Adam Stacoviak: That’s how we power the house, you know? So back to this unique thing - basic attention token… It’s so foreign to me. Ethereum-based, we’ve done shows on this - I get it; I understand blockchain, I understand cryptocurrencies, I keep hearing more and more that they’re coming of age, they’re about to be mainstream, or they are mainstream and there’s a small mainstream, who knows…? What’s the state? Where are coin in general being used? What is the state of this cryptocurrency system?

Adam Stacoviak: Using the new Brave it’s pretty interesting, where you can in the right-hand corner see the known BAT triangle icon, and you click that and it pops down and shows you your wallet, and it actually shows you that you’re on a Brave-verified publisher. Of course I’m on Changelog.com talking about this, so if you’re listening somewhere and you wanna open up a tab - sure, why not? Go ahead and go to Changelog.com and follow my direction here… But it’s pretty interesting, you have a button here that says “Send a tip”, and I find this kind of interesting, because we’ve always kind of been pulled into this “Pay with your attention” model, and it’s never been attached to something so foundational to a user’s day-to-day usage and/or software habits like a browser. I don’t know – Jerod, am I crazy? This is the first time, right? …a cryptocurrency available to give it out; it’s based on attention. Right now my attention is slightly higher, towards Changelog.com over Brave.com. I’ve been to two sites in this new Brave browser app installed, so 61% of my attention is going to our site, Changelog.com, and the other 39% is going to Brave.com… So does that mean I’m automatically giving those two sites just my attention and I can divvy it later, or is there some sort of auto – like you said, it’s auto-contribute, right?

Adam Stacoviak: Nice… Okay. We have talked to Kevin on the Changelog as well; when we first heard of Gitcoin, we thought it was actually a coin, and it took us a little while too to unravel the onion and say “Oh, okay, so this is about funding issues through cryptocurrencies, because ConsenSys is essentially very bullish on the cryptocurrency markets, specifically Ethereum.”

Kris Brandow: …but I feel like AI is exactly like cryptocurrency. I feel that the underlying technology –

Kent Quirk: This won’t surprise any of you here, but my unpopular opinion is that cryptocurrency is worthless. Having spent two years doing it, I came away convinced – and trying to create what I was thinking of was going to be the ethical cryptocurrency, I came away with the conclusion that pretty much everyone in cryptocurrency is either a crook, or an opportunist who wants to take advantage of the crooks, or hopelessly naive and not paying attention to what’s actually going on. There is no useful use case for cryptocurrency still, all these years later. Web 3 is a disaster, and blockchains are not useful tech. And so yeah.

Adam Stacoviak: [01:01:54.18] You were explaining – I think where we were trying to go is like real-world examples… Like, who’s using it? Why are they using it? What are they building with it? We kind of know some examples now, like we’ve mentioned ICO’s and fundraising and different things like that… Actual value, which is a cryptocurrency, or in this case maybe the FedEx example, or eBay for example, but other examples might be pretty interesting to share… Can you share more about what people are building with it and why they’re building with it? Maybe what the value is happening in a certain community, around what they’re building?

Liz Rice: [58:13] Yeah. And then on the security side - so Falco was probably the first security project, certainly in the CNCF, that was using eBPF. This project from my former colleagues at Aqua called Tracy… And then in Cilium we have a sort of Cilian family, we have a sub project called Tetragon, which is allowing you to create low-level security primitives almost in Yaml form, and apply them to your Kubernetes cluster. And you can do really cool things with Tetragon. I get a bit overexcited about this, because if your kernel is modern enough, you can not just detect that something is potentially malicious behavior, a process is opening the wrong file, or connecting to a cryptocurrency miner, or whatever malicious thing that you’ve detected - you can kill that process synchronously from within the kernel. And what that means is the process gets killed - to me it’s not like you have to go and tell somebody and then eventually your process gets killed… It’s happening right there and then, and it stops the attack before it happens… Which is super-fun to demo. I love it.

Andrew Stanton: I do. I don’t think we’re necessarily all the way there yet, but I think there’s a bunch of indicators that are really positive for it. First, big companies are starting to depend on it. Dropbox, for example, uses Rust in their storage layer, because they need reliability. Facebook came out earlier this year with their cryptocurrency which is written in Rust, because they need the security. Microsoft a few days ago published the result of a project where they were trying to replace core pieces of their Windows 10 code with Rust, because they got tired of security flaws. So we’re kind of seeing a lot of big companies starting to adopt it, at least in prototyping cases, and more seriously in the case of companies like Facebook.

I think that it is going to be really valuable, because the one thing that I’ve noticed for us is that there’s really kind of three payments that you have to make when it comes to building machine learning. There’s the cost of development, there’s the cost of running, and then there’s the cost of maintenance. And Python does a great job at the cost of development. We have this rich ecosystem, filled with libraries out there, you can just PIP install them, they just kind of work… You can prototype an idea very quickly. But it’s not an efficient language, so that cost to run starts adding up, especially when you have to clone it over hundreds of machines to train these very large models.

[40:08] The final piece - and this is the more insidious one - is the cost of maintenance. What is the cost of failure in your systems. And when we think about things like production, whenever our learning to rank models go down, we start losing money really quickly. We have 18 million results for jewelry, right? It turns out that there are better answers in there than others. So we need to be very cognizant of what that long-term effect actually is when it comes to building out “production”.

Brian Behlendorf: No, no. But then – so learning more about it, and spending time with Jim Zemlin, who’s the head of the Linux Foundation, spending time with folks who are part of pooling that together, from IBM, from a startup company called Digital Asset, from a bunch of others, learn more about the rationale and said “Okay, this is important to work on.” It’s not against the cryptocurrency side of the world, but in fact there’s ways that we need them and ways to work together, but maybe there’s some real opportunity here to build something that has some lasting impact.

Steven Pyle: I mean, part of it, I guess, is - yeah, the models have never been perfect; we know that. That’s part of it. The other part is the environmental aspect of it. You’ve got all these servers… It’s a reminder of cryptocurrency, where it’s using all…

Kris Brandow: Yeah, I’m curious as to like – because you know, it’s easy to be hyper-fixated on what the hype cycle is currently occupied by… I’m interested of “Well, what happens when the hype cycle inevitably moves on?” Which I think everybody right now - because we’re in the hype cycle - is just like “The hype cycle is not going to move on.” But I don’t know, I’ve said this before; it’s been an unpopular opinion before. It’s just like, Bitcoin and cryptocurrency was also the huge hype cycle that everybody was talking about, and it’s now been replaced with AI. And before that – there’s always a hype cycle, and there’s always something.

[01:17:56.10] I think there’s a lot of interesting things that come out of the – like, once the hype cycle has moved on, you’re left with these technologies, of like “Well, what can you actually do with these technologies? What are interesting applications of them?” I think that’s happening now with blockchain and how people are thinking about blockchain in kind of new and novel ways. Or even the components or the concepts of a blockchain, irrespective of the implementation of it. And I think that AI as a space has the same sort of potential, where it’s like once we can actually really kind of get into the nuance of it… So I feel like that’s kind of what follows those big hype cycles, is the hype cycle is the lack of nuance, and that lack of nuance allows us to imagine literally anything. And people doomerize it, people are super-excited about it, but everybody is just like – there’s a whole lack of nuance around most of it. And then you start getting a lot of that nuance, and the hype comes down, and you start being like “Oh, these are the interesting places where this technology can actually go.”

And I think that Go as a language, it’s positioned in a special place, and I think it is – I mean, as the language of the cloud, I feel like it’s gonna play a role in all of this, but a different role, a complementary role. So I don’t think it’s gonna be the language you use to necessarily write large language models, or do the things that other languages are currently good for. But I think it’s going to be somewhere in there, probably playing a very foundational part. Kind of like with the cloud. Go is certainly not – you don’t have to write all of your services in Go. In fact, you might not even write Go in general, but most definitely, it’s all over the place in the cloud world, and in that whole thing. So I think there’s space.

But yeah, it’d be super-interesting to explore that on the podcast, and have people in… I think it was – I don’t remember who brought it up, but someone brought up the idea of not necessarily debate, but different viewpoint episodes, where you bring in people that have not necessarily opposing views, but differing views, and you have a conversation. I think that would also be interesting in this space, but also just in general for us as a Go podcast, because there’s a lot of things in Go where it’s like “I don’t know, the answer is it depends.” There’s two episodes basically of this on Changelog & Friends. There’s the “It dependencies”, and then there’s an “It depends” with me. So listeners, if you like that kind of “I don’t know, the answer is It Depends”, go over to Changelog & Friends and listen to that.

Matthew Boyle: [06:04] Yeah, I think just to go back to a comment Mat made, firstly it’s about - like, obviously, you’re going to build software to model your problem space. What else are you going to do? And I think that’s such a fair comment, because it seems obvious, but I think one thing that domain-driven design gives you is it gives you sort of principles, ideas, and a framework to work through to help you figure that out.

So I think there’s so many key language and sort of ideas that domain-driven design sort of gives us, that helps us to think about what is the domain and what is the problem space. A couple of examples that I use in the book is - I think I build like a coffee shop system, right? And you may have a domain around ordering; like, ordering could be an entire domain in itself, because it’s like a vertical slice of the business, that has all its own rules, it has its own language, it has its own sort of ideas and thoughts… And so to slice that off into a domain makes a bunch of sense.

And then you may have subdomains off of ordering, like maybe order fulfillment is a subdomain of ordering, and that’s something that you want to kind of model separately, but use some of the same concepts you have in ordering. So that’d be one example.

I think another example I use in the book is like a wallet. You have the sort of domain of a wallet, and within that wallet you might have various different debit, credit, and - less relevant these days, but maybe cryptocurrency would be in your wallet, too… And modeling all those things separately makes sense, but they all kind of belong to this higher order idea of a wallet.

Nader Dabit: Yeah, those are actually fascinating points to me, because it is fundamental to how this space actually works. It also applies to a lot of different things in the Web3 space; the idea of ownership. Being an owner of Ethereum means you’re part of – you could almost think of it as buying stock in something like Tesla, or something. You want that company to win. I think one of the main differences here is that you’re part-owner of some token, and you might want that token to succeed, and you might be out there hyping it up… But you can also build software and applications with it. You can transact in the way that you might have in the bank with the U.S. dollar, but you probably don’t go out there and try to have the U.S. dollar succeed or fail. It’s not even probably an after-thought. Of course, you don’t want it to fail, but yeah, with the cryptocurrency you can kind of do all of these different things with it…

And it does provide incentives, in good and bad ways… So if the Graph token, for example, which is one of the tokens – I would say it’s the token that the Graph Network uses to make it function… People that have bought it in a speculative way might come into one of my tweets and say something stupid, because the Graph went up or down too much that week or that month, and they’re upset. So you end up having people that are interested either in the technology, or they’re not and they’re just speculating… You have all these people in the same discussions, in the same communities, and it’s definitely different.

Feross Aboukhadijeh: Well, I was really sad when Mozilla shut down Firefox Send, because it was such a nice, simple product… And I was like “There’s no reason that this thing couldn’t be rebuilt and done in a way that is sustainable.” So that’s kind of what we wanted to try, was to make a version of it that wouldn’t get shut down. And it was also a really nice application of all the tech that I’ve been working on throughout most of my career. The peer-to-peer stuff from WebTorrent really does improve it, even beyond what Mozilla did with Firefox Send, because we can do that streaming that I mentioned, so the downloader can start downloading right away… The user experience stuff that we did too is actually another thing that Firefox Send didn’t do.

Here’s the thing - when you’re building a security product, you can either sell it to people as going into all the security details about why it’s great, like I just kind of did with you guys, and told you all about end-to-end encryption… And then you’re gonna get some people who are like “Oh, I am worried about people getting access to my files, so I’m gonna use this thing that is more secure.” But I think most people just care about the experience of using it… So I see this mistake made a lot when people are building, especially in the decentralized space, or even like in the cryptocurrency world you see this a lot, where people are making stuff that’s so hard to use that they only get the true believers, the really ideological people, who would use it no matter how terrible and hard it was to use… And everyone else who’s not really that bought into the ideology of “Oh, everything needs to be decentralized” or “Everything needs to be - whatever the buzzword is.” Those people are not gonna be sold by it, you know what I mean? If you’re making a Twitter clone, for example - a Twitter clone, but it’s decentralized. But it’s like ten times harder to use than real Twitter. You’re not gonna get any normal people to use it, “normal people”.

[16:12] So that was the thing I wanted to be really careful when we were building Wormhole that we didn’t do. I didn’t wanna be like “This thing is worse in a lot of ways, but it’s most secure”, and try to get people excited about that. It was like “It has to be better, even if you don’t care about security.” That was part of the challenge, like “Can you even do this?” Because usually, there’s some kind of compromise when you’re trying to increase the security or whatever… You’re making it harder in some ways; it’s more challenging from a technical perspective. And sometimes that tech leaks out in a way that the user sees it. Like “Oh, you’ve gotta save this key”, or whatever nerdy thing you have to do in order to use it. We’re trying to do it without any of that. So just be better, faster, simpler, quicker, get you the link right away, not at the end of the upload, all that kind of stuff. And then hopefully, they perceive that as being just way better. And then also, icing on the cake, it’s end-to-end encrypted. We can’t see your data. In that order.

Break: [17:04]

Raul Jordan: Definitely. I think Preston and I consider ourselves pretty pragmatic engineers, so when we first started on this journey, we were just excited by this concept of having this global computation platform that anyone can use. It was like “Wow, it’s so cool, the things that this can enable for the world.” And of course, blockchain cryptocurrency has had quite a negative reputation on a lot of fronts, especially because there have been a lot of nefarious use cases for the technology. And you can say that about any sort of new technology.

Why we got excited about Ethereum is because Bitcoin was, of course, the first blockchain, the original blockchain, and it serves its use case as basically a store of value; basically, the predominant narrative is that Bitcoin is like this digital gold. But on Bitcoin, interestingly enough, you can’t really run code, or run these decentralized applications that are Turing-complete. You can’t write a custom piece of software that basically can do whatever you want. Ethereum is the first blockchain actually that allows for these decentralized applications, for basically decentralized computation of software; and to us, that was really cool.

Ethereum has the most network effect, has the most developers, the most people building on it; it’s not controlled by a single company or a single person. People say it has a soul… It’s basically community-owned, and that to us is really awesome. That’s kind of like why we got into it.

Lea Anthony: [laughs] I guess one of the key targets for this application would be somebody who wants the power of Go, somebody who wants to perhaps do low-level stuff, USB stuff… I’ve got a couple of projects that I noticed that are working in the cryptocurrency space, so there’s a lot of [unintelligible 00:09:36.27] that happens… And it’s for people who want to just present that sort of stuff visually, where you perhaps otherwise maybe use a web browser to connect to, but that’s doesn’t feel quite right… And it’s something that you wanna interact with in real time, but still have that power of go. That’s really your target.

There’s a keyboard company in Japan called ergodox-ez. They create these amazing keyboards, the ones that have like a million different parts and they’re super-ergonomic… One of their main developers contacted me a while back, and he was using WebView to do that firmware flasher. He’s recently ported their application called Wally, which is their firmware flasher - he’s ported it over to Wails. I’ve worked pretty closely with him; we didn’t have to do much. He gave us some great feedback, and… Yeah, I was really happy about that, because it really shows the depth of application you could do.

Matthew Hodgson: Some of the Rust community went to Matrix, rather than Discord… I’m hoping that the rest of the Rust guys will see the light and come back, eventually. So it’s honestly a weird mix of open source – lots of cryptocurrency projects, like Tezos, and Status, Ethereum, obviously Parity (that I already mentioned), open source cryptocurrency types… Also activists and people who really care about privacy and encryption, and then on the flipside governments… And that’s kind of the main uptake right now. And we haven’t yet shifted into the more mainstream people who are using Slack today, but that is the big challenge that we’re tackling really over the course of this year… Trying to get the UX glossy enough in Riot, trying to make sure that features like Communities are there.

In terms of first-time user experience right now there’s quite a lot of wtf moments, so to speak, for new users… We just haven’t quite polished it as much as we should, or we’ve let a bit of the protocol shine through into user interface. For instance, right now you can invite users in DMs. If you’re used to it, it’s really useful. Somebody DMs you, and you think “Aah, you’re not talking to the right person! Let me invite in the right person to direct you to!” Or if you want to invite a bot into your DM, so that it can translate it for you, or something like that. It’s actually kind of useful.

But the level of panic that some people have when they’re coming over from Slack and discover that the one-to-one that they were having with somebody can actually be magically upgraded, not even into a DM, but into a private chat, and it can even be made a public chat if you wanted to, just because you can, is a classic example of “Just because you can doesn’t mean you should.”

Paul Frazee: But I think they had a different concept of it. I tried it out, and whenever you went to a torrent, it actually sort of acted like a typical torrent client, as opposed to behaving like a website. I don’t know if they were quite tried to use it as a drop-in replacement to HTTP, which is what we started doing with Dat and that was one of the differences…

But back to Feross’ original question, the good news is that there’s a lot of projects out there right now that are attacking this problem. So for anybody that’s really into decentralizing the web, you should be happy to hear that. A lot of it has to do with blockchains. Beaker and Dat - we don’t use any blockchains; we’re really actually just seeing how far we can get with Dat… And there’s some technical similarities, because Dat has this append-only log behind every Dat archive, which retains all the history and it uses signatures on that log… But unlike almost every other blockchain, there’s no decentralized consensus, which is the major feature of a lot of the cryptocurrencies.

[16:23] But the Ethereum project has been really explicit about how they wanted to build a next iteration of the web with their tech; they even started, I believe, the Web 3 Foundation.

You know, I guess almost all of the blockchain-based solutions have payments at the core of what they’re trying to add to the web, which is cool… Payments are important; it’s important that people can make money off of their work, and it’s all sustainable… But the main difference, I feel like, between what we’re doing and what the cryptocurrency-based projects are doing is that by not using the blockchain, by not having decentralized consensus, we actually reduced costs even more. You can do this all with hardware that already exists, you don’t have to pay any money to transact on this network, and you end up getting a lot of the same kind of benefits, where you’re detached from any proprietary cloud, which is really what’s at the heart of all this.

So that would be probably the most significant difference between a lot of these projects - is there a cryptocurrency involved or not?

Chris Beams: That’s right. And for completeness, I should actually say, it tends to be in practice that one side of every trade in Bisq is Bitcoin. We’ve actually recently added support for additional kinds of what we call “base pairs”. In the situation that I’ve just described, Bitcoin is the base of the trading pair, but we also support Dash, which is another popular cryptocurrency we support, Litecoin, we even have support for Dogecoin, which you guys may know and other people may not know… It’s kind of a meme, kind of a joke on cryptocurrency, but people use it and it’s even possible to trade against that as a base pair.

Anna Derbakova: Yeah, so maybe you can announce that. But you know, just to explain it, Blockchain is the system that powers Bitcoin. A lot of people have probably heard of Bitcoin, and that’s the one assumption I made in my talk, that that’s something people know about… It’s a cryptocurrency that’s really widespread.

Cory Doctorow: [39:55] There’s two parts to that; the first one is a very practical thing that’s going on right now. As apps gained ascendancy, browsers lost some of their power; they became less significant to technology ecosystems, and this made browser vendors and the World Wide Web Consortium (which standardizes browsers) pretty desperate. In 2013, the W3C decided to add DRM to the core set of HTML standards for HTML 5, and something called encrypted media extensions. What these mean is that for the first time ever, the person who runs a website will be allowed to tell your user agent, your browser, how it must perform, how it must render content and whether it can render content. That’s never been the case; this is why ad blockers and pop-up blockers work, but also why I have bad low contrast vision, and when I get to gray on white type, I turn on a thing that turns the gray type black. All of that stuff only exists because users are able to configure their user agents to display the web in the way that’s convenient to them, and so what this is doing is it’s setting up this regime where it’s a felony to change the way your browser is configured, if that conflicts with the interest of the people who serve the content to your browser.

We have proposed to the W3C that it should take its existing policies and extend them to cover DRM. Right now the W3C, if you join, you have to promise not to use your software patents to attack people who want to employ W3C standards. The W3C’s position on patents is that standards are more open if you don’t need to license a patent to implement them, and so they have this policy. We’re saying, “Okay, well there’s this new right, the right to tell people how their browsers must work, that you’re creating by doing this DRM standardization. Surely, a standard is more open if you don’t need someone’s permission to implement it, because of DRM, just as surely as it is with patents. You should have the same policy for patents”. So that’s coming up for a vote very shortly.

We’ve been joined by some pretty significant parties there, the browser vendor Brave is in with us, so is the Royal National Institute for the Blind and Oxford University. They’ve all signed on, along with most of the crypto-currency and Blockchain companies that are W3C members, because as cryptographers, they are like “Yeah, of course people should be allowed to report vulnerabilities in browsers.” It’s a terrible policy to say that companies get to decide who can report vulnerabilities. That’s a thing that’s live and underway, and if you work for a W3C member, you should talk to your rep about supporting us there, because that is coming up any day now.

Max Howell: Well, the main purpose of T, at least what I wanted to accomplish when I came up with the idea, was to use cryptocurrency to fix what we call the Nebraska problem after that famous XKCD comic. You know, the [unintelligible 00:55:46.21] representing all of open source, as it’s stacked on top of each other… And those little projects near the bottom that are fragile, because the people who maintain them don’t have the time or the incentive to do so, and yet it’s holding up so much critical infrastructure.

[00:56:04.06] So yeah, it’s a cryptocurrency project that uses a unique tokenomics model in order to give open source developers token rewards on a 24-hour basis. And a lot of the other pieces of it are designed to attract the interest of typical crypto investors, or just like normal developers who want to show real support for their open source projects. A key differentiator between us and most ways of supporting open source is that there is no donations in our system. You can buy a token and then stake it against projects, so both you and the project is gaining from this. There’s no gift. It’s more like an investment.

Rich Burroughs: That would have still been when I was at the ISP, because even there we were dealing with – it’s kind of funny, when the whole cryptocurrency stuff happened… Because in like probably like ‘97 or something we were dealing with this thing called cyber cash, that was like one of the early encrypted currencies… And I was big on – there was this email list called The Cypherpunks email list, that was like the crypto nerds, and I was very big into that stuff. I would go to like PGP key signing parties, and stuff… And so it was funny when the cryptocurrencies stuff came up, and everybody was like “This is brand new.” And I was like, “No, it isn’t.”