CoinGecko Podcast - Bitcoin & Cryptocurrency Insights

Working Towards Permanent Information Storage with Sam Williams, CEO and Co-founder of Arweave - Ep. 30

March 23, 2021 Bobby Ong Season 1 Episode 30
CoinGecko Podcast - Bitcoin & Cryptocurrency Insights
Working Towards Permanent Information Storage with Sam Williams, CEO and Co-founder of Arweave - Ep. 30
Show Notes Transcript

In this episode, Bobby Ong, co-founder of CoinGecko is joined by Sam Williams, CEO and co-founder of Arweave. Bobby interviewed Sam on the background of Arweave, its AR drive, its AR token and the use cases, as well as Arweave’s plans.

[00:00:02] Intro
What is Arweave?
What happens if the AR token price falls?
Differences between Arweave and the other decentralized storage solutions?
Where does the idea to build a permanent storage solution come from?
Use cases for Arweave
Can private information be stored on Arweave?
[00:30:55] Profit sharing communities
Tokenomics behind the AR token
[00:39:56] Upcoming plans for Arweave

Quotes from the episode:

“So being able to, for example, commit a blogpost and know that that blogpost, your ideas, your writing can be perpetuated literally far, far past your death, like far into the future... That's a pretty cool thing to be able to do for like 0.40 cents per megabyte.” [00:17:09]

“One day someone is going to turn that off and then eventually your NFT is going to be worthless. That's crazy. It doesn't need to be this way. Just do it on Arweave and then it's there forever.” [00:21:15]

“The idea is that with Arweave, because you can start smart contracts from the tags associated, where they file. Then now you can associate the contract of an NFT directly with the file of NFT in one thing, and then refer to them by the same transaction ID.” [00:33:16]


Arweave -
CoinGecko -
Arweave (AR) -

Social Media



Bobby Ong [00:00:00]:
Welcome to the CoinGecko podcast. I'm your host, Bobby Ong. Each week we will be interviewing someone from the blockchain industry to learn more about this fast moving cryptocurrency economy. If this is your first time listening then, thanks for coming. The CoinGecko podcast is produced each week to help you stay ahead of the curve. Show notes can be found at I highly encourage you to join our newsletter, where we send out top news in a crypto industry every Monday to Friday. Come back often and feel free to add the podcast to your favorite RSS feed or iTunes. You can also follow us on Twitter and Telegram at CoinGecko. 

Welcome to the CoinGecko podcast. For today's episode, we have Sam Williams, CEO and co-founder of Arweave, a decentralized storage startup. So very happy to have you on the show. Welcome to the CoinGecko podcast, Sam.  

Sam Williams [00:00:51]:
Thanks for having me. It's good to be here. 

Bobby Ong [00:00:53]:
Yeah, to start things off, Sam, maybe can you give us at CoinGecko,the community here, a simple explanation of what Arweave is and would like to know more about it. 

Sam Williams [00:01:04]: 
Yeah. So Arweave is a permanent information storage system, I guess your audience is very aware of blockchains. And so they're probably aware of the idea of content storage. Storing something in the blockchain itself. We can think about Arweave basically as on-chain storage that scales and has appropriate incentives to me that were literally permanent. So the way it does that on technical side is a new mining algorithm, which rewards people for storing data, rewards them more for storing data that is stored less. And that kind of auto levels the storage over the disc space that is provided. And then on the economic side we have, what is essentially an endowment structure. So you paid for essentially 200 years worth of storage upfront when you use Arweave, which sounds expensive, but it's about 0.40 cents per megabyte, approximately. And then over time as the cost of storage declined to essentially gain interest in the form of storage purchasing power on that initial 200 years. Yeah, the grace period you're using it essentially. Just like with a, you know, university endowment, you make some interest on the principal that's put aside and you use the interest to pay for people to go through, you know, get degrees. Yeah, it's the same principle, but just applied to data storage. 

Bobby Ong [00:02:09]:
So, yeah, I kind of get the idea, where instead of paying for one year or one month, you pay for 200 years upfront and then you got to store your storage essentially, almost permanently. I guess my question is, how do you come up with the 200 years number? Like why 200? Why not 100 or 500 and I guess this is too far to ask but what happens after 200 years? Like do they disappear? 

Sam Williams [00:02:30]:
Right. Exactly. So over time, the cost of storage declines, and you only take from the storage endowment, what you actually need to pay for storage. So essentially what you're getting is interest on that 200 years in the form of storage purchasing power over time. So imagine, and this is the average rate, the cost of storage declines at a rate of 30% per year. That's what it's been on average for the last 50 years. So, year zero, when you put it in, we've got 200 years with the storage purchasing power up front. At the end of the first year, now we actually have 260 minus one years worth of storage purchasing power for the principle that we've already put up. And you can see as this compounds, you actually end up with more storage purchasing power, not less. So the question of why 200 and it's actually a reciprocal, which is why 0.5%. So why expecting a rate above 0.5% for storage cost to clients? It constantly comes down to conservatism. Like we think that blockchain protocols are the successful offer a zero to one service, something like Bitcoin, for example to digital scarcity from zero to one. Ethereum took smart contracts from zero to one. These are not just copying business models that you can do in a centralized world in the decentralized world. These are giving you the ability to do something you just couldn't do before. And we think that permanent storage is another one of those.

It's something that you couldn't trust a centralized company to give you at any point in time. You can however, trust a protocol to give you this service because it won't change. And the participants can come in and out over a very long period. And there's no corporate governance to worry about. There's no profit motive or incentive for the network exactly in and of itself and so on. So it's the kind of thing that you can allow a network to do, but in order for it to takeoff, you need it to be sufficiently conservative. And when you say, geared towards achieving it's one fundamental goal for people to be able to trust it. So we could have probably set it at 20 years and that would probably be okay. But if we'd done so then people might not trust, it might not catch on. You might not actually about to achieving one core important, you know, zero to one, if you will, change that it allows. So we picked a much more conservative number. 

Bobby Ong [00:04:31]:
I guess the question then is if it's conservative, it almost feels that if I want to store something on Arweave now, I'm sort of overpaying for the endowment fund. Right. And this cost of 200 years, someone's going to make a profit from this, right? So who is going to take the profit from me overpaying on this endowment fund? Is it the protocol, the miners or who's gonna take this profit here? 

Sam Williams [00:04:57]:
Yeah, it's an interesting question. Well, at the moment, what happens, what will happen and I don't think this will ever change is just the funds just go into the endowment, where they just stay. Forever, basically. 

Bobby Ong [00:05:07]:
Okay. So the protocol has this endowment, which controls all this funds, I suppose.

Sam Williams [00:05:11]:
Exactly. Yeah. So it's kind of like burning with the caveat, you know. [inaudible] if for some crazy reason in the future, the storage costs actually increased then the amount taken from you [inaudible] minus could also increase as well. So "burns*" if you will. And [inaudible] I probably gets benefits token holders. 

Bobby Ong [00:05:31]:
And the endowment, I mean, when I pay for storage on Arweave I pay them in a form of AR token, right? And these AR token fluctuates with the USD price. So if the price of Arweave goes up in the future, that means, yeah, your endowment will have a lot of money and then... 

Sam Williams [00:05:49]:
That's already happened a lot. 

Bobby Ong [00:05:51]:
Yeah. I've been monitoring [inaudible]. 

Sam Williams [00:05:53]:
The endowment has like, I think at this point, maybe $250,000 worth, $300,000 worth of storage, worth of USD inside it to pay for storage. And at the current time we're talking about like five and a half terabytes or something, which in temporary storage terms not a lot, but in permanent storage terms actually it's quite a lot. But even still, you know, now the expected storage [cost decline rate], I calculated this recently. I wish I could remember the exact number, but it's the expected storage cost decline rate to keep this to work perpetually at its current rate, is something new or it's a 0.01% storage cost decline per year and then the thing will run forever. But yeah, there's definitely another reason that we chose 0.5%, which also gives that reciprocal 200 years, is because it just gives a lot of slack for potential changes in the token price, which we all know like early crypto networks, there are changes.

Bobby Ong [00:06:47]:
Do you see of any problem, like, when token prices rising, like everything works out well, right. But touch wood, but like, what happens if the token price goes the other way around instead then like [inaudible] 99%, for example? I hope it doesn't, but I mean, if it does then will it still work out then?

Sam Williams [00:07:02]: 
Yeah. So I think 99% is a little bit extreme, but like 90%, who knows, yeah. That's why we use such conservative value. So that's just another reason why basically. Yeah, this point, obviously, you know, there's so much in the endowment and it's interesting, cause it goes both ways. So you put a lot of tokens in the endowment and then the price moves up and now those tokens are worth even more, but obviously if they go down and then you still have this extra buffer that you've generated. So now it's 0.01% can afford to go to 0.5% and again or whatever it happens to be. It's an interesting question. 

Bobby Ong [00:07:33]:
Cool. So decentralized storage is not essentially a new concept. I mean, I've been in this space for some time and it's been attempted by several teams a few years ago, since 2014 or so, or 2015, like Siacoin, Maidsafecoin, Storj and more recently Filecoin with IPFS. I think you kind of mentioned briefly that the differences between this different decentralized storage is that Arweave is kind of going up to the permanent storage solution, whereas I believe the others are not going after the permanent storage, they go more for a fixed time storage solution, but I would love to hear from you, what are the differences between Arweave and all the other decentralized storage solutions?

Sam Williams [00:08:08]:
I think you put, what did they say, knock the nail on the head or whatever it's. Yeah, it is a permanent storage system, which none of the others offer. And no one is actually in this lane at all. Frankly I think it's kind of surprising because it's just such a big, obvious and big, opportunity. I personally find that, so I was invested in the storage ICO actually, way back in the day. And I think like temporary storage system interesting. They're not zero to one. Right. They're going to have to go and try and compete in Amazon on cost. And that's going to be really, really tough, like really tough. I'm not sure that that's a winning strategy for decentralized networks. I think you have to do something new and that's kind of why. Yeah. Arweave is focused on this permanent storage angle. And then, also on top of Arweave, which we haven't really covered yet, is what we come up with. So it's like a collection of protocols basically expose all the information in the Arweave to people's web browsers. And that's like a whole different game. Like it's literally permanent decentralized web that lives inside the system. Yeah. I don't think [inaudible]. 

Bobby Ong [00:09:07]:
So how'd you get this idea to build a permanent storage solution? Everyone's kind of building on this temporary or fixed time storage solution, but you came about and say like, "Let's build something truly game-changing, a permanent storage". Like no one's ever thought about it before. How do you get idea? Why do you do it and all? 

Sam Williams [00:09:24]: 
Yeah. Well if you've ever read 1984, yeah, you know, the idea of the memory hole. Okay, for the listeners that haven't read the book, basically the book is about a dystopian future or would have been the future, it's actually written in 1948 and it was talking about the future as 1994. And dystopian future in which authoritarian regimes essentially around the world. And they did so largely through the control of information. And part of that control of information was controlled of access to information about the past. And it's through this mechanism, they're essentially allowed to control the way people think about the present and subsequently the way that they act in the future. It's just like a really key insight. I think from that period of time is that actually how you control large numbers of people is not necessarily through force. You do so through control of information. And you know, this book was trying to exemplify that idea. And so the main character in the book, Winston and also the smallest sub-unit of Arweave token is called the Winston, after this character.

The main character in the book, his job is to censor records of history by throwing them down what he calls, the memory hole, which is essentially a fire pit. So his job is to collect up all the records of a past historical event that is now unfortunate or the regime doesn't want people to know about and to burn them all. Yeah, we essentially wanted to make that impossible. And we saw that the blockchain technology, you quite a lot of the way there, you got replication of a piece of information all across the world without a central leader. And you've got the means through which to add data to that. But of course, when you didn't have was the means to add arbitrary amounts of data and that's the incentives to make that storage same and consistent. Like people think at the moment,  even if they do store data on Ethereum, some NFT people try and do this. Yeah, like I know it's just crazy. It costs like thousands and thousands of dollars for 32 kilobytes, but whatever people try. Yeah, there's no actual incentive for people to store that data. The miners are only incentivized to store just the minimum amount that is required to take part in the mining process. Even if it did blow up the size of the chain to store more stuff in it, then people aren't incentivized. 

So anyway, we figured, okay. We saw the world moving in this direction where it would make more sense to have a decentralized reliable archive of history. And we saw the kind of technological components that would be needed to do that. And the first thing we set out to try and do is actually build a newspaper archive. Yeah, it was just basically trying to get newspaper articles and make them live forever. So we didn't forget history, but of course we realized pretty quickly, as soon as you solve those problems for a newspaper archive. Well, now you have a new kind of archive you can use for everything. Yeah. and that's kind of the Genesis of the idea.

Bobby Ong [00:11:57]:
Interesting idea how you taught about this dystopian future and how governments will try to censor things, and you try to build a web where someone in the future can sort of reference things from the past and see for themselves that this is what it is. Kind of like the internet archive in a decentralized manner. Yeah. 

Sam Williams [00:12:13]:
Yeah. [inaudible]. Yeah. Interesting. 

Bobby Ong [00:12:20]:
So you mentioned briefly just now about how the other decentralized storage solution were kind of like competing against trying to compete against Amazon S3 and that's kind of like, not a good idea because there's no way one can drive efficiency gains and compete against Amazon S3. So maybe to give us some perspective of cost, I'm not so sure if you're aware of how much the others are charging like how much it costs to store an IPFS versus how much it costs to store on Amazon S3, and then how much does it cost to store permanently on Arweave, for example. 

Sam Williams [00:12:49]:
Yes. The others will be somewhat cheaper. I think people have done like the crossover calculations. Like at what point does it become cheaper to store on Arweave that kind of thing. It's like a few years. But I really think that the benefits offered by permanent storage for information. So being able to, for example, commit a blogpost and know that that blogpost, your ideas, your writing can be perpetuated literally far, far past your death, like far into the future. Presumably would forgotten after the last human or human-like intelligence has gone. That's a pretty cool thing to be able to do for like what, you know, 0.40 cents per megabyte. And for a normal blogpost. And this is an interesting question, like, what data doesn't make sense for normal blogposts? That's a power that Arweave gives you for, like, you know, let's see, if that blogpost is 30 kilobytes or something, in 0.001 cent, someone's gonna check my Maths. 

Bobby Ong [00:13:38]:
Yes, it's cheap. 

Sam Williams [00:13:39]:
With tiny, tiny fraction of a cent. Yeah, so that's actually a pretty damn good deal. But of course, if you've got like, I don't know, an archive of really, really, really unimportant video, something that's like a very, very large file format and you want to store a lot of it. And the density, what we call the information value density is low, basically. Then you might not want to store it on Arweave. You might want to store it somewhere else. And that's totally okay by us. Please go store it on Storj. My Storj tokens, which I still have or rather haven't picked up yet. Anyway. Yeah, it makes total sense to use another kind of system for that. But there are many, many things we would argue, almost everything that people publicly commit to the web that they really want this component of like spreading it as far and wide as they can. If only the functionality were available to them. And that's what Arweave does. 

Bobby Ong [00:14:29]:
I'm just curious, right, for Arweave, like, do you have any stats on how much data that is currently store on the Arweave network, now?

Sam Williams [00:14:37]:
Yeah, it's under like five and a half terabytes or something, but the difference is that this is permanent data. They're like replications of this far, far more than you would expect in a normal system. So I would say at least until the like 250 to maybe a 1000 range at this point, quite possibly a lot more than that. Yeah. But also again, it's about the information density. So I would say this one is probably 15 million, maybe 20 million different pieces of data on the network as well.

Bobby Ong [00:15:04]:
You mentioned information density, but I don't think I really understand what you mean by information density. So maybe you can elaborate a little bit more about that. 

Sam Williams [00:15:10]:
So you've got a file that is one terabyte and it's value to you might be $5, right? You've got another file, like this blogpost is 30 kilobytes and it's value to you is a few hundred dollars. So you can take these, these factors and you can calculate it approximately a value density for the data, if you will. Does that make sense? 

Bobby Ong [00:15:31]:
So, when you store data, you can sort of put a value on how much you value the data or... 

Sam Williams [00:15:36]:
It's an innecessary heuristic that people go through in their head. So it might not make sense if the value data density is low, for you. You'll instinctively be like, I don't want to pay that price because the files aren't worth that much to me, basically. But if it's something like a blogpost, then the value density is much higher and you see like, yeah, that's like, you know, hundredth of a cent or whatever it was, is a great price to store this thing forever. Do you know what I mean? 

Bobby Ong [00:15:58]:
Okay. Yep. Yep. And I'm just curious, right? You mentioned about five and a half terabytes of data that's been stored on Arweave right now. I'm just curious, so who are the guys that are storing data on Arweave now? What kind of use cases have been shown to be using Arweave now?

Sam Williams [00:16:15]:
Yeah. So one of the first ones that started getting picked it up most recently, well, I'd say there's three key areas right now. People starting NFTs. It's one that's growing excessively rapidly. Like there's a real growing understanding that like, "Hey, if you've got an Ethereum contract, which has got a URI in it, right, and that URI is stored somewhere that is not permanent, then eventually inevitably, all of them will break. Like they are all stored on someone's centralized server. And one day someone is going to turn that off and then eventually your NFT is going to be worthless. That's crazy. It doesn't need to be this way. Just do it on Arweave and then it's there forever. It makes sense. People are starting to do this on mass very, very rapidly, which is really cool to see. Another one is actually chain storage. So one of the interesting prompts you have with blockchains is that they're essentially ever-growing ledgers that don't have any incentive to store as I was describing before, actually. This is kind of a problem. Eventually you get to this point where the chain is just too big. I was actually just talking to someone, they're like, the Terrachain or something like this, I believe it's like 200 gigabytes and the software after syncing it is not necessarily so optimized for giving me that number of transactions. And so it's taking like a year to sync to the latest block. That's a bit of a problem. So what you can do is you can store all that data on Arweave and you can have them nodes very, very rapidly collate off the Arweave network where it's permanently stored, incentivize forever and sync extremely quickly with the network.

So that's pretty useful. That started with SKALE, I think were the first to do that. And then very rapidly after that, it was Savannah. And then after Savannah, it was Polkadot and now there's like three or four more people we haven't yet announced that have come on board to do this. And this has grown very rapidly into uh, projects on top of the network, actually a startup called KAI. And KAI, basically the idea is to build permanent verified data feeds. So essentially they were just trying to solve this problem with like we put together a co-funded grant with Polkadots where we're saying, "Hey, let's get all the Polkadot data onto your Arweave". Make sense. It's not so complicated. And they're like, "Well, what if the person that's running this bridge is actually malicious? How are we going to know that they're not tampering with the data". Then they build have this really elegant solution using SmartWeave, the smart contracting platform on top of our, to reward people for essentially running these archiving nodes and verifying each other's data. And so now you can get these like verified data feeds of almost anything. So that's really, really powerful. Pretty exciting to see. The third component is people building what we're calling profit sharing communities. This is something in the Permaweb ecosystem. So in the stack of protocols on top of Arweave that expose all this stuff to people's web browsers. I think this is super, super powerful, and we're only seeing the beginning of it, but it's starting to take off very, very rapidly.

They've only been, this ecosystem has only existed for like six months. And I used to count how many dollars have been committed to funding these rounds of that building. It started as like, you know, half a million and it was a million very rapid, again, it was 3 million, now it's 5 million and it's like, it's increasing at an exponential pace. So that's pretty cool. But the basic principle is building decentralized web services. So services top to bottom. They're running in a decentralized way. You can imagine like a decentralized Twitter, for example, where the community owns the profit created by the application and the entire thing runs on decentralized infrastructure. I think one of the things that is kind of missed in the crypto community sometimes is actually a lot of these dApps that we use. They like decentralized one small component and everything else is centralized. These are the opposite of that system. This is like taking it to the nth degree, what if absolutely everything ran on like censorship resistant distributed infrastructure.

Bobby Ong [00:19:42]:
Interesting, the three main use cases that you mentioned. NFTs is like, a lot of people don't realize this, like you mentioned like this the images are store on some Amazon S3. And at some point somebody tried to rug some of these NFTs and they replace all the NFT with actual rug images. That was quite hilarious. I mean, we minted some NFTs before this and we were kind of studying the architectures, like what we found was that yeah. I mean back then it was at least kind of a pain trying to put onto IPFS, so we kind of, okay, let's just put it on S3, just put it out and try out and experiment with NFT. But we see how this was unknown to many people like, where is the actual image stored and I'm glad that people are talking about it right now, but we saw this a couple of years ago. 

Sam Williams [00:20:19]:
Yeah. It's rapidly becoming, I mean, we've been chatting about this for a long time for it's rapidly becoming like the known thing in the industry. There's this effect called like an information availability cascade, where like it goes from suddenly almost no one knows about an idea, a few people know it and then very suddenly everyone knows. It kind of going through that mode. Yeah. Pretty interesting to watch. 

Bobby Ong [00:20:40]:
Yeah. Interesting use cases on the nodes as well. Cause like yeah, it is kind of, it's very painful to try to sync a full node and I can imagine that moving forward, like instead of like syncing on your local machine, you probably sync it to a decentralized Arweave note, which constantly gets added with new blocks and then you can kind of [inaudible]. And then in the third one was the profit sharing community. I have some questions on that, but keepI keep it for a bit more, but let me ask you this question next. Do people store like private information? Like I know you were sharing me some of these decentralized application built on top of Arweave. One of them that I was looking at, was like a Google drive wrapper, like an app that looks at Google drive where you can upload like private documents and store them permanently. 

Sam Williams [00:21:21]:
ArDrive. Yeah. 

Bobby Ong [00:21:21]:
ArDrive. Yeah, that's the one. Can you store private information, I dunno, encrypted private information on Arweave? Is that a good idea or that's kind of a pretty bad idea to store like private information on Arweave?

Sam Williams [00:21:33]: 
This is an interesting question. There's a lot of different schools of thought. Like obviously you should not be storing private information publicly permanently on the Arweave. That's like saying, you know, look before crossing the road but whether you should store Ciphertext on the Arweave, my personal opinion is if you look carefully at the way, it is architected, pretty much anyone can get access to your lines of communication. Right. And that's sending huge amounts of Ciphertext, and then you're [inaudible] all the time. So really the difference between that and storing the information on ArDrive in a very heavily encrypted way, not actually that high. Yeah, that's what I would say about that. I think that it's complicated and people are going to have different risk models using some kind of quantum resistant storage encryption system for your stored data makes sense. But that only pushes the question back further, basically Yeah. So I think that comes down to people's individual risk tolerance as to what makes sense for them. 

Bobby Ong [00:22:25]:
Yeah. I think at the end of the day, it's that, how strong is your encryption, like what will be the future quantum, like you say quantum computing can it encrypt, can it decrypt whatever, can it brute force any of the current encryption technology in the future. And essentially if you use ArDrive, you are trusting ArDrive with the encryption. What if our driver is malicious, for example? 

Sam Williams [00:22:46]: 
I mean this is one of the interesting things about profit sharing communities in decentralized Permaweb apps is that basically they are permanent by design for the utility that they offer. It has these quite interesting effects, like you can audit the code of ArDrive once and have it associated with the TXID so it sent to the network. And then if you go back to that TXID and that TXID has been audited and you're comfortable that it is safe and is doing what it's purported to do, then you can be confident that that service will never change. This is kind of vaguely useful in the case of ArDrive. But really very significantly useful in the case of for example, a email system, an example I always use. So when you sign up for an email account with Gmail, you basically sign over your ability to move away by having your identity [inaudible] what you give to other people. So your email address you give to other people, now you're stuck with Google. And they can change the service they're offering to you all the time. Same with Facebook and so on. Like once you get locked into the service, then the company can basically do whatever they want. And you've got to remember that these companies aren't necessarily incentivized by the things that make the customers happy or successful.

This is really, really dangerous. I would even argue that it's like the hidden danger with the Web 2 world is that basically the incentive structure is that you track people into using a new platform and people bankroll this process. And then once they're all in the new platform, they can't really easily leave. Then you turn on the money tax, one way or another. And this normally means that the quality of the service rendered to the users, the clients yeah, and people ended up being somewhat exploited by the system. Whereas on the Permaweb we can do something much, much different. We can say the relationship between the developer and the user is fundamentally, the developer can release new versions of the software, but those new versions must increase. There must be strictly better than the prior versions. Otherwise people just don't update and they continue to [inaudible] software. So the trust model and the relationship that was just fundamentally different. I think that's actually really powerful. 

Bobby Ong [00:24:48]: 
Let's jump into the profit sharing communities and I understand that some of these profit sharing communities, they issued of profit sharing tokens to the members. So I guess my first question, right, are the profit sharing communities similar to decentralized autonomous organization, DAO? They sound the same to me conceptually, but I wanted to hear from you. And then after that, maybe can you share some examples of interesting profit sharing communities that are in the Arweave community?

Sam Williams [00:25:13]:
Yeah, absolutely. Okay. Are they the same as DAO? They are like DAO plus profit sharing basically. In fact, I used to call them, before we did some branding on it, I used to call them psDAOs, P-S-D-A-O. Because basically the same principle, you have a decentralized autonomous organization, but the critical part is that that autonomous organization, the members are, [inaudible] giving reward direct [inaudible] from just holding the tokens when people use the application. And so now the profit sharing tokens associated with the application become a, you know, API yielding asset for them. And they can just see, like, if I hold this for a year, how much relative to my investment, am I going to get back as essentially interest? Yeah. So the profit sharing in nature rather than accruing value to the treasury. Now what are some interesting ones in the system? There's so many. But I would say for example, hard drive is a great example, to be honest. It's just a permanent decentralized file sync application. You can easily replace Dropbox and Google Drive for most people. And if you own the token, then you take part of a 20% tip that's added when people upload data. Very simple, very effective, generates a lot of revenue. And take another somewhat, what would you say, their self-referential because [inaudible] itself a profit sharing token exchange. 

One of the things it does is it allows people to issue images as NFTs or what we're essentially calling atomic NFTs. So, this is a little bit novel and interesting, but basically the idea is that with Arweave, because you can start smart contracts from the tags associated, where they file. Then now you can associate the contract of an NFT directly with the file of NFT in one thing, and then refers to them by the same transaction ID. So they have, for example, an atomic NFT exchange, basically. That's a little bit like OpenSea mixed with Binance, somewhere in the middle. Or you know, OpenSea and Uniswap, perhaps. And if you own the [inaudible] profit sharing token, then you get a small tip every time someone takes the transaction. Or it makes a transaction. Now, suddenly you financialized in a very real and, literally make real time financialization of the profit created by web applications, which we think is really powerful. Now as people just buy into the profit streams created by the applications that they like, when they do so by supporting the things that they can [inaudible], they get a portion of the profit in the future. And the founders the team get another way of financing the development of the thing they're building. 

Bobby Ong [00:27:41]:
And what about the Arweave token, like what's kind of the tokenomics behind the AR token?

Sam Williams [00:27:47]:
Yeah. Well, I mean, it's pretty simple in some way. You use it to pay for storage, but when you pay for storage, instead of the token going directly to a miner, like you would in Bitcoin or Ethereum, Or actually before the new Ethereum upgrade, but it goes into the endowment and then the endowment, essentially, [inaudible] release from the endowment, when the block reward doesn't cover the cost of paying for storage in the network or on its own. So, essentially right now you're not taking any tokens from the endowment because the profit isn't high enough. And that gives us time to basically bootstrap that endowment. And then over time you just take tokens from the endowment relative to the amount required to pay for storage, plus some profits for the miners.

Bobby Ong [00:28:27]:
So the miners, I mean, for Bitcoin miners, they run ASIC machines, for Ethereum miners, they use GPU machines. What about Arweave miners? Do they run like massive data warehouses with terabytes of storage, or kind of miners are there for Arweave?

Sam Williams [00:28:41]:
Yeah, certainly terabytes of storage. So it's a mixture of compute and execute. You store some and you also do some computation. It's actually what we call a proof of work sandwich, this called. Technically it's for us, to people outside of the deep tech world, we just called it a proof of work sandwich. What's happening is like you are doing a piece of computation and the output of the computation is a random place in the network that you have to prove that you have access to. And then you grab that data and you package it again and you hash against [inaudible] compute and you check the output of that, to see if that essentially satisfies the difficulty required. So if that is a winning candidate that you're trying to mine. Yeah. And so this is the system. 

Bobby Ong [00:29:20]:
And then I suppose they difficulty and then every block can only be mined by a single miner, I suppose based on the difficulty level like Bitcoin or yeah? 

Sam Williams [00:29:30]:
Yeah, it's just like Bitcoin in that way. So, I mean, technically you could mine two at the same height, very occasionally happens, but again someone would mine one on block, not one on top, and then you get a fork of it.

Bobby Ong [00:29:40]:
Yeah. What about the inflation model for Arweave? Like Bitcoin started with 50 Bitcoin per block every 10 minutes. What about Arweave and how does it taper off to asymptote, for example, like a max supply of AR token?

Sam Williams [00:29:52]:
Yeah. So the maximum supply is like 66 million. There's a few million left to mine, something like that. It's, fairly low inflation system. 

Bobby Ong [00:30:01]:
66 million and about 43 million has already been mined. So maybe roughly three quarters have been mined. And we're left with the remaining one quarter that we mine in the next, I don't know how many more years? I'm just curious. Yeah. 

Sam Williams [00:30:12]: 
It doesn't stop. We use an exponentially decaying curve. So we actually recalculate the block reward per block rather than halving. We actually think halving is like a super negative component to having a network like this. It just makes a lot of uncertainty in the markets, like what will happen after the halving? Who knows? And you see it in Bitcoin every time people start to panic. I mean, you could argue maybe it's good also because it brings some attention to the protocol. Like I dunno. But we try and keep things stable, and predictable, and unsurprising. And so, yeah, we just have this like decaying curve per block. 

Bobby Ong [00:30:45]:
Yeah. Because like the halving is such a big change in reward for the miners. It's like first you got like 50, and then 25, it's basically half. Like there's always a question what's going to happen to all this hashing power. Will miners stay on or it's no longer profitable and they will remove everything. And then difficulty comes in and then like, is the networks still secure once the hashing power goes down. And then miners only mine if the price goes up. So the price sort of have to double every four years to sort of make sense, otherwise there's so much economics and numbers behind it. Yeah. I see where you're coming from. 

Sam Williams [00:31:19]:
Yeah. It's very interesting component. I don't think Satoshi did it deliberately, is my guess. I think it's just like a, well, the math is easier. No, which is no offense. Like you don't want to screw something like this up. And I know my impression from the Bitcoin white paper which was an amazing piece of work, technically. Not diminishing that. It spawned an entire industry that we now will work in, which is worth like $2 trillion right now. Unbelievably cool. But my impression with the paper was, it's almost unfinished. Like if you really get a feeling when you're reading through it and he gets to the last page or two, it tapers off and like, I dunno, like "I've got something else to do. I'm going to write this later". And then he just stops. It's really quite something.And you get that, what it's worth with the Bitcoin code base is also pretty similarly, like, with lots of sensible things, but they didn't have a block size limit at all, for example, and no DPoS protection until like a week before Satoshi disappeared. It was an amazing system, but it's like sort of cobbled together in some sense. 

Bobby Ong [00:32:19]:
So I guess one of my last questions, I suppose. would be what's kind of the plan for Arweave this year, next year and next few years? Yeah. 

Sam Williams [00:32:26]:
Yeah. I mean, what's really exciting about the ecosystem at the moment is it's just reached this very clear inflection point where the founders of protocols, some of them, but startups in general, profit sharing communities on top of the network, who've taken the reins of like developing the things on top. And really our job is as the core Arweave team is just become like supporting these people as much as we can. So almost all of our effort now goes into founders support, for people building on top of the platform. And it's working really, really well. Like amazingly well, it's so cool to see people, they quit their jobs and they raise capital and they go and build the opponent decentralized web together. It's fantastic to see. Yeah, so we're just focused on supporting that. Everything the ecosystem needs to make that happen more effectively. Even more if I had to be, frankly. Yeah. That's the focus in terms of the core protocol itself, it's pretty much complete. I don't want to change too much about it at this point.

There was some things we would like to iron out. We'll keep developing little bits, but we're very much of the mind that good protocols mature and then stop changing. They get to the point where they will be scalable. They effectively solve a problem. And then they mature and really like freeze. You'll note that all of the important protocols in our daily life basically don't change that often at all. Like power sockets, for example, same format, dozens of years. Yeah. And there's actually an interesting reason. This is very important, which is basically people need to trust that the protocol isn't going to change In order to embed it in things and the more trusted it is, the more I can embed it in stuff. You imagine, it's an interesting code experiment, if the USB standard hadn't changed like five times in the last, you know, 20 years, I guess, how much sooner would we have seen USB ports embedded into a power sockets that are now found. I think almost instantly, actually It's an obviously useful thing to have, like you've been carrying around these tiny adapters for many, many years. And that I think is because people couldn't trust the protocol not to change under their feet. And so there's a very important thing that needs to happen. You see it with Bitcoin now. Basically it's impossible to change Bitcoin in any serious way. Because we've all bought into, this is the archetype of what Bitcoin is until always be. Yeah, I think that process needs to happen and it is happening. That means that our attention now is on the layers on top of the base protocol. Things like SmartWeave, the smart contracting platform allows you to run arbitrary amounts of compute without gas fees with actually just storage fees. Yeah, all sorts of stuff like this and what not.

Bobby Ong [00:34:52]:
Interesting perspective there you brought up about how protocols shouldn't change. Seems to be, I mean, Bitcoin is, you know, it's impossible to change, but it seems to be quite a different perspective with Ethereum, where there seems to be constant changing of the protocols with the recent EIP and the miners are getting angry right now. And there's ETH 2 massive, massive changes coming along as well. So quite different philosophy of how things are going about. 

Sam Williams [00:35:15]:
Well, and I think part of that is because everybody in the heads knows that a huge change has to happen to Ethereum for Ethereum to survive. That is ETH 2, basically. And so we've got in a heads, yeah, you can change whatever you want about the protocol. We're not going to buy into it that deeply because we know it's going to shift. And we've been thinking it will shift in a year, for the last three years.  

Bobby Ong [00:35:35]:
And the thing about it is, ETH 2, I mean, you brought up a point, like, my mindset is set that I mean, Bitcoin's never going to change. If it change, someone's gonna fork it and go back to its origin idea. Whereas ETH is just going to be constantly changing. There's ETH 2, but there'll potentially be ETH 3, ETH 4, ETH 5 and they will constantly evolve, where it's kind of like the USB port analogy, like, I won't even be surprised if ETH keeps evolving over the years.

Sam Williams [00:35:58]:
Yeah, but I think if you do that it's not going to be able to really decentralize properly and people won't trust it in the same way that they do Bitcoin. Bitcoin has a shot at being digital gold, if not just digital currency, full stop, because people can trust it. We think that archive are essentially the same thing. This idea of like, "Hey, we give you the opportunity to store your blogposts in perpetuity, literally forever, replicated all over the world, no one can alter it, no one can take it away. No one censor it at any point in the future". We offer that for like a fraction of a cent. And I think that we're going to need to come to like a conclusion on what the protocol is in order for people to be able to trust that. In the same way, to loop it back to the beginning, we were talking about, you know, this 0.5% or 200 years thing, you need to give it some kind of really robust conservativeness in order for people to adopt it and use it for its intended purpose. Yeah. It's interesting discussion for sure. 

Bobby Ong [00:36:52]:
Yeah.  Thanks a lot for taking the time to explain Arweave to us. Definitely very insightful. I learned a few things over here and I'm sure anyone listening to this will also learn a few things about Arweave. 

Sam Williams [00:37:03]:
Yeah. Thanks so much for the conversation. It's great. 

Bobby Ong [00:37:06]:
Thank you. 

All right. That wraps up the show. Thank you for listening to the CoinGecko podcast with Bobby. If you like our show and want to know more, check out Or please leave us a review on iTunes. If you have any feedback, do drop us an email at Join us for more next week. See ya! 

Disclaimer [00:37:26]:
This podcast is provided as part of the overall information on cryptocurrency contained on our website, is for your general information only and does not howsoever constitute any endorsement, financial or investment advice nor any solicitation or offer of securities or other financial instruments. CoinGecko and the podcast presenter makes no warranties, implied or express of any kind in relation to this podcast, including without limitation the accuracy and updatedness of its content. All opinions and recommendations there in the podcast are based on the personal opinion of the presenter. Please conduct your own research and procure professional advice should you at your own risk, decide to howsoever invest or trade in relation to the content contained in the podcast.