadcast across all of the satellites and arrive to all," }, { "speaker": "adam_back", "time": "09:56", "start": 596.48, "text": "Satellite receiving nodes. And it's, so it's a broadcast, but if you want to use it to do a point-to-point message, so a message from one person to another specific person, you can do that by marking it for their attention and, encrypting it for them." }, { "speaker": "stephan", "time": "10:12", "start": 611.69, "text": "Okay. Okay, cool. So what we're doing now is basically putting in a function that enables users to send messages anywhere around the world, and they can use encryption to selectively disclose that message to party- is who they choose, rather than the, you know, the sender, you know, what sort of, if you could comment on the encryption that's used on these and, you know, can the sender or the receiver or the content of the message be known by third parties?" }, { "speaker": "adam_back", "time": "10:41", "start": 641.11, "text": "well, if it's encrypted, then only the recipient will be able to decrypt it. So I think that's the, the advantage there for sending a direct message. There could be use cases also for messages which are intended for an application. So, some people have Started working on connecting Blockstream Satellite download, you know, the, the full nodes that people run when they receive Bitcoin data via the satellite on that node you run a Bitcoin node, so they've started connecting those to mesh networks, so it's a way to sort of replicate that data, data and share the uplink, let's say, in a village, or in a town or a marketplace in emerging markets. So when you do that, you might, you know, you might have a application specific to that, like let's say a wallet or something, and it may miss some auxiliary data. So, for example, maybe you want to know, the Price of Bitcoin, like the exchange rate, and that could be sent by the application developer who maintains the wallet, he could send that data, periodically, and the application could receive it using these mesh networks." }, { "speaker": "stephan", "time": "11:49", "start": 709.48, "text": "Right, so as I understand it, then you're suggesting that people could combo this technology with other technology, such as that GoTenna mesh networking technology." }, { "speaker": "adam_back", "time": "11:59", "start": 719.29, "text": "Yeah, exactly, or like a WiFi hotspot. I think some people have explored the idea. There of setting up, a Wi-Fi hotspot that gives access to Bitcoin in metropolitan areas, like in, in cities or in remote locations. So it could be Wi-Fi technology or mesh net technology. People have also looked at LoRa for longer haul, sort of low bandwidth. Longer distance radio, where the mesh network, is typically short distance, way to share internet connectivity and data." }, { "speaker": "stephan", "time": "12:32", "start": 752.05, "text": "Fascinating. And then you mentioned the use of Lightning Network as well, so can you touch on how, logistically, how, how does that work? Like, is, is the individual connecting to the Lightning Network through a different means with the internet to make the payment?" }, { "speaker": "adam_back", "time": "12:48", "start": 768.28, "text": "so actually, Lightning is for sending, so you would need a normal internet- Internet connection or a bi-directional internet connection to the message sending API. So Lightning is, a bit interactive. It's, it's, you know, when you want to, if you have a Lightning wallet and you want to send to another Lightning user or send to a service that accepts Lightning payment, there are a sequence of messages that come backwards and forwards as you establish the connection, negotiate, your software negotiates the payment. So the Lightning I'm referring to is For the application author or the message sender, it's in, connects to the lightning net, connects using the lightning network to the satellite API, pay for a message and then have that message go through the satellite. Now, it would, it would be possible to use the satellite received Bitcoin data as a basis for doing lightning payments. Now, so lightning is As I say, an interactive protocol, but nevertheless, the cost of receiving a Bitcoin data can be expensive, whereas with Lightning, you're only incurring cost related to the payments you make yourself or payments you route for other people, and so those tend to be, smaller, smaller amounts of data. So, you know, for a main chain Bitcoin transaction, there are people that run SMS gateways so that you can send a transaction via SMS and it, it will make its way back to the Bitcoin network. So it's kind of gateway function. Also, you know, you could, make an interactive lightning payment informed by the full node that, where you've received that data from the satellite. And So it can be the case that people have, you know, two point five G or three G connectivity, it's just that it's expensive to receive the full Bitcoin data on an ongoing basis, right?" }, { "speaker": "stephan", "time": "14:45", "start": 884.72, "text": "I see, gotcha. So it, it, the intended, you know, use here might be that you down-- you use Blockstream Satellite to download, you know, to have your full node, to download all the blocks and the headers and so on. But in terms of broadcasting a specific transaction to make a payment, you might, you know, the end of individual might be able to use their mobile phone connection or one of those gate-SMS gateways you mentioned to do a payment." }, { "speaker": "adam_back", "time": "15:10", "start": 909.81, "text": "Right. And, I mean, the You know, the Bitcoin data on a monthly basis can be, sort of in the gigabytes, right? so maybe ten gigabytes or more, if, you know, that would, that would be if you- Sort of put it into a low bandwidth mode where it didn't resend transaction data and act as a relay in the Bitcoin network if you were just receive only, you know, you could be as high as ten gigabytes a month. And so You know, for areas that have restricted bandwidth, that's a lot of data and could be expensive to receive. However, even if you're paying at quite high data rates, you know, using GSM data or there are satellite based, bidirectional services that are expensive per megabyte. you know, to send a single Bitcoin transaction, which could be as small as two hundred and fifty bytes or something, it, it could still come out quite cost effective, you know, a couple of cents or so, where to receive all the transactions sent in a month might be prohibitive. So you can also marry it with other forms of connectivity that have A higher cost, but then you're only paying for your own transactions. And, and only on the sending side, the receiving side you would see back through the satellite." }, { "speaker": "stephan", "time": "16:27", "start": 986.95, "text": "Right. And then you touched on this earlier, so what sort of cost would we be looking at to send a message? And let's talk in terms of dollar terms, and then also in Bitcoin or in Satoshi terms." }, { "speaker": "adam_back", "time": "16:39", "start": 998.63, "text": "So the satellite service is generally for smaller messages, so a few hundred bytes, a few kilobytes, that kind of thing. And I would say You know, one cent, ten cent, that kind of range, depending on popularity. So if it gets congested, the price might go up a little bit. Right, right. I mean, I think generally it would be sensible to consider it similar to Bitcoin. You know, there's OpReturn, a small data field in Bitcoin, intended for, protocol extension messages generally, but you can think about it as similar to that because if the cost of sending via the- The API was expensive relative to that, people would have an incentive to put data into the, into the blockchain itself, and that, that would also get sent via the satellite, right?" }, { "speaker": "stephan", "time": "17:29", "start": 1049.25, "text": "Yeah, yeah, and that's another whole vector for attack as well. Okay, so what sort of-- Let's talk a few examples in terms of what sorts of censorship resistance does this enable? As, as I understand it, it's, it's an alternative, to sending messages through the standard internet or if the internet is blocked. Locked. Can you talk a little bit about that and maybe what are some potential, you know, what would be some typical uses of this service?" }, { "speaker": "adam_back", "time": "17:57", "start": 1076.89, "text": "Yeah, so I mean, I guess if the internet were shut off, you would still be able to receive Bitcoin data and application message data. And another distinction with the satellite data is it's broadcast, so it's not kind of directed at an individual user as far as people observing the network data, where internet data is point-to-point routed so you can see Which IP address it's intended for or something like that. So it provides potentially better privacy characteristics and also it's, it's, inherently a broadcast mechanism as well. So if there's data that needs, you know, that would need to go to many people, it's lower cost on a per recipient basis. So The, Bitcoin Satellite data, doesn't it increase in cost if, if, you know, a million people receive it or ten million or a hundred million, so on, right?" }, { "speaker": "stephan", "time": "18:50", "start": 1130.43, "text": "It's very scalable in that sense." }, { "speaker": "adam_back", "time": "18:52", "start": 1132.21, "text": "Yep." }, { "speaker": "stephan", "time": "18:53", "start": 1132.81, "text": "One topic that confuses some people is, they might wonder, okay, well, is this satellite system very centralized, right? If I'm a user of the Blockstream Satellite, am I trusting the transaction and blockchain data received via this?" }, { "speaker": "adam_back", "time": "19:09", "start": 1148.62, "text": "Yeah, that's, interesting. Interesting question with a interesting technical answer. So now you can, you can look at analogy of how the Bitcoin payment network, transmits information. So typically a Bitcoin node, has eight neighbors, so it's establishing or maintaining connections with eight sort of randomized other nodes in the network, and then a transaction data is relayed through those. And you could ask the question, you know, if, if I had eight connections to my computer, and one of them's very high speed, you know, like ten megabits or s-something that's gonna get data quite quickly to me, and the other one's a very slow, you know, sort of, 2G modem speed or something. Now, you could ask the question, \"Do I have to trust the high-speed node compared to the low-speed ones?\" And the answer is no, because the data is, Sort of, checksums and the proof of work, is verifiable to your node software. So even if the high-speed connection tried to give you an invalid set of data, your software would automatically recognize that that's not the longest chain and, you know, discard that and focus on the correct Data with more proof of work. So it's something similar with the satellite, so you can view the satellite as a fast, node, and then you can cross-check it, you know, you can cross-check it with a low priority, slower connection or receive headers only over a, a, a standard connection. Or there's also, also was a, A phone service, a sort of, robotic voice you could call up and it's kinda like the time service, it tells you the current block height and the checksum of the latest block. So with something as low bandwidth as an SMS, you could check that you're on the right chain, and share that information onwards and have other people verify that it's correct. So from that point of view, you're not really trusting the provider of the data because it's very low cost to cross-check it and to verify locally that it's correct." }, { "speaker": "stephan", "time": "21:28", "start": 1287.51, "text": "Fantastic. Alright, so similar to a devil's advocate question that I've asked an earlier guest, Rajul, Rajul Walsh, on this whole topic of trying to build tools that help enable censorship resistance, that sometimes, you know, a detractor might ask this question or they might say, \"Oh, look, you Bitcoin guy- Guys, you're just, you're just larping, you're live action real role playing, you know, they might say, \"Look, why you, you know, why are you guys being so tin foil hat paranoid? You know, that sort of censorship wouldn't ever happen here.\" How, how, how would you sort of respond to, you, your person who, who just generally doesn't feel the same way?" }, { "speaker": "adam_back", "time": "22:07", "start": 1326.82, "text": "well, I think, there have been a number of Events in recent history that show that, censorship isn't hypothetical. You know, if even in Western countries we've seen deplatforming of unpopular opinions, that's a very recent topic, right? And So people have been disconnected from social media applications, there are whole segments of the economy w-who are essentially unable to obtain banking service, even though the products or services they're providing are legal, just due to Kind of political pressure," }, { "speaker": "adam_back", "time": "22:49", "start": 1368.64, "text": "and, you know, I think also as a general backdrop that the revelations that Edward Snowden made, you know- Over the last decade are, very illuminating of people who had previously been considered to being too paranoid, having shown to be not paranoid enough, maybe. vindicated. At least in the sense of, you know, a government intent to have the mechanisms to surveil and index and catalog information. So it's not that you have to be guilty of anything in particular, but they're just, you know, hoovering things up and- There's a general sentiment, I mean, to what you were saying, some people will say, \"Well, if I've, I've got nothing to hide, so I don't care,\" but, you know, I think what's, politically Popular changes over time, and it's also a matter of principle, right, that people should be able to speak without feeling surveilled. I think there's some psychological studies showing that people will almost start to self-censor if they think that, you know, things they're saying are being collected and indexed. By a government entity. So that's a kind of, sub-subconscious self-restriction of what you might feel inclined to say or discuss." }, { "speaker": "stephan", "time": "24:09", "start": 1449.22, "text": "Yeah, I think that's a very insightful comment that really whatever is popular right now may not be popular in ten, twenty, thirty years time. And also just the point you make around self-restriction and self-censorship, and, you know, maybe we can't be our true selves if we don't sort of feel free to speak, you know, to truly speak freely. Okay, so let's talk a little bit about how, how we would actually set up, you know, Blockstream Satellite connection. Can you talk the listeners through just sort of a high level, you know, what sort of equipment you use, how, how would you go about setting it up?" }, { "speaker": "adam_back", "time": "24:45", "start": 1484.78, "text": "Right, so you need, A satellite dish, and that could be one that has, you know, a used one. There are lots of satellite dishes that get upgraded or left unused, so you can certainly pick up used ones or reuse one that you already have. And, a number of parts that come with it, so an LNB, which is the, the piece that goes on the end of the arm, you see on a satellite dish, and so there's a parts list, but basically, a small, power source that amplifies that. And, the main other piece being a software-defined radio, so it's a kind of USB key and the" }, { "speaker": "adam_back", "time": "25:31", "start": 1531.22, "text": "the analog signal from the satellite dish gets fed into that, and it converts it to digital, and then your, the software you run on your laptop does the signal processing to convert that into digital data and then feed it to a Bitcoin node so that you're, you, you can run a Bitcoin node that's not, so directly internet connected as it were, except by the satellite. so there's a bit of, it's a, in terms of the complexity or skill set to set that up, I would say that pointing a satellite dish, if you haven't done it before, is a little bit fiddly. You know, if you, if you're off by a degree, that translates into a great distance at geostationary. satellite height. But, you can also, you know, engage somebody who's familiar with satellite setups, so like somebody with a satellite TV installation business to set it up for you. So that's one way to, to approach that. And then the, the software part of it is at this point, Linux software. So if you have some kind of power user level skills or follow along instructions, if you have a Linux installation, that, that's a, you know, it's not so hard, but, If you've never used Linux before, you might find that a little tricky." }, { "speaker": "stephan", "time": "26:54", "start": 1613.66, "text": "Okay, so pretty much you need to be reasonably comfortable with command line, and just basically power user level." }, { "speaker": "adam_back", "time": "27:02", "start": 1621.9, "text": "Right, and that that should get easier over time as well, because, you know, we've put some effort into this release into making it easier, more visual to Tune-a-Satellite, so there are tools where it shows you a graphical visualization about the signal strength and you can adjust the dish. so there are a few parameters about the dish, its polarization, its, angle of rotation and the, You know, sort of how high it's pointing and how far east or west it's pointing. So there's a tool on the website which you can put in your, GPS coordinates, latitude and longitude, and it tells you which satellites you should be able to see and what angle they're at to point the dish." }, { "speaker": "stephan", "time": "27:47", "start": 1667.04, "text": "Right. And have you had any feedback from customers who've used Blockstream Satellite so far? What's the sentiment been there?" }, { "speaker": "adam_back", "time": "27:55", "start": 1674.53, "text": "Yeah, people have been pretty enthusiastic about the idea. I mean, it, it, it sounds high tech Great that, that Bitcoin data is being broadcast down from space, and with this release on, you know, much more of landmass, so more people are technically covered, including Australia, New Zealand, and most of Asia, Europe, Africa, North and South America, and so on. So, we've had a number of people, sort of technical people, trying different things, like, making their own satellite receiving kits, including electronics and power supplies and things. And And as I mentioned, looking to connect with mesh networking, and I think, you know, also some people appreciate the privacy pros-prospect, so that they're interested, you know, they may have high-speed internet connection, but they're interested to connect to the satellite network to have greater privacy about Their participation with Bitcoin. So generally speaking, it's, if, if you're looking for sort of asset protection and security and privacy, it's better for you if, you know, the internet at large or at least, you know, random Internet nodes on the network don't know that you have Bitcoin, 'cause that will tend to leak your IP address, which can be geolocated to your physical address. And then if you have- Let's say some Bitcoin backed up on a device in your house, somebody might try to figure out where the Bitcoins are located by, sort of navigating information from the network. So this, this would, provide protection against that kind of thing because it wouldn't be obvious to the network. That your internet connection was, participating with Bitcoin." }, { "speaker": "stephan", "time": "29:42", "start": 1782.28, "text": "fantastic point. Yeah, that's sort of like, you know, how in the movies, where they, look at where people are using a lot of electricity to try and figure out where's, where's the hydroponics being used, kind of thing. So similar sort of concept there. Yep. Okay. are there any other, you know, security precautions or warnings or just some advice for users who wanna go and pick up Blockstream Satellite and try and use the product?" }, { "speaker": "adam_back", "time": "30:06", "start": 1806.36, "text": "yeah, there's a It's online and I think you, you should follow the, the LNB. It sometimes matters which LNB you have and which LNB you need depends on which part of the world you're in, 'cause we're using different satellites to different locations, so pay attention to that part of the instruction. It's not necessarily a generic LNB. And the other, the other thing to watch out for is the equipment, so the laptop or desktop that you connect it to needs to be something like an i5 class machine. The signal processing, so we get the cost of the equipment down very low because it's using a software defined radio. The, USB software defined radio parts go for sort of as low as fifteen, even ten dollars. But that has the side effect of offloading the signal processing to a laptop, and you really need like an i5 laptop, an ARM like Raspberry Pi or something isn't, isn't powerful enough. so that's something we're, you know, we've, we've put some work into optimizing, but it's not, an ARM isn't quite able to keep up. So bear that in mind when you're figuring out what equipment to use and for people who are interested and want to compare notes, there are-- there's a IRC channel, which is the, Blockstream Satellite, and there's also a satellite channel on the Bitcoin Core Slack called Blockset, so you can, Go onto those and compare notes and ask questions with other people who are trying it out." }, { "speaker": "stephan", "time": "31:36", "start": 1896.39, "text": "Fantastic. And then you mentioned around the coverage, so, I actually have a friend who is keen to use it, but he, he mentioned that Australia wasn't available yet, so it's, good thing you mentioned that Australia is becoming available. so what, what are the timelines, or is it available basically now?" }, { "speaker": "adam_back", "time": "31:53", "start": 1913.06, "text": "Yeah, it's available now. So it was actually, running in test mode, but we hadn't published the specifications, so people didn't know it was Broadcasting Bitcoin data at them, but that's been going on for some weeks now. But yes, it's, immediately available, so, if you're, battling- The Australian, ISP problem, this will take some load from your internet connection and let you, you know, do your internet things without having Bitcoin drawing from your bandwidth." }, { "speaker": "stephan", "time": "32:24", "start": 1944.39, "text": "Fantastic. anything else you wanted to say on Blockstream Satellite, Adam?" }, { "speaker": "adam_back", "time": "32:29", "start": 1948.9, "text": "no, that's, that's cool. I mean, what, what we're interested to see is for people to, try different things with it and experiment. It's, it's kind of a part of, the sort of infrastructure story where, when, you know, when you release These pieces of open source or extra infrastructure, there are many developers actively working on Bitcoin these days, and they will sort of release applications or adopt your technology and build other things on top of it. So we're pretty interested to see what people come up with using it." }, { "speaker": "stephan", "time": "32:57", "start": 1977.49, "text": "Excellent. Okay. I had a couple other topics, that might, be interesting for the listeners as well. So obviously, as I'm sure you're aware, Timothy May has recently passed, and, I, I know, you, you did, work with him or, liaise with him in the past. I'm wondering, can you offer the listeners any insight into your interactions with Tim May and, what impact did he have on you?" }, { "speaker": "adam_back", "time": "33:24", "start": 2004.41, "text": "yeah, I mean, most of it was electronic because I was living in Europe and, I guess, he was living in the Bay Area for most of his career and, since he retired. So he retired kinda young, I think, age thirty-four, as it was a, engineer at Intel and invested in a number of tech companies with his, so kind of lived a frugal lifestyle on a professional salary and invested in lots of tech stocks, so he was able to kind of, get the timing right on that. And Then used his free time to focus on things that interested him, which, you know, he came across, public key cryptography and Thought about the use cases for encryption, cryptography, electronic cash. So he discussed, a number of ideas that, you know, only really came to fruition. Most in a, in a re-sort of robust or wide scale sense where people sort of noticed and it came to mainstream attention with the advent of Bitcoin. So, you know, he was talking about these things in the early nineties already." }, { "speaker": "adam_back", "time": "34:39", "start": 2079.29, "text": "So, yeah, I thought as a, you know, very interesting guy to interact with online, quite forthright in his opinions, and, not, not politically correct, so, as a kind of personality interaction style, but very, very insightful, And so I, I think I met him only twice in, in person, like when I was in the, in the US at some point, for a conference or something like that. And more recently in the Bay Area a few years ago as well. so, you know, very sort of pre- and early thinking about a lot of topics that became available with Bitcoin. So, you know, he talked about something he called BlackNet in the early nineties, which was sort of extrapolating from the question of, okay, you know, what if we had a- A electronic cash system and the ability to sell things on the internet, then you would see some kind of, dark net market, which he called black net. So it was a kind of thought experiment, the technology to build it didn't really exist at the time, but he could see that the technology would likely evolve and then was able to comment politically about what that would mean. So it's kind of foreseeing Silk Road and its successors, of which I guess there have been multiple, so that was kind of interesting and You know, there are a number of technologies relating to that that people on the SaferPunks list were discussing. So here's one of the co-founders of the SaferPunks list and-" }, { "speaker": "adam_back", "time": "36:21", "start": 2180.66, "text": "it was quite interested in the other privacy technology that evolved, like, remailers for to claim, to sort of reclaim, Freedom of speech and freedom of association in the online forum. So in a sense, in the online world, typically you would end up with less privacy and rights that are typically protected by laws and regulations, but those would tend to get eroded by The fact that a lot of online information is logged by ISPs and as a side effect of the technology, so he and a number of people thought it was interesting to try to reclaim that and that the online world also provides You know, some rebalancing of power between the individual and the state in favor of the individual. So, I mean, I think he, you know, sort of helped, helped start this kind of conversation, and it was a lively discussion list. And he was, he was one of the more active posters, sort of bouncing ideas. So it was, it was a pretty interesting time, you know, a lot of rapid technology evolution and brainstorming about things that cryptography could do. And how that would be good for individual liberty." }, { "speaker": "stephan", "time": "37:32", "start": 2252.43, "text": "Fantastic. And if there were any key articles or pieces of work of his, you know, if there was one thing you, you would ask the listeners to read by Tim, what would you sort of point them to?" }, { "speaker": "adam_back", "time": "37:43", "start": 2263.34, "text": "well, he's written a number of th- a few things. So the Kryptonomicon is a pretty long- Kind of, summary of the ideas that he developed from the building blocks. So that's kind of, sort of fairly full exploration of Different things and implications, political, geopolitical implications and implications of the technology and how they fit together, and, also more recently. did a presentation at the, Pirelli Polis Hackers Congress, and that, that was based in Prague in Europe, but he, he was, via Skype or something, so, but the, the video is online, so that's, You know, probably a faster way to catch up with his thinking and reflections on how, what he'd been talking about has, you know, what parts have come to pass and his thoughts about that." }, { "speaker": "stephan", "time": "38:41", "start": 2320.82, "text": "Excellent. Okay, let's talk a little bit about Bitcoin privacy. So I know this has been, something you've been interested in for some time, obviously. There were some, there was some discussion online that I saw around this concept of pay-to-end point. could you outline a little- A little bit around, you know, what that is as a concept and how that might work if it were to be, you know, used with Bitcoin." }, { "speaker": "adam_back", "time": "39:06", "start": 2345.8, "text": "yeah, so I mean, the discussion-- so we ended up having a- Kind of in person, a couple days brainstorming, with a number of people interested in improving Bitcoin privacy on a practical basis. So" }, { "speaker": "adam_back", "time": "39:25", "start": 2365.34, "text": "during this conversation, we sort of, tried to look at CoinJoin and ways that CoinJoin could be used in wallets, per Pa- Or and the protocol, new protocols you might need to do that and, you know, how they would work with wallets. And it's quite difficult to, make privacy work because there's a lot of data that gets sort of linked as a side effect, and so Paychex Endpoint was One thing we hit on that seemed to be practical, 'cause we wanted something that was, you know, simple enough that we could implement it and have it not require, Base protocol changes, so not require like a softforking or change in the Bitcoin protocol, which would have a longer timeframe to see adoption. And so Pay-to-Endpoint was, was the idea that, when you are paying a service or a person, there's an opportunity to sort of do a coinjoin with the payment, and it provides ambiguity in part because, you know, normally speaking, I mean, so I guess I should restate what coinjoin is, so it's just a Bitcoin transaction has multiple inputs and multiple outputs, and people would naively assume that the inputs belong to one user's wallet Bitcoin being organized as coins, and then the outputs being the person or persons you're paying and one of the outputs being change." }, { "speaker": "adam_back", "time": "41:05", "start": 2464.68, "text": "So CoinJoin is the observation that, the, the coins that are being used to spend don't have to belong to one person, you know, so two people could make two different payments to two recipients and Join the transactions together. So then, to somebody trying to analyze the blockchain, they wouldn't know if it was one person paying two people or two people paying two people, or so it creates some, more ambiguity. but one challenge with that is the size of the coins, because, because the coins have a clear text value, the, the values can become unambiguous, you know? So if, if one, if there are two inputs and one coin is nine bitcoins and one of the payments is ten bitcoins, then you know that that coin couldn't by itself have paid that output. So it's probably the other, let's say the other input is twelve or something, then you know that the nine couldn't, couldn't have paid it in isolation. So if you tried to do a coin join in that scenario It doesn't, it doesn't provide much ambiguity. So Pay2end point is the observation that actually if the person you're paying, like, a web store or somebody else's wallet You're making a payment combined with a coinjoin, then it's, it's more ambiguous because the, the amount that's being paid is being subtracted from the payment, and somebody with a wallet might want to Reorganize their wallet, sort of, sort of combine change, as it were, which people do to sort of optimized wallets." }, { "speaker": "stephan", "time": "42:43", "start": 2563.44, "text": "Right, I see. And then another- Technology is obviously confidential transactions. Now, my understanding on that is that there are, depending on the way it's implemented, there are Some different trade-offs with the way that could work. So if it were to be merged into Bitcoin, we could have the amount of a transaction being blinded, but depending on the construction, we may have reduced assurance or comfort on the fact that there will only ever be twenty-one million bitcoins. could you comment a little bit on that? And, for Anelia, I guess has commented on that as that trade-off between perfect binding and perfect hiding." }, { "speaker": "adam_back", "time": "43:25", "start": 2604.6, "text": "Right. So There's a, sort of mathematical trade-off where you can't have perfect binding and perfect binding simultaneously, and so there are, It's, it's possible to choose which one though, and then there's some discussion about which one'll be preferable. And, so the perfect binding being that You know, you're, you're looking forward, casting forward to a hypothetical very far future where the cryptography we're using today, with the key sizes it's using, would be broken, most probably by extremely powerful computers or quantum computers at a scale that With current technology, we're not sure well, can be built physically even in the future, so sort of some question marks there. But if, if that amount of- Computing power becomes available, then hypothetically with perfect binding, they will be able to remove the binding and see that, this coin had this value. Now, With perfect blinding, on the other hand, when that cryptography is able to unravel things, you, you wouldn't learn which, You wouldn't know the amounts because all amounts would be equally possible. Now, the problem would be at that point that, people could forge amounts, and that, that would be bad for the coin's security. It would, you know, allow inflation or changing the cap and so on. So that's obviously something that would have to be fixed. And so that introduces a third possibility, which Tim Roughing called, switch commitments, which is you start with, perfect blinding and at a future time at which you start to question Whether the, quantum computers or algorithms would be getting closer to making it possible to decrypt this information, then you, switch To perfect binding, so you kind of get the benefits of perfect binding, in the short term, and then in the future you switch if, if that becomes necessary. And so they're, they're, I think reasonably accepted cryptography to do either of those two things using Pedersen commitments for perfect blinding and Elgamal commitments for perfect binding. And the other piece of technology which relates is Bulletproofs, which are sort of more compact, more space efficient. however, they, They don't, they're not able to provide perfect blinding, so, sorry, perfect binding. so they, they have inherently the, kind of inflation risk. So I would say that the technology is evolving quite quickly, but, you know, so new, new inventions are coming to light and bulletproofs weren't known about, you know, until they were invented. Also, in the last year or so. so we may see some further progress in, size and performance and assurance provided, but that's, that's the Kind of trade-off. And, you know, strangely there, there are different ways you can look at this. You know, if you take a privacy-first kind of viewpoint, you might say it's better that privacy is pre-pre-protected at all costs, but then in terms of the, you know, security of the network, if- You ke-- if people keep privacy, but the coin becomes unusable, that's a different problem in the future." }, { "speaker": "stephan", "time": "47:05", "start": 2824.74, "text": "Right. And, I suppose the other aspect is also the getting consensus on, on, you know, e-- which direction we go, right? Whether we go blinding or binding or for the switch commitments. is it possible that, you know, there may just be some nasty trade-offs either way, whichever selection, that make full privacy o-- for on-chain Bitcoin infeasible and just not likely to ever get merged in? Do you have thoughts on that?" }, { "speaker": "adam_back", "time": "47:32", "start": 2851.62, "text": "I mean, I think that the, you know, from, from what I've seen, Bitcoin users seem to be quite interested in improving privacy and fungibility of Bitcoin, so I don't necessarily see a kind of political will or like user intent or market interest problem And technically, I mean, w-with the elements and Liquid sidechain, we've deployed this technology for, so that people can use it actually on a sidechain to see how it works and look at, kind of look at the trade-offs. It's sometimes interesting to try things firsthand rather than, you know, just looking at papers and explainers, so you can kind of see it in action, and that's using the perfect blinding variant. But I think, I think the switch commitments are- relatively reasonable trade-off. probably the thing which kind of muddies the waters now is that Bulletproofs are quite a lot, more compact, and so it's kind of, more scalable. Yeah, so it's kind of a tr- attractive to, to consider that option as well. So maybe there's a way to sort of start with Bulletproofs, you know, perfect blinding aspects, and have a switch commitment to ElGamal commitment Later, because, you know, the prospect of, the cryptography being broken or sufficiently powerful supercomput- you know, quantum computers to attack this in, in any kind of short term seems quite unlikely to me. You know, I think you're talking decades or more Out and, you know, with the switch commitment, you, just want to see the system upgraded before that comes into effect, and you got to bear in mind that if we-- if quantum computers existed and were available That could, you know, undo these kinds of transactions, it would affect everything, you know, internet security in general, online banking, encryption, all, all kinds of things would change, right? So" }, { "speaker": "adam_back", "time": "49:38", "start": 2978.5, "text": "I think, we would know about it and we would have time to, to react and we'd see indications that quantum computers were, the technology was improving, which to date, you know- Has, there's been progress, but it hasn't been to very large parameter sizes and practical speeds and that kind of thing." }, { "speaker": "stephan", "time": "50:01", "start": 3000.94, "text": "Yeah, thank you very much for, out-explaining that. That's, a great explanation on that. and also you mentioned, your thoughts on the concept of sufficiently powerful quantum computing being decades away. Could you outline a little bit around that and then what the likely responses from a Bitcoin point of view would be, such as a switch? To post quantum cryptography, if, if it were to ever occur?" }, { "speaker": "adam_back", "time": "50:25", "start": 3024.88, "text": "Yeah. So, I mean, I think if, for people who aren't following the kind of, bit- bit discussions and the discussions about, you know, incrementally upgrading different parts of Bitcoin, they, they may think it's more difficult than it is to introduce a new signature scheme. So, you know, as a, as a point of, of, current topicality, the Schnorr signatures are under serious discussion, there are drafts and, implementations and work on sort of serialization aspect of the signatures and security of multi-signatures using them. And so I think it's reasonably Straightforward on the scale of Bitcoin changes to introduce, post-quantum signature scheme, such as Lamport signatures or other variants as a, you know, third signature option. So I, I presume that the, if we, presume that the Schnorr signature goes forward and gets introduced into Bitcoin like next year or the year after, that will be an ex-sort of a live example of Bitcoin having introduced A new signature scheme, and that same process and experience could be used to introduce a post-quantum signature scheme. So something similar to that, you can, you can think about the, switch commitment as well, that, you know, you start off with one scheme, and you anticipate that you may need to upgrade to another scheme in the future." }, { "speaker": "stephan", "time": "51:55", "start": 3114.51, "text": "Excellent. And also that brings up the question as well around, hypothetically again, in the future, maybe decades away, if there was a need to switch to quantum-resistant cryptography, would that mean that, say, some of the earlier inputs, so say Satoshi's coins, could they move?" }, { "speaker": "adam_back", "time": "52:15", "start": 3135.27, "text": "I think so. So John and Lau had introduced, an idea for migrating existing coins. Into the new signature scheme. So his idea was that basically, you, you would, you know, add a feature to Bitcoin to do a two-stage thing where the first operation is migrate the coins and the second one is spend them. So in the migrate the coin step You basically prove that you know the keys for the Bitcoin, but without revealing a signature on the key as such, and then use that proof to intro- to Record your post-quantum address and public key, and then you go on to sign with that using a post-quantum crypto. So there's, there's possibility for many of the stored keys in Bitcoin to make proofs about the, The public key corresponding to the address without revealing a signature. So that gives the, opportunity to sort of, make proofs with that and then migrate. and I think also as a general principle, you know, Bitcoin could migrate longer term to- Significantly different storage formats or representations. So today's blockchain is composed of Scripts and public keys and a ledger of the recording of transactions, but, you know, the, the future can hold other representations of the, of the data. So I tend to think of the blockchain is more of an artifact, sort of cryptographic artifact to prevent double spending. And so if its format evolves over time in ways that provide better privacy or fungibility, I think it's plausible that even if a, you know, a newer scheme some years into the future is found that is difficult to incrementally migrate, one could, you know, make a new, a new, a new chain chain using the new format and import the existing chain or have you know a migrate coin feature. So," }, { "speaker": "adam_back", "time": "54:32", "start": 3272.34, "text": "I think we're not, we're also not in principle constrained to, you know, stepwise simple changes. We can also contemplate a more major format change. So that, that generally means that, you know, whatever technology is Developed and evolved Bitcoin's UTXO set, current ownership set could in principle be imported into it." }, { "speaker": "stephan", "time": "54:56", "start": 3295.83, "text": "Fascinating stuff, yeah, that's really interesting. Okay, lo- I've got another hypothetical just around Bitcoin privacy. So let's just imagine, say, three years from now, you know, Joe Bitcoin, the average Bitcoiner, the man on the street who, who likes Bitcoin and maybe he's, he's making use of CoinJoin technology, he's using some kind of wallet like Wasabi Wallet. Wallet, he runs a full node at home and let's say Dandelion has been implemented and we've got some level of, you know, protection from network de-anonymization attacks, we've got Lightning Network operating a little more fully and obviously Lightning Network has onion routing, so that helps improve privacy, and then I suppose just generally individuals out there who are just using a smartphone wallet, well, they're using Utrino-style wallets, they're a little bit better for privacy than the current-day SPV mobile wallets. Do you think this might be, quote unquote, enough for the average Bitcoiner just to give them some level of basic privacy?" }, { "speaker": "adam_back", "time": "55:56", "start": 3356.23, "text": "yeah, I mean, I think, a lot of, you know, there's a lot of activity in Bitcoin development in general at the moment, but there are significant things happening for privacy as well. So Lightning is, one thing you mentioned, ano-another thing is Dandelion, which is a kind of main chain, better privacy for when the transaction is- Sent into the network so it's less obvious which IP address originated the transaction, and so it doesn't seem like so far that there's a silver bullet, so something that f-improves fungibility and privacy completely in one shot. So it seems like we're in a kind of incremental Improve things where we can and, move the ball forward. So there's a lot, a lot of things come together there, and Lightning is quite good for that too, in part because it sends less data to the blockchain, so blockchain being the broadcast part of the picture with Lightning, and then the actual payments being point-to-point. So if there's a high degree of recirculation, which- People hope to see with Lightning, then much of the transaction information never actually makes its way onto the blockchain. So, you know, so you can think about the privacy as a side effect of the scalability. So if the rest of the network doesn't see the transactions, they can't analyze them basically. They only see the netting when a transaction is eventually, when, when a channel is eventually closed. So I mean, I think things are moving in the right direction, and, you know, hopefully we also see some kind of, More powerful upgrades like confidential transactions in, in the midterm as well, which, sort of interacts nicely with other things in improving coin joins, for example, they become much easier with confidential transactions because you don't have to worry about the, making sure the amounts. Add any ambiguity?" }, { "speaker": "stephan", "time": "57:56", "start": 3476.06, "text": "good point, good point. Okay, excellent. So look, we're, we're in a time where, you know, some people have a bear market blues, but at the same time, there's a lot of exciting things going on from a, you know, technical fundamentals point of view. do you have any other comments around, you know, technical fundamentals of Bitcoin?" }, { "speaker": "adam_back", "time": "58:14", "start": 3493.98, "text": "Yeah, I mean, I think the, you know, for people in the technical domain, at startups and individual contributors and people working on different protocols like, Lightning, Neutrino, sidechains like liquid elements and rootstock. there's actually, a, a lot of development going on both at the protocol level and in the application level. So, you know, for Lightning, there's a lot of application level work going on in wallets and, tools to integrate into web frameworks And people writing applications like games or websites that, start to show the new capabilities. So if you think about the micropayment level payments as being, A new capability for Bitcoin because it, it allows faster, sort of more instantly final. Like Bitcoin transactions can take twenty minutes or more to be final. randomly. so Lightning also supports much lower payments because, you know, there's essentially no practical minimum fee, it's, such a small transaction. so I think, you know, overall the pace of innovation and application development is higher than ever. So if you look at that as a technology fundamental, there's a lot of, Momentum and technology adoption and building out for sort of next phase of adoption. And also I think something, you know, there's quite a lot of things happening in the kind of exchange Ecosystem with, for example, Liquid providing more exchange settlement benefits, more privacy, faster clearance, better liquidity, and there are a number of co-companies entering the space with, more fiat coins, different types of market participants like, established players like Ice, their backed service. And Square as a retail payment platform that is, that has added Bitcoin, and Robinhood and Abra. So I mean, there are many new services coming online and markets opening by new, new entrants into the Bitcoin And crypto asset space. So I think if you look at it from a technical fundamentals point of view, everything is, on a very strong trajectory right now. And I think also people kind of, should generally zoom out, not, not focus on short-term price too much, because it's, it's a kind of recurring pattern with Bitcoin and crypto assets that the price goes through high volatility and you know, over a three year period, for example, as, as far as I understand, that Bitcoin is the, you know, highest return asset class on the market, you know, amongst stocks, pink sheets, a-anything basically. So, you know, to look at something that is down You know, sort of eighty percent range when it's, in, in the, within twelve months ago, it, it was up. A hundred times from three years before that is a kind of, you know, cherry picking sections of the graph. If you zoom the graph out, it's generally Increased over a long period, and that shouldn't be something, an unusual concept for investors because, you know, it's a common thing if you talk to an advisor or anybody in the investment space or experienced, like, personal experience trading or making investment decisions for themselves that, you know, people will ask, \"Well, what-- well, how quickly do you need access to the funds?\" If you need them within a month, then you probably shouldn't put them in the stock market 'cause it's short term volatile. But if you're willing to, to, to, you know, put them in there for a minimum, sort of three to five years, then the stock market statistically has outperformed For example, deposit accounts and things like that. So, you know, Bitcoin has the same kind of, volatility trade-off. Over, over the long term, it's been, essentially the best return asset class, but over the short term, it's extremely volatile, and those things are related. There's a kind of risk-return trade-off." }, { "speaker": "stephan", "time": "01:02:54", "start": 3774.7, "text": "Totally agreed on that. Look, I think that's, so pretty much all we've got time for, but Adam, I just wanna thank you. It has been, an honor to host you for a discussion. I've enjoyed this discussion so much, and I've honestly learned a lot. I think what you guys are doing at Blockstream is really, you know, it's innovative, and you're really building out the infrastructure. You've got, you guys have got your eyes on the prize. So I'll just make sure all the listeners know where to find you. So the links for Blockstream Satellite page, that's blockstream dot com forward slash satellite. Obviously, adam three dot, adam three dot us is your website, and Adam three US is your Twitter handle. Have you got anywhere else that you would like to point the listeners towards?" }, { "speaker": "adam_back", "time": "01:03:37", "start": 3817.79, "text": "those, those are good places to go. There's also elementsproject dot org, which is the open source, GitHub repository for much of the technology we've talked about." }, { "speaker": "stephan", "time": "01:03:48", "start": 3828.68, "text": "Excellent. Well, I'll put it, I'll put the links for that in the show notes page. Adam, thank you very much for coming on the show. Likewise, thank you. Check out the show notes for this episode on my website stephanelivera dot com. If you enjoyed it, remember to subscribe so you don't miss out on the next episode, and please share the podcast with your friends. You can also follow me on Twitter, my handle is at stephanelivera. Thanks for listening." } ] }