Adam Stacoviak: That’s why the bigger the problem that you solve, the more money that can exchange the hands. Oor the value, however you want to term – currency, bartering is how we began, then it turned into things you can actually exchange… Those things became true currency backed by local governments, and nation states etc.

Preethi Kasireddy: Yeah, sure. That’s a great question. I’ll be honest, one of the things that I’m always excited by are problems that are not figured out, and blockchain is definitely one of them in the sense that there is this amazing technology that got created, but the actual applications and use cases have yet to be created, and there’s so much room for creativity and innovation. I’d say that’s one of the main reasons why I’m so attracted to this industry… Because I naturally have the entrepreneurial spirit, I wanna solve unsolved problems, and that’s one of the reasons why I’m gravitating to it. But at a more fundamental level, I think the technology itself is super interesting.

I think earlier you were saying how blockchain is interesting as a digital currency… I’d go a little bit further than that and say I actually don’t get excited by digital currency per se, because digital currency is just one application of the blockchain. There’s so many other applications that I think have potential that haven’t been created yet, and I’m more excited about those applications. In my opinion, it’s really hard for the current currency system in any nation – for example, take the U.S. For cryptocurrency or digital currency to replace fiat currency, that would take a lot. The government has a really tight control over the monetary policy and I just don’t see what benefits digital currencies provide today that would make people switch over, that would make the government friendly to this, and so forth… Because in the end, what digital currency is - it’s completely decentralized, completely open, completely censorship-resistant, which is not something any government really wants.

So for me, I’m more excited by what else blockchain enables. At a fundamental level, blockchain is a decentralized technology where it’s maintaining a ledger of transactions or basically state changes, and the state changes are cryptographically secure because they are maintained by something called the consensus process. This consensus process in Bitcoin or Ethereum is called proof of work, where a bunch of miners are there and they run a bunch of complex computations to validate these transactions, and they get paid for validating this blockchain. So the entire chain is decentralized, it’s trustless, because there’s no need to trust any central party to maintain this database or this ledger, and then it’s completely censorship-resistant because you can’t have really any central authority trying to censor it… And it’s completely secure, because it’s cryptographically secure. So with those primitives you can do just so much on top, and that’s kind of what I’m excited by.

Mikeal Rogers: This also happens to your currency in your pocket. There’s inflation, there’s currency exchanges, and how valuable your currency is when compared to other currencies… I’ve run into this a lot, because I have a stack of money from other countries, so that when I land in those countries, I can spend it… And this has not been a very good last five years to be holding currency not in dollars, so that kind of sucked.

Anyway, so for blockchains there’s this whole side of things that’s like currency speculation, and it’s a lot of ostensibly gambling, like fun gambling, a lot of money flowing into it, and gambling on these exponential increases in the value of these digital coins. But the underlying technology can solve a lot of problems outside of just currency exchange, and outside of just things that we need a currency to do.

There’s an element of transparency and provability without a centralized owner that is really important for a bunch of use cases. What Brave is going after with this basic attention token is essentially the ad market. If you’ve ever read a bunch about the advertising market, especially online, there’s a huge amount of fraud, a huge amount of fake clicks… There’s everything from click farms to people just generating crazy wild numbers for what has and has not happened on different services and what not.
If you wanna try to solve that, you need something that is provable and has a lot of these elements of transparency and provability baked in… So what they’re looking at is this basic attention token. This is a provable way to show that you spent some attention on something. That could be used to prove that you saw an advertisement or saw some content… Or what they’re probably betting on a little bit more than that is you can prove that people spend time on a site, and then you can inject capital and money into where you spend your attention, and that can be doled out as micropayments to those sites.

[12:19] If you’ve used the Brave browser, they already have this feature baked in where you can do these micropayments. You can say “Look, I don’t wanna deal with advertisements on website, I hate all that kind of stuff, but I do wanna support content creators, so I’m gonna pour $20/month into wherever I spend my time.” What happens is that Brave tracks that in a way that is anonymized and protects your privacy, but it also allows them to dole out some of that to all these different places where you’re spending your time.

Adam Stacoviak: Well, isn’t that what they do? They’re trying to hedge their bets against fiat currency, that’s controlled by government?

Chris Benson: I have a follow-up to that real quick, if you don’t mind. Would it be right to think – we think of stores of value in terms of money, and we’ve been talking about money in recent years, we’ve looked at cryptocurrencies and we’re starting to think of those as stores of value and forms of currency themselves… Should we be thinking of data in a direct way? Because we’ve kind of talked like one step removed so far, but is data money in the way that we should be thinking going forward?

Jack Dorsey: I don’t think there’ll ever be a replacement for any of these things, but I do think there will be one that’s more dominant, and I think there’s a very strong potential that the U.S. dollar loses its global singular reserve currency status. And there may be a second one in the Chinese yuan. And there might be a third one in Bitcoin. And I think that’s a net positive. I’m obviously rooting for Bitcoin because of the properties behind it, because it’s transparent, because no company or individual controls it, because no government controls it, because it’s resilient, it’s secure, it’s never been hacked, it’s never gone down… That’s what I want out of my money. But I want the transparency, I want it to be owned by the people as well. So having a world reserve currency that is owned by the people and developed in the way that Bitcoin is developed I think is very, very powerful, and I think it would be ideal for everyone on the planet not to be controlled by the dominance of any one government’s currency. But I do think that we’re moving away from a time when there’s one, and there will probably be many, and then maybe one will become dominant.

Jack Dorsey: I don't think there'll ever be a replacement for any of these things, but I do think there will be one that's more dominant, and I think there's a very strong potential that the U.S. dollar loses its global singular reserve currency status. And there may be a second one in the Chinese yuan. And there might be a third one in Bitcoin. And I think that's a net positive. I'm obviously rooting for Bitcoin because of the properties behind it, because it's transparent, because no company or individual controls it, because no government controls it, because it's resilient, it's secure, it's never been hacked, it's never gone down… That's what I want out of my money. But I want the transparency, I want it to be owned by the people as well. So having a world reserve currency that is owned by the people and developed in the way that Bitcoin is developed I think is very, very powerful, and I think it would be ideal for everyone on the planet not to be controlled by the dominance of any one government's currency.

But I do think that we're moving away from a time when there's one, and there will probably be many, and then maybe one will become dominant.

Chris Benson: Yeah. And that, once again, comes back to what is their intent. What are they trying to do with it across their use case spectrum. And probably many are very reasonable use cases, but there are some – this article certainly leaves that ominous overtone in terms of how it might be supporting the social currency system in China… Which obviously – there’s a value difference right there between Western values and Eastern values, at least there in China. That’s the kind of thing that would scare more Westerners in a variety of Western countries… I think even in places like in the U.K, in London, where people expect to be surveilled, I think they would expect that the way the information is being used is not so nefarious.

So it’s interesting to see as we look at global trade how we’re going to reconcile some of these differences in terms of how we approach society, how we approach business and trade, and I think some of these companies – they may be acting completely appropriately based on an objective standard. I think one of the scary things, certainly for the American perspective, is just the ability to verify that, the ability to understand that they’re dealing with a vendor that does not have an ulterior motive, and stuff. Whether or not that would deserve to put them on the list here or not - I have no insight into that. It’s a tough thing, and I don’t think we’re anywhere near solving that.

Adam Stacoviak: …to use it in the book, and Tara Hunt to write a book on it, basically, but it’s – you know, you do good things out there, you get reputation, and that’s a form currency. That’s what I mean by that.

Adam Stacoviak: Welcome back everyone, this is the Changelog and I am your host, Adam Stacoviak. This is episode #222, and today Jerod and I are talking to Gavin Wood about Cryptocurrency, Ethereum, and this big world of unregulated digital currency.

Gavin is the founder of Ethereum, the creator of the Solidity contract language and the founder of Ethcore, the company that created Parity, an open source Ethereum client.

We’ve got two sponsors today - Code School and Rollbar.

Jerod Santo: So how should you store currency inside of Postgres? What’s the best practice?

Thomas Eckert: Taylor, have you ever tried to use the plastic babies as currency, like get a cup of coffee? Be like “How many babies do I owe you?”

Lukas Egger: That’s a really tough question, and I will try to answer it as best as I can, but I will be a bit – a bit of heresy and blasphemy coming in, because my pedigree is machine learning as well, but I think especially when it comes to machine learning, or we call it AI these days… I mean, when I started, we worked on conditional random fields. That’s how old I am, right? So I think there’s some sort of – I don’t want to say fetish. But we think that as soon as AI gets applied, there’s an intrinsic value that makes it more worthwhile. And to me, the problem behind it is focusing on output versus outcome. So you want an outcome, and the outcome, loosely speaking, is you want well informed, data-driven decisions, fast, in a way where you can adapt and quickly iterate or learn of past behavior. That’s what you are striving for, right? And the output might be model, but it could be something else as well, right?

There’s a currency in every organization - how we make decisions, who gets involved, and so forth. I think here there’s culture at play, and if you establish a culture of where the outcome is, we want to make the right data-driven decisions and quickly learn, based on previous behavior data, and instill that into the organization. It’s more likely that people will gravitate towards, let’s say, ML solutions. But at the same time, you’re not creating weird incentives where you’re saying, “I only think of your outcome in a positive way if there is some machine learning thing deployed”, right? And that’s what I would like to avoid.

In terms of then how to get the right people involved - once again, make it easy for people to talk with each other and set the right incentives. But for me, at the core of this, or where I focus on my attention most, is to make sure that we talk about the right outcomes, and not focus on the output so much, if that makes sense.

Nader Dabit: …they’ve now enabled payments within the game. [laughter] But what is the problem with that? Well, I wouldn’t say that this is a problem, it’s more of like a limitation. The limitation there is like when the kid outgrows the game, then you’ve now dumped a thousand dollars into this thing and nothing has come of it. Maybe they’ve had a great time, and maybe that’s enough; that is a good thing, right? Maybe having fun is good enough. But what if instead, people can actually take the value that they’ve put into the platform, and maybe get some of the return of their investment back? Or maybe they can give it away to someone else and have them use it, and have a really, really easy way to do that.

So Axie Infinity kind of took those ideas of what Fortnite has, along with some of the Web3 ideas, and they’ve built this game that allows players to earn in-game currency; and they can also buy it. But they can also resell that on a liquid market. So as you play the game, you’re getting paid, essentially, to play… And it’s not a ton of money for maybe someone like a software engineer, but for people in other parts of the world, they’re actually able to quit their jobs, and they’re making more money, sometimes orders of magnitude more money just playing this game, than they would be able to working in like a local area, using their skillset.

So when I started learning about stuff like this - this is an example that kind of excites me, just because it’s so revolutionary… And applying these same ideas in a bunch of different areas maybe in the future is also really exciting. And I think the idea is – like, having the combination of the technology to make it happen, the actual implementation that’s successful, and then all of the different areas that have yet to be applied is very exciting, because that means we’re kind of still early in on making these changes, but we can see that they are actually things that can be applied, and they can be successful.

So yeah, that’s kind of like – I don’t know if I went too far into a tangent there…

Quincy Larson: Yeah, well YouTube does that automatically. For becoming a member on the channel, which was what you had to do to be able to chat – we just set that up because, again, we wanted to see if we could get some donations… Essentially, the way YouTube works is they keep 30% and you keep 70%. I think with Twitch it’s a 50/50 split, if people become subscribers on Twitch…

So I think some of that may be scaled based on the country. I’m not even sure how popular Twitch is outside of the U.S, but YouTube is definitely like a global phenomenon. By usage, I think it’s the number one site in the world right now. I think it’s passed Facebook. So if you wanted to, there’s something called The Big Mac Index, which is – The Economist puts it out every year; they take the cost of a Big Mac, which is available in pretty much every country, and they scale it… And they use that to determine the multiplier that you should put on a price of like an online good if you want people to all be paying about the same thing in their local currency, based on their earning power.

Mikael Cho: Yeah, I think we’re still figuring out the use cases where blockchain would be so obviously better than anything that you could do with fiat currency. We’ve seen some of these consumer platforms where people get paid to contribute, or these sorts of things, and there’s tokens that are created in exchange… But you can also ask the question of how much better is that – if Reddit Gold, the token system that is inside of Reddit, what if that was run on blockchain? Would it be significantly better than the system that there is now? I don’t necessarily see that as being significantly different or better, but I think there could be a combination of things that happen, and we’re looking into a bunch of little interactions that happen on Unsplash.

We look at the download, and the interesting thing that we have with our download is, because people don’t have to credit, the photos can end up anywhere all over the internet. So is there a way that we could potentially figure out the line of creation? There’s this original photo that gets posted and then all of these things get made with it after. Is there a way that we can connect all of those people somehow? Right now, technically, it’s uncertain, but could there be something within blockchain technology that could allow us to do that better? Maybe.

There’s also different things that help the Unsplash ecosystem function really well, for example photo tagging. As you were mentioning, when you search for a photo, the library is getting better and better. We actually have a lot of people who will tag the photos just as their contribution to Unsplash. So is there a way that we could help those people be recognized? Maybe there is a token ecosystem where we can break that down into micro transactions that would be significantly better than doing it with cash or some sort of fiat system.

Brendan Eich: The story of my life. I was a Unix kernel hacker at Silicon Graphics before I ended up at Netscape, and I always worked on platform code… I think you see - it’s pretty explicit now - open fintech through the Symphony Foundation and other things… You see a lot of companies realize that open source is better for quality assurance, recruiting, lots of things that traditionally they would have to pay for themselves, so they can share the cost of platform code, or what Georgios Kontaxis calls evolutionary kernel code. This is the sort of stable code that’s conserved like the best DNA in a population. It’s like the TCP/IP or JavaScript - once you stabilize it, everybody can build huge systems above, and sometimes even below it. You can have multiple link layers and go from Ethernet being 10 Mbps on copper, all the way up to fiber (metropolitan Ethernet or whatever ATM cells) and still have this TCP/IP in the middle, and sure, IPv6, but it’s not really taking over, and it’s all evolutionary. JavaScript ES6, here we go again!

The platform code, the evolutionary kernel code that’s sort of “the commons” in the best sense of the word, is a cost center. When I was at Silicon Graphics, as I developed hot, killer graphics, workstations and then high-end multi-processors and low-end desktop graphics workstation machines, eventually to be killed by the PC and the GPU in the ’90s, the kernel group that I worked in and the network software group got kicked around. It was a cost center, it was an albatross, or else it was a source of talent for building out something important for the multiprocessor business. So they got kicked from the “hot product” group to the “not hot product” group and back. I think it even got divisionalized a little bit, not fully forked. HP did the same thing.

I see a pattern here, where open source is serving the commons, it’s not serving the differentiated, risky or for-profit innovation, that for better or worse some of that stuff stays proprietary. But anything that starts to become a platform, starts to become a cost center and needs to have its costs shared if it’s of interest to many other players. And how do you fund that? I wouldn’t say it’s exactly like publishers, because publishers often are for-profit. But not always - Dow Jones was a long-time family-owned and subsidized… In some ways we need Carlos Slim and Jeff Bezos to prop up the nation’s number one and two papers of record, right?

[01:00:22.27] Newspapers have been in a decades-long decline, and they always relied on advertising and subscriptions, and subscriptions never paid for the whole thing; they were always a minority of the revenue needed to run a newspaper business, even back in the heyday, the golden age of newspapers. Because people would subscribe, and there was some revenue you made there, but advertising paid the bulk of it, and that’s still true.

The way I look at this is not to say “We must have advertising. Advertising is always good.” There was a TV executive I heard about in the ‘50s who said, “It’s inconceivable that television will ever be other than free and advertising-supported”, and of course we have Netflix now, so never say never. Maybe the Brave donation model, in some future frictionless micro-donation, micro-payment, micro-royalty world will suffice. We still have free television with ads for sure, even with Netflix. But if you look at how costs are covered, you have to look at what’s happening today. If 70 billion is spent on ads in the U.S. (I think this year, or maybe it was last year), and Facebook and Google are taking a lot of it (they’re taking 80%) and the increment in spending from last year - maybe it went from 60 to 70 - of that 10 billion increment they’re 90%, that’s not leaving a lot for the publishers. And if you look at how the publishers do their ad businesses, they have to pay if they sell direct ad space. If they do indirect, they’re at the mercy of malware, like I said, but they’re also getting far less, because there’s so many people in the middle, cutting out from the pie; by the time the pie gets from the marketing side to the publisher, there’s very little left - 35-40% or less.

Still, that’s a lot of money. That’s billions of dollars a year, and these companies need to get it, so how would you go about replacing that? Assume for a moment things need to be replaced as is, that we won’t get a better model, we won’t find fusion energy, like Sam Altman thinks would make electricity free - I kind of doubt that - but ceteris paribus (all else equal), how would you replace that 70 billion? I think about that a lot. First of all, I think a lot of it is wasted on ads that never are viewed. This is the big scandal that’s been breaking for the last year or so, thanks to my friends at White Ops Security; there was another group whose acronym I’m forgetting - ANA I think, that did a study that showed there was a lot of fraud and kickback nonsense going on…

[01:02:56.04] A lot of ads aren’t being viewed. Facebook recently announced that its video ad metrics were off, way high from what they actually were. And they were charging accordingly, so people are kind of mad about this. But we have computers, we have smartphones; we could theoretically do a very private platform that measures what you’re interested in without giving away your data profile or your privacy, and matches valuable opportunities to you and give you a cut. That makes me think there’s a way to fund the web even if it’s not a commons. But certainly, for things like a publisher site that is more of a commons - obviously, Wikipedia is an example, but there are others - or all the open source software that everybody wants to share the cost of, because it is a cost centre, it is even an evolutionary kernel in some sense, and it has to be sustained by everybody who’s chipping in. I think there are ways to fund it, it’s just we haven’t found the ways to do it. That’s why Brave’s doing Bitcoin under the hood. It’s not because we love Bitcoin; we don’t want everyone to learn about Bitcoin. We do not intend people to have to become Bitcoin gurus. We haven’t announced yet, we’re doing a deal where you can easily just trust us with your credit card to do a recurring small charge to get Bitcoins; you don’t have to think about it at all. Currently, the way you fund your Brave wallet while we’re doing this Brave payments, beta is with Coinbase (we’ve partnered with Coinbase), but you still have to think a little bit about Bitcoin. And the publisher side, they’re getting Bitcoin out; we’re gonna make that easy to get fiat [currency] out.

We’d like to use something like Bitcoin though because we think there’s a future where you have a frictionless system - no interchange charge, none of the hidden charges that are associated with the credit cards where fraud sticks the merchant with the overhead or the cost of having funds clawed back to the bank. The interchange charge is like 2,15% or something (it varies), but the hidden cost of fraud is high, and a lot of merchants have to eat it. It’s not a good deal for them, especially the small businesses.

[01:04:57.10] So I think there’s something coming to the web in terms of frictionless payments, whether it’s Bitcoin, or Ethereum classic, or son of daughter of redhead’s stepchild of both… There’s something coming there, and the important properties are the permissionless property, no intermediary, frictionless property… Ideally, it would be anonymous and capable of doing micro transactions with Bitcoin as not currently. But that’s why Brave has this Brave payment solution - we solve that ahead of some next-generation solution that is coming to Bitcoin. And we want to, again, make it just work with your native fiat currency. If that happens, then I think it will be easier to micro-tip, micro-donate, have micro royalties.

Think about the Ted Nelson’s project Xanadu vision, and now think about VR if it ever takes off, or AR, because it really should be in our sunglasses. In ten years it probably will be, then all the great stuff creative people build for the augmented or virtual world - you can’t really DRM it. It’s a shared world, there’s too many eyeballs to ray trace and path trace to. You can’t say, “These are encrypted pixels and you cannot touch them, or we’ll put you in jail under the DMCA.”

All your models and your texture art - they’re gonna be out there, just like they were in Second Life. How do you protect that stuff? Well, you can watermark it; that’s a traditional method, it goes back to real-world paintings and documents. That is more of an identification system for prosecuting gross copyright violations after the fact. What if you could just have automatic micro royalties, like

Ted Nelson envisioned? People are looking at or using, or borrowing, or creative-commons-ing, meshing up some bit of art - there’s a micro royalty associated to the artist. That can be automated too, and that’s another thing that I think you can do with cryptocurrencies if you do them right. That should be part of the web standard, future AR web. I’ll pause there because I’ve said a lot, but you can see that there’s a big vision here, and I hope it’s exciting to everybody because it goes way beyond just browsers.

Eduardo Cuducos: Yeah… So I think it has a lot to do with the kind of illegal usage of public money we expect to find using data science. Because when there’s a big corruption scandal of millions, billions, maybe trillions of whatever currency you’re using, probably someone already spotted that and someone is working on it.

When we use big data, probably we are seeing a lot of small cases - a lot of small, a massive amount of small cases - that hardly ever would be spotted by a human being (that was our hypothesis), and I think it has a lot to do with the Toblerone Affair. It was a level of corruption in terms of monetary value very low… So this is the main reason.

There is a second reason that in Brazil when our (let’s say) Federal Bureau of Investigation is investigating something, of course they can’t say “I’m investigating this case” with a very meaningful name, so they just give random names; I don’t know how it works elsewhere. So calling our project Serenata de Amor was a kind of joke with those random names our FBI uses. It’s usually “Operation Something”, and something that makes no sense. Operation Sand Castle, Operation CarWash, and then we have Operation Serenata de Amor.

Adam Stacoviak: So your current trajectory was a plan B. You were just expecting to take some time off, live off the savings, and just enjoy your craft, essentially.

Beyang Liu: One year ago – 2023, right? 2023, Copilot was king, they seemed unassailable… Everyone wanted autocomplete. No one even cared about chat, or context awareness in that chat, right? Because autocomplete was where it’s at. You want it to be fast, you want it to be snappy… You want it to be a little bit context aware, but people just wanted it to be fast, right? And that was the UX paradigm that was dominant at the time.

And then in 2024 the paradigm completely shifted. I think it first shifted with GPT-4, where - okay, that was a model that had a step function improvement in its ability to one-shot applications. So now it’s like, why would I sit there, autocompleting stuff, when I could literally just generate an entire mini app from scratch? It also got a lot better at incorporating context, so now you can do kind of context-constrained code generation, which we deployed, and it had massive success in the enterprise. But then there were still certain things that it couldn’t do well. The code was often sometimes broken, or didn’t quite compile.

And then when Sonnet rolled out, that was like another step functions. Now all of a sudden these things that were kind of just beyond the capability, are now firmly within – like, I can trust this thing to reliably emit JSON, for instance. That used to be – that’s a solved problem now. People used to write entire thought pieces around “How do you constrain the model output to produce a valid JSON?” These days you don’t even think about that, because the model has been trained to treat that as a solved problem. And I think now we’re seeing it in terms of in-editor agents. That is the future, and that’s what we’re building towards. A year ago, I think if you were pushing agents, there were certain companies that were like “We’re going to build the agent that eliminates the developer entirely.” I think now people recognize that they were kind of selling beyond their existing capabilities at the time. It made for a great kind of marketing moment, and they got their name in the headlines, but there was a disappointment in the actual product experience.

Now, with the newest set of models, we’re seeing this approach, just like – this is kind of the new paradigm. That’s what we’re building for. We have kind of like a wait list of people who want to get access to the thing that we’re currently building in collaboration with a lot of these frontier shops… And it works very well in large, messy codebases as well, because again, that’s the problem that we’re solving for.

And so I think the future is very bright, and I don’t – at some point I think we might plateau in terms of model capability, but at least for the past two years, there’s just been this regular cadence of step function improvements in model capabilities. It’s just been really exciting to see that, and sort of ride the wave of the development and maturation of this technology.

Jerod Santo: Sometimes, yeah. Well, that usually happens more with an existing system that I’m trying to understand, versus one that I’m currently building. So there’s kind of two contexts. You seem to be a lot of times modifying a world that exists and is already gnarly… And so discovery is like top of mind. Oftentimes I already have most of the system in my head, and so now I’m extending and changing, and I’m making assumptions and make sure those assumptions are true or not, and so that’s the discovery… And then I’m, of course, testing out the code that I’m adding to the system as well. So there’s a little bit different mindsets there, I think.

I’ve only ever come into a system – like, remember your best/worst codebase conversation… Like, I’ve had one of those that I was just comparing it to the whole time. But mostly – because I did rescue projects for a while. Ruby on Rails rescue projects contract years ago. But most of the time the systems that I’m working on aren’t so unknown to me that I need to do too much discovery, except for when I’m trying to discover a solution perhaps to something new, you know?

Jerod Santo: Well, that leads us back to trade-offs… Because while I completely agree with you, Johannes, on all that, sometimes the best app is the one that you can actually build. And I think that’s some of what James is hitting up with… It’s like, what trade-offs is he willing to take as a solo dev trying to build whatever it is that you’re currently building, James. You’re still working on actual as an open-source, right? Are you moving on to another product now as well?

Beyang Liu: No, we’re very mindful of that. We actually integrate models in a couple of different ways. So we do it for kind of the chat-based autocomplete, there’s a separate model we use for code completions, and there’s another model that we use for embeddings-based code search and information retrieval… And it’s kind of a mix and match. Sometimes we’ll use a proprietary off-the-shelf model, other times we’ll use a model that we fine-tuned. But for the models that we do rely on external service providers for, we’re very mindful of the kind of evolving legal and IP landscape. And so one of the things that we’re currently building is basically copyright code – or copied code detection. And if you think about it, Sourcegraph is a code search engine. It’s kind of in a great position to build this feature. It’s like, if you emit a line of code, or you write a line of code that is verbatim copied from somewhere else in open source, or even in your own proprietary codebase - you might be worried about just code duplication - we can flag that for you, because we’ve been building code search for the past 10 years.

Mat Ryer: Hello. You’re a software engineer at Polar Signals, right? And also a Prometheus and Thanos maintainer, and you’re currently building an open source eBPF-based continuous profiling solution called Parka. Welcome.

Mauricio Salatino: It’s pretty wild, but I’m actually writing the last chapter now, and it’s all about that, like the thing that we mentioned before, measuring platforms, and how do we use, for example, DORA Metrics to measure performance on software delivery. So I’m currently building a small POC that will go with the book, some tutorial that basically you can just create, some sort of way of calculating these metrics on top of a Kubernetes solution. And that’s been an interesting thing, because again, as most of the things that I’m working today, they involve different communities, and the collaboration between different groups. So I’m really looking forward to finish that chapter, because it has been a lot of work, but also because it will have something that you can actually try on your own.

David Flanagan: So I am currently building the Rawkode Academy website, because I wanna make my content more discoverable. So far I’ve relied purely on just word of mouth and YouTube’s algorithm, and people searching for Kubernetes content to find it and I’ve only recently decided that actually I really want people to be able to google our Kubernetes content and get results, and that means I need a website, and building up SEO… I also wanna do more with the content. I started hiring people to type up show notes, and do editing to dense the format down to small, hopefully viral videos we can share on YouTube. So I’m starting to invest more time and energy into making the content consumable to more people… Because it’s quite hard to set aside 90 minutes or two hours for some of these episodes, and I appreciate that… But again, there’s so much to learn from the way that people handle these problems that I just wanna be able to make it consumable in maybe like 15 minutes or 30-minute chunks, for people that just want the highlights.

[32:22] The Rawkode Academy website is part of that… I have been building it, and it was online for a couple of weeks, but I’ve also been migrating all of my DNS and domains away from Cloudflare, so currently, things are in motion, but currently very, very broken.

Jerod Santo: So why did you decide to do it? I mean, you say you kind of asked yourself that question repetitively… But the very first time that you asked yourself the question and said, “Yeah, actually, I’m gonna do it”, it’s like, what was in the way? What was missing? Or what was in the way of you doing what you want to do, or you currently were working, or currently building?

Merve Noyan: In Hugging Face, there’s a certain group of people like me that do not really have specific things. In Hugging Face nobody actually tells you, “You should do XY.” You just go ahead and pick a responsibility, and that’s your thing from that moment.

What I did so far - I did the Tasks, and I did a Keras integration in which I have done model cards and a Tensor board, and stuff. I really become so happy whenever I see a Keras repository with a model card inside, because I know that people actually find it useful and just keep using it.

I am currently working on how we can use Hugging Face hub for the tabular data. I’m working with an amazing scikit-learn core contributor that is currently in Hugging Face; he’s Adrian. We are currently working on a package that is focused on how we can improve the production capabilities of scikit-learn… Because you use, for instance, Pickle, which can run arbitrary code on your machine if you just pulled any Pickle and just de-serialize it. It’s a bit hard problem to tackle. We want to post whatever information we can have about the model. Currently, I’m working on that site.

For instance, if it’s a tree-based model, you can visualize the tree. If it’s a clustering model, depending on what type of clustering model that is - it can be like a dendrogram, or visual, like with a PCA… Or if you have a linear model, you can put the hyperplane… I am trying to put those stuff, and also, what model has learned through feature importance, Shapley values and everything. What I want is I want people to call one line of code and just push their models on the hub, which will create these model cards with several information. We are all supporting Gradio, and I’m also working on how we can leverage Gradio for tabular data stuff as well. Because previously, Gradio was mainly focused on the modalities, like text or audio or computer vision. And the components, if you take a look at the documentation, are focused on that; you can just drag and drop a dataset and just automatically visualize everything regarding that dataset, and that’s quite magical.

I remember the first time I used pandas-profiling, and also data analysis baseline library… There are a couple of libraries that enable you to profile your datasets, train baseline models, do an AutoML, like TPOT… I remember using them, and I was like, “This is witchcraft. This is so good.” [laughs] It saves so much time. Recently I realized, we can actually do that on the hub.

Currently, I am building two tools. One is profiling a dataset; it’s like a Gradio interface that does it. We have recently released in Gradio something called Blocks, which is more flexible than an interface. You have tabs, and you have rows, and you can put multiple stuff inside.

I’m currently building two spaces, like I said. I either build, I maintain, or add something to the Hugging Face hub library, or something else, or do demos… Currently, I’m working on something that is like an auto EDA, like a pandas-profiling. Another thing is an AutoML tool, sort of.

These types of things also save a lot of time and also lower the barrier of entry, I think, because you have a baseline model and it will push your best baseline model to the hub, and maybe you can create a space really easily on that, which you can later go and tell your local data scientist, “Hey, I want this, but improved version.”

I really love this. I was a big fan of Hugging Face for - I don’t know, since I saw Thomas’s video. I think it’s been three years or something, I don’t remember. And I’m still a big fan of Hugging Face. I go and talk at Python conferences and people approach me and say, “I’m a big fan of your conference.” And I’m like, “Me too.” [laughter]

Adam Stacoviak: Definitely an interesting space, that’s for sure. I mean, I never really considered how complex it could be. It sounds like what you’re saying though is even by current measures, current time and market even, you’re innovating faster than Calendly. And that’s not you talking crap, it’s just simply–

Oleg Nenashev: Yes and no, because I’m currently building a prototype which integrates Jenkins, OpenTelemetry and Captain… But for me, the main objective is to actually expose more information about quality gates, so when we deliver software, we can verify that all items are basically delivered that meet all matching criteria. So currently, Captain is mostly built around cloud events, which is probably a topic for separate discussion. Captain exposes OpenTelemetry metrics on its own, so you can understand what happens inside Captain when you analyze, for example, quality gates etc. But it would also be great to have integrations in other directions. So when CI/CD systems supply information about the status, metrics and all deployment parameters, tools like Captain, so that they can make decisions, but the system is compliant with the expectations of our CI/CD admins.