Fork This: Fixing Open Source’s Broken Incentives in a Web3 World | Max Howell - Tea Protocol

good morning everyone uh thanks for joining me here this morning I know it's pretty cold in here I wasn't planning on wearing a sweater uh I'm Max how co-founder of tea protocol and creator of home brew and I'm here to talk about fixing open source's broken incentive model in our web 3 world the first a bit of story time uh some time ago when I was young and foolish bit naive I thought that I could write a package manager it was 2009 and I was working at startup in London and we developed apps for six different platforms Windows Linux Mac Android Linux uh Android iPhone phone and even a Blackberry app and we used Mac as this unifying platform to develop for all of the other platforms it was the one that you could do everything on but the package Management Solutions weren't that great I used to complain about it a lot to my co-workers to the point that they got fed up with me continuously complaining about it and challenged me to do something about it so I went home that day and made the first commit to home brew this is a video showing the commits to home bre over time as you can see initially it was just me cuz all open source starts off with just one person with an idea something that maybe could turn into something bigger but I managed to get some attention and suddenly it was overwhelming there was thousands of people contributing to home brew and uh it was very rewarding to have this happen but uh I started having to dedicate evenings weekends lunchtime getting up early doing extra work so that I could work on home brew and handle all of these pull requests and all this amazing amount of contribution after a few months of that even even that free time wasn't enough and I decided I wanted to quit my job so I could work on Homebrew full-time which became this cycle in my career where I would uh save up some money uh quit work on open source and home brew full-time run out of money and then find a job again uh usually a contract work or maybe working for a startup something fun but you know what I wanted to do was work on an open source and over the years it became apparent that uh web 2 was having a success built on top of Open Source open source was the layer that allowed them to make uh the the world that we know today you know 15 years ago open was still considered kind of Niche and it was over those years that it became more and more the foundation of the world we know today meanwhile while they were making uh fortunes I was uh working paycheck to paycheck in order to maintain this open source and my story is far from unique almost all the big open source projects have a very similar founder story and you know something's just not right with that there's so much money in the software industry but so little of it trickles down to the open source Foundation open source is everywhere nowadays it's been to Mars it's at the bottom of the ocean it's in every pocket of every human on the planet and yet uh the uh open source developers I've spoken to over the last couple of years as I've developed outt protocol uh are usually earning like very very little money at all if anything from donations or bounties or any of the existing systems that are there even though 97% of Commercial Code contains open source as we know a recent study concluded that open source is worth almost $9 trillion a rather impressive amount but really it makes sense that's the case right and uh the money that is being made here is not trickling down this is a famous XK CD comic that's actually almost 20 years old at this point uh this is a problem we've known about for an awfully long time everybody knows about this problem the fact that all modern digital infrastructure is built on top of this shaky Foundation of Open Source that's maintained by people that are barely able to find the time to do so the comic points to this pillar at the bottom that is thanklessly maintained by some random person in Nebraska and so we've adopted this we call this the Nebraska problem the Nebraska problem is a fundamental uh problem with open source it's making the internet fragile it's making it insecure and this is just coming more and more to a Crux every year that this continues I uh love this quote from Dennis pushkov who is the maintainer of Cs CJs is a typical Nebraska project buried deep in the stack of every single node application that exists you can't really make know application without it it's downloaded billions of times a week and yet he uh so you know uh slaves away maintaining this thing with almost no support from the millions of companies that build on top of it ffmpeg tweet this every week all modern video software that we use to watch Netflix and the like is built on top of their work they've been building this stuff for more than 20 years it's in incredibly impressive software uh they have a tongue and cheek attitude about it as all open source maintainers kind of do cuz at the end of the day we didn't make open source to make money that wasn't the point we made open source because we wanted to make something cool that we hoped other people would find useful push the uh the bubble make the world a better place make World a more capable place and yet it would be nice to be able to work on open source full-time that's what we want I would love to be working on open source full-time right now I started T protocol because actually I didn't want to start T protocol I wanted to work on open source fulltime around the time we were getting funding this classic exploit happened which was great for us getting funding instant where uh the whole internet ground to a halt for a few weeks because there was this massive exploit found in log 4J just some logging software that's used on like just about every Java server that exists and like Java servers were rout exploitable in day one the maintainers received a huge amount of flack on social media about this they uh said look it's our fault there's bug uh but we barely have time to maintain this thing we fix the bug but maybe some of these companies that use our software could you know send us some donations or help fund us to work on it more time after we fixed it and yet the world has basically forgotten about this Nebraska project few years ago I was once again trying to work on open source fulltime and I looked to see if there was anything new in the space to try and make it so people like myself could do that and I didn't find anything new and I tried the existing Solutions like begging and trying to get bounties applied to my software and then making a patreon and begging people to fund me I uh had a patreon that got up to about half my rent at one point but half my rent is as good as zero rent really so uh I realized that there was nothing new but maybe it was time to try something that was new and maybe the person who would have to do that would be me so I went deep deep in the rabbit hole I didn't really know much about crypto before I looked into this but I discovered that crypto had special characteristics that maybe could change this situation could maybe fix the Nebraska problem so I co-founded T T's mission is to fix the Nebraska problem to sustain the open source that sustains us we're big Believers in open source and its power and its utility and its need to have these problems fixed T's got four main pieces main one obviously is to incentivize maintainers by getting them paid we got onchain governance which you know most open source projects have some kind of governance it's what emerges as the projects get bigger the need for someone to be making decisions in a democratic fashion we have a system that we're calling trai which is one of the key pieces it ranks all 10.5 million open source projects according to how impactful they are to the entire world's open source usage and we're also doing gpg commit rewards people have been signing their commits in open source for more than 20 years but those and that gpg signature is like the developers identity we're tying that into our Network so that we can actually reward People based on contributions in the future and in the past we've been running a test net now for a while but we recently launched it as a layer 2 and we're Amazed by the results we have more than 160 million wallets have been created this is just over a month old 460,000 daily transactions going on and we've had 17,000 open source projects register over the last year with our protocol so that they can get rewarded by the the uh the network incentives when we launch main net which should be happening pretty soon CH as the uh the key piece when I uh was coming up with the ideas for what T would be I realized that there was fundamental reputation data encoded into open source all open source projects have dependencies where they depend on other open- source projects that have done some of the work they need so they don't have to reinvent the wheel this dependency statement is like a reputational statement you're saying that your dependent uh your reputation depends on that of your dependencies the things you depend upon you wouldn't make those choices casually it's important you're not going to pick something that's going to have an exploit in it or that you trust the maintainers to keep exploits out of their code and not inject them you trust the people that you're depending upon to do a good job with API stability ABI stability these things mean that you're project is more important more impactful so we assign a t rank to every project in the graph or 10.5 million of them from 0 to 100 giving the relative impact of each project allowing us to reward projects based on how important they are you can't just spin up a new open source project tomorrow and hope to be rewarded the gpg signature technology that we developed is uh really quite exciting I I'm very pleased that we came up with this and we're applying it cuz the gpg signature is like a developer identity and it means that if you commit across lots of different projects or just like throw little commits in here and there you're contributing to open source and we can recognize that you don't have to create projects a lot of people just use projects and fix them those fixes can be rewarded there's no need for the project to figure out how to reward people who submit pull requests also means that past contributions can be rewarded I'm uh looking forward to finding out how much L tools might be uh on the line for considering he's been signing his commits since he invented git and before when he was working on the Linux kernel one of the most important developers of all time and he will have his uh just reward for it in order to encourage people to sign their commits because honestly not enough people do sign the commits we recently released this t-base which is what I call the developer cockpit it uh enables you to very easily make your commit signed and secures your gpg private key in the most secure place that you have on your Mac which is the secure Enclave in the keychain it also does a bunch of other stuff that is just really useful for developers or it enables new workflows and new possibilities we love open source and we want to empower open source both financially and through utility by making new tools that help another thing that we've been doing is building out uh a series of tools for the packaging ecosystem I was this time around old enough and perhaps foolish enough to believe I could make an even better package manager than home brew and I believe we've succeeded package X and its Suite of tools is a wonderful packaging experience you just type what you want and you get it and it runs we have four main tools in the set package X which runs packages really fast re written in Rust recently so it's just like lightning fast and Tiny this Standalone binary you can download now we have Dev which allows you to have virtual environments for each of your projects that you're working on Super useful for open source people but also for teams working in businesses that can like then uh make sure the versions of all the tools for the whole stack they're using are pinned we have MASH which is a package manager for scripts cuz as we wrote package man package X we realize that script scripting was one of the most powerful parts of what it can do you put package X in the script shebang and you can add anything from the open source ecosystem into that script without having to worry about how the end user of your script is going to install things and finally we have package M which is basically Brew 2 so thanks for attending my presentation we're building T um which is designed to fix the Nebraska problem for open source source to make it so that people who make open source are sustained as they sustain us and look out for our main net which uh is really happening any day now thank you [Applause]