Towards decentralized alternatives for code collaboration

Transcript

Wow it's just like all put your heads up at the same time that was super nice I think we'll wait like one minute or maybe we'll just get started it's 5:16 I think I'm supposed to start at 5:15 maybe I'll wait one more minute for some last-minute people to show up like that person who just walked in the door you

started um hi everybody thank you for coming I really appreciate it I'm here to chat with everybody about peer to peer code collaboration and actually the journey towards peer to peer alternatives to code collaboration so to start off my name is Abby I'm a full-time contributor atman addict and I work on the radical project which is building a secure peer-to-peer network for code collaboration so I do all things product all things community all things talking so I'm excited to talk to you guys today about what we're building at radical but before I start talking about peer to peer collaboration I'll

probably start with code collaboration as it stands today and so let's talk about code first right and so the basis of code collaboration really exists first on version control right so version control is a version control control systems are the software tools that allow us as devs to manage changes to source code and so a brief history of the evolution of version control system so we first started out with SCC s which is source code control system I think very functional name which was like a proprietary in like local solution to a version control and then throughout time we actually saw the evolution of version

control systems becoming open-source so moving from per to prior Terry to open source tools but then we also saw the shift to this client-server model right which was inherently kind of centralized and so the concept of only having a central single repository that developers could access at one time so meaning people would have to check out code so this is like with SVN check out code and then that piece of code would be locked to any other developer until that developer checked their changes back in and so around 2005 developers were like why do we need why does this need to happen like this you know we realized that we could actually have every developer could have their own

working copy of the entire source code and change history that they could push changes to continuously without having to check in or check out of a central repository and so that was kind of like the emergence of the distributed version control and so to back up this I have a short story so we all know Linus the father of Linux kernel and so Linus was actually developing the Linux kernel on this distributed version control system called bitkeeper and so bitkeeper was free for open-source projects but closed source and so linus was using this because it was the best thing to serve the really distributed nebulous development process of the Linux kernel and so what happened

was one day a Linux kernel dev being the free software vigilantes that they are decided to reverse-engineer the networking protocols of bitkeeper in turn leading bitkeeper to in spite slash the free license giving linus access to the bitkeeper product and so in that and in through that situation you know get was born right so linus then in turn developed git which is the basically distributed version control system that is like the de-facto foundation for code collaboration that was developed by linus specifically for the Linux kernel community and so with the development of git

we actually saw specifically in like 2005 I think 2008 like the rise of a bunch of other distributed version controls so mercurial kaneo bizarre darks and fossile all joined the contingent of distributed version controls and so this kind of evolution from centralized to distributed was a really big advancement in software development that did a lot of really great things for increasing the efficiency of you know how we were writing code together so well so what's great about yet I'm gonna focus on get specifically there's a lot of great things about other distributed version controls but I feel like that's a whole nother presentation so it gives great because it's fully distributed right so it allows users as I said not to work

off of this central one working repository but actually allows them to work productively offline so I can work locally I can push changes consent continuously and then I can do cinq centrally when it's needed and so this leads without having to rely on a central server for more fast and efficient development process so they don't have need to sang sync with the central server all the time I'm able to push my changes continuously and these large nebulous distributed projects are able to kind of operate as every developer is just working off of their local working copies and then pulling in people's changes you know when they're developing and so that has a really strong support for nonlinear development which Linux kernel is famous for right but the git branch model is like really

cheap and like really easy to merge so it's much easier to create Forks it's much easier to again work on your own branches and these working copies that we all have on our own machines kind of act as these remote backups for projects so we have git which is great right but what goes on top of git right so like where if we are able to now manage changes in code how do we collaborate on those changes how do we share them how do we communicate with other developers so we can actually collaborate instead of only working locally and so the collaboration layers built on top of get started with the og mailing list right so this is still the de-facto code

collaboration method for the Linux kernel development right and so that mailing lists are really great for a lot of reasons because they're self hosted right so you're not relying on any central server it's still within the ethos of having distributed workflows and developers can maintain their local working copies just Network them through another open protocol which is email and so this email layer gives us the networking so we can now start collaborating but then it also gives us identity so now a developer's identity to other developers becomes their email address and so this was really nice for again this bizarre style development in which everybody was maintaining their own branches and communicating patches

and changes and also get patches were like designed to be communicated via email so but I don't know if you've ever tried to use a mailing list or I've seen the Linux kernel mailing maybe I should move but it's super shitty to read and to manage so obviously the usability was some of these like og hacker solutions aren't really up to par and so that's why we've seen you know this the growth and the the rise of hosted platforms like github gitlab and source Hut so these hosted platforms are essentially hosted meaning you're not really self hosting your instances unless you're using like a self hosted get a git lab instance and so the source code that you're working on and these

copies that you're sharing and networking are actually networked on central servers and central repositories and so they often follow this like software-as-a-service model and so the identity isn't an email address but it's actually kind of this online identity so your identity is like locked into these platforms but it takes on this whole new you know concept of like what connections and what data is actually associated with that identity so it creates this a global networking layer that anybody can create a profile on and anybody can access but anybody can find anybody as well and that does a lot for accessibility and discoverability but of course you know these hosted platforms are really walled gardens so although they're built on open protocols and of

course leaning on distributed version control systems they build these like really unique user experiences and interfaces that give you all these new features but are like usually closed source like github so hosted platforms have obviously changed the game for code collaboration like specifically github we all know I'm sure all of us in this room have used github at least once but probably many times I mean I know I use it every day and it's changed the game for like many good reasons right so it's made these protocols use version control systems like really easy to use in these collaboration layers really easy to use because it's

transferred it's taken the standard as email lists and turned it into these like super rich web interfaces that we can look at we can we can understand and we can communicate with communicate through sorry but it's also created like a very accessible standard way of working so like once you know how to contribute or use the github UI for example you know how to contribute to any other project so like what github did was kind of break down the barriers for what it means to contribute to a project by making and giving everybody the ability to do it which has been huge for the growth of free and open source software and so on top of that if anybody can communicate from anywhere in

the world really you know how this like huge global community of developers who are all collaborating on software whether it's in public repos or private repos but everybody's building together and so you have this whole social coding aspect that has become kind of this you know now ephemeral layer that builds on top of like version control systems and that incorporates profile stars contribution crafts and like follows that totally enhances the way that we develop software together and then finally in a more functional level github actually like has enhanced our workflows right because these things like PRS and how we manage issues are now able to be not just connected to the code but also to

the people so we now have these new workflows that are like so natural to how we collaborate on code that were entirely defined by how github presented it to us and so that's like a really powerful thing especially when you start getting into like CI and like code reviews you know like github didn't just like make collaboration easier they introduced new means of collaboration and so we see that everybody really likes is like this is done a lot for software development so like there's over 40 million developers on github right now that have created over 44 million repositories and over 41 countries like that's pretty massive you know that's like pretty unreal of how global this domain this

means that production has become and it also has done a lot for free and open source and we meet it like we mean there's like very wholeheartedly right like github has done so much for the growth of free and open source software across borders across the world so this year one point million first-time contributors joined github and made their first contribution like that's huge you know and like over 350,000 github users made over five million contributions to the top one thousand projects on github you know this is not just because people want to code more you know this is because the tools in which we are collaborating and coding are becoming easier and more accessible to people around the world

and so because of this github has now become the largest repository of free and open source code not just code but collaboration right and this is where everything is happening this is where it's hosted and this is where you know people are going to to get engaged with the free and open source community so as a quick reminder does that get I think a lot of people forget this is that git is like really nothing without I mean github is really nothing without get right I mean github is essentially you know introducing all of these new collaboration things but it's also making it easier for us to write code and usually write code and manage those changes through git but it's non get

features have really overtaken the developer experience so as I said these non get native you know UI features these new enhanced workflows are absolutely overtaking and now to sucking in developers because they're creating new means of collaboration which are great because they're making everything more accessible and easier but they're also inherently creating a closed garden built on open soil right and this is extremely similar to you know these other global platforms like Twitter and Facebook that we have seen capitalize off of the emergent of like digital communities and spaces and that the early protocols free and open protocols

that the internet enabled and so although github is free you know for open source companies at the end of the day like it is still owned by a massive US corporation that is Microsoft and so I kind of want to like drop this decentralization like decentralize everything attitude because I think it's like silly to be only proposing an alternative based solely on an ideology right like of course microsoft owns github but also github works and it's done so much for the free and open-source community and also just like global software development but what we're proposing and what i want to talk about you know in

its presentation is that there are actually like real issues with hosted platforms and those issues right now are just like bubbling under the surface you know some people care about them some people don't but as we see software development like keep growing and rising as faceted pace it is today and our digital ecosystem is like ever like you know transforming and trying to like encompass like the means that it's like eating the world you know these issues are gonna become critical points of failure if they're not solved so we categorize these issues into three categories so that being platform risk vendor lock-in and security risk

so platform risk really relates to the platform that you're using to host your code and to hoster collaboration being owned by one corporation which gives that company the ability to influence your user experience and manipulate it and control it in ways that you might not really know right so github can terminate an account at any time which and they have been forced to by governments they can ban entire countries which they have based off of us sanction laws and they can change their Terms of Service at any point in time without user consent and they can also discriminate based on user behavior you know for example if a user has

visited or downloaded a certain project or logged in from a certain location these are all things that github is done right and now people are starting to realize like how that's kind of messed up to ban developers from an entire country based off of political intentions that are not related to the software development at all but it is the nature of a platform being owned by corporation so the second issue is vendor locking and so this kind of touches on how github has created this new means of collaboration right is that these non get native social artifacts and collaboration artifacts like issues comments and reviews are entirely hosted on github so you can say if github went

down I'd still have my source code I'd still have my change sets because it's local cuz you know get is distributed and I'm always having my working copy but like what would happen if you lost your issues what would happen if you lost your conversations that you're having around the pr's around the you know the issues like what would happen if you lost all of the actual collaboration that goes into code collaboration it's not just about the code anymore because what we've done is we've expanded on what code collaboration is and that expansion what we've used to extend is now entirely defined and hosted on github and also that social network so as I said that global community of developers it's just like the Facebook community right like

all of our means and our social connected connections as developers as people building projects and building code and writing code are all unique to an entirely hosted by github so your stars your likes your followers your contribution graphs you know what projects you're associated with again entirely hosted by github and then finally since github is you know a for-profit company and corporation they also really like to introduce additional offerings that try to encompass the entire development experience so once Microsoft acquired github they launched something called github actions which is offering you know CI servers so you can have every github integrated into your entire development process and so the

growth of the offering just keeps introducing like more vendor lock-in for different pieces of the process but also like more attack vectors and so that kind of leads into the third issue which is security risk so we all kind of know this you know I don't like you know security shaming anybody but if github was compromised like every single user account and project would also be compromised as well as what's shown in the UI and that's like a really important thing because not only could you access for example like all contexts of a private repo but also you could have all these different attack vectors and these are just three on this like document that we have of like ten I just didn't want to overload everybody but like for example if your commit signatures aren't verified

locally and you're just like relying on that like green check in github you know a hacker could introduce commits into the source code that appear to be verified or signed by one of the maintainer x' because it's based on the UI code reviews which aren't signed and people are starting to like sign their reviews now because of this security risk can be impersonated so a hacker can impersonate a maintainer and improve pull requests and then also like changes to the actual source code like pushes to master a master could be have appeared have succeeded or a range of different things in which a hacker would manipulate the UI you know to show a user something that they haven't so I mean I think what I'm the point of this

diving into these issues is because we have this belief that as you know more and more the digital our digital world or Dutra eco system relies on free and open source software like we believe that our dependence on centrally hosted platforms and corporations for the distribution of open source infrastructure is inherently unsustainable like if there are all these attack vectors and issues with hosted platforms and our repository of free and open source software is only growing and only becoming more critical for the resilience of our digital ecosystem how can we rely on a platform that's that's controlled by one corporation and so that leads us to one

what our thought is right is if we remember back to the beginning of the presentation git is inherently a peer-to-peer it's inherently distributed we saw a version control go from centralized client-server models to decentralized version control systems and so this shows this concept of peer-to-peer right so we believe that peer-to-peer is a new hope in the fight against the centralized hosted platform Empire that's because peer-to-peer means interacting without intermediaries right so without third parties so peer-to-peer systems are ones that we don't have to rely on a central server to operate or to use and so the coolest thing about peer-to-peer is really this concept of

resilience right so um peer-to-peer systems are economically resilient because they don't need to make a profit and they can distribute the costs of operating or using a system across an entire network they're politically resilient because they're not subject to the will of authorities and they also can't be captured and they're very technically resilient right like peer-to-peer systems are harder to attack because there's no central point of failures so they're really hard to take down the whole point is that they can survive the zombie apocalypse right so they have this robustness that makes them like very technically resilient and so these this resilience really vibes with the free and open source values right that the free and open source

movement was built off of but they also really vibe with the needs of the web 3 community in a domain in which we're building disruptive technology that is inherently combative to current systems we need to be building resilient infrastructure that can resist or may you know sustain in the resistance of central systems of people trying to fight back you know if corporations if institutions if governments want to take control how do we keep them from doing that right and and so we need resilient peer-to-peer infrastructure to secure and to ensure the health and the resilience of the free and open-source ecosystem but also the web 3 ecosystem

so on top of that the peer-to-peer ecosystem is growing like so fast you know like we're seeing not only like peer-to-peer infrastructure but also like peer-to-peer applications and communities growing and then all these block chains are popping up everywhere like peer-to-peer is like where it's at right now and the ecosystem has seen more development than it ever has before and so like a really good example of this that I want to dive into quickly to kind of like define what peer-to-peer looks like impact some and also talk about you know one that's like really um you know maintaining a lot of the peer-to-peer values is secure scuttlebutt right sis SSB is a Brazilian peer-to-peer social network that gives us as people trying to connect and share pictures and

messages with each other the ability to do so without a central server like Facebook or like Twitter and so how SSB works is through this paradigm kind of like called a social overlay so to ensure that I'm getting the data that I care about that I'm interested in what I do is I actually me as a user the protocol replicates data based off of my social graph so basically my friends who I follow who I tell the protocol I want to follow and so what the protocol does is it replicates not only my friends data but my friends of friends data so I'm getting this replication graph that is resembling also like my social graph so who I'm interested in who I want to follow the

additional layer on this is all these little nodes are backed by cryptographic appear identities so it's not an online profile defined by a certain platform you know these are defined to my device or define to me uniquely as a user and they're backed by cryptography Xand advancements in networking cryptography have given us the ability to have this additional layer of security that centralized platforms aren't really aren't able to access so um SSB is great and I know that I use it a lot more than I did before and I think it's being adopted in a lot of different sectors which is really great but we're also seeing still some challenges in peer-to-peer networks and so the first one is like availability right so though

peer-to-peer networks are always available and the data isn't always available and this is because of the nature of this like new replication you know paradigm that we're introducing so there's problems with replication lag I don't know if you've tried to open patch work after not opening it for a long time but it takes a little bit eventual consistency and out-of-date content limit peer-to-peer networks ability to maintain consistent and predictable available availability of network artifacts which is like a really important thing you know on a peer-to-peer network you need to have your data available and the second one is usability which kind of builds off the data availability thing but these protocols are able to work on offline without reliance and servers but the applications built on them lack the

ability for users to kind of collaborate or connect on a global scale right and that's because these protocols are separate of the network's and they lack the third challenge which is this concept of global seamlessness because these networks are again existing as little instances like little they call it like the cosy web like little communities that are connected to each other and replicating data based off of their networks they lack the network effects of global platforms because they're built on these separate protocols but peer-to-peer can really solve most of the problems with those platforms right so it solves platform risks because you don't have to trust intermediaries to use the network and it can't be censored by a third party so

you know Microsoft can't come in and just tell you to ban developers from an entire country and that's like a really huge deal it solves vendor lock-in because users own control and moderate their own data and this is really cool because although there's an issue with global seamlessness and peer-to-peer networks I'm also in control of the data that I'm disseminating and replicating across the network and also what data I'm replicating on my own machine and so this gives users control over their network instead of just relinquishing control to a central platform and then finally security risk right so this introduced a introduction of cryptographic peer identities like no central servers to rely on and really

inherently like local first data that's built on and built up from the local users machine introduces this new layer of security that as I said like central platforms just can't access right now and actually solves a lot of these security risks that we see when we're relying on platforms to deliver us close source experiences and that are hosted and controlled by one server so to return again to the beginning of this presentation as with the rise of distributed version control get really did decentralize code collaboration but then github resent realized it and so as we see peer-to-peer tech evolve in advance

we start to see the unique functionality that hosted platforms deliver like network effects rich user interfaces like community displaced by peer-to-peer systems because we have the ability to build usable and functional robust networks that totally disrupt what centralized platforms built their entirety of their you know their product on so we ask the question if peer-to-peer tech is evolving and if code collaboration was inherently peer-to-peer in the first place why can't we have peer to peer code collaboration now so we want to talk about how we have peer to peer code collaboration so to do that we talk

about peer to peer code collaboration in theory we believe that a peer to peer code collaboration appear to be our alternative for code collaboration must satisfy a certain set of requirements right so it must work offline and not require DNS or online portals to function it must not require trust in a third party or intermediary to use must exceeds current security standards set in place by mainstream platforms because they're just not good enough right now and it must meet the usability meaning like responsiveness functionality user experience that user interface that is just kind of the standard for cota collaboration now it must meet those standards that are set in place by hosted path perms and then finally must be audible transparent and unsensible

and it of course it must be free and open source software so if we have these requirements in mind how would we design peer-to-peer code collaboration and practice and how are we at radical doing so so it all starts with git as you can tell we really love git and that's because git is actually such a great foundation for peer-to-peer replication because it's offline first so git is inherently optimized for full replication and it doesn't assume having an online server it actually has networking built-in which a lot of people don't know but this offline first quality means that get gossips really well it's also designed for the bazaar model which is

very similar to how peer-to-peer networks are structured and designed right now and then finally PAC files which is more technical reason but the way that data is stored and get changes in code or stored or through these like really awesome things called PAC files that make the protocol really fast and makes thinking really fast and so PAC files are actually the reason why us at radical keep the storage layer of git instead of replacing it with something like ipfs because git is just really really good for storing code and so how would we replicate git repos across the network so again we start with git right so it stays the same we still have this

local working copy that we push changes to but it's paired with a peer identity which is based off of a cryptographic you know like an ID right and that is linked to my device key so at the start everything is the same when we want to replicate it what we do is we don't just replicate this local working copy we actually create this like hidden replicated copy that not only includes the original change set the original source code but also the remotes meaning other people's you know other people's working copies of an original source code we include the remote so we include that data and then we also include the remotes of the remotes so just like in

SSB as you're replicating the friends and the friends of the friends we're replicating the remotes and the remotes of the remotes and then it's all still tied to the peer identities of the original authors of the code that's being replicated and so what this does is it gives us a social overlay forget right just in the same way that SSB replicates data through friends and through your social graph we're replicating through the social graph of the code so of how people are taking the original source code forking it creating their own working copies making changes how people develop you know code right now right the different branches and everybody has their own change sets and

we're replicating that all together so when you're descent dating data across a network your disseminating not just the original working copy of you the user but also everybody else's that's associated with it and so to to do that you know we designed this kind of object called projects right so if you want to publish a git repo to a network what you as the founder of this repo the original author does is initialize a project and so what this means is that you're linking your original working copy or original chainset to this unique like metadata file that is not only links it to your peer identity but introduces more actual

metadata that can be associated with the project which results in you can see at the top of their a project ID and so this project ID gives you the ability to also have other information on their projects such as maintainer keys so if you have multiple maintainer sewer projects you can actually have those linked to the project metadata file giving us as someone viewing a project the ability to not only verify the authorship of a repo or a project but also the maintainer and also any contributors associated with the project because contributors when they contribute also have their own signing key so what you need to know from this is that projects which are replicated

across the network not only just include the code but include all these connections to different peer identities which give the network the ability to verify the authenticity of the code of the people writing the code and how it's been replicated across the network and so really what you do is you announce your changes you push them to the network and just using get like a get fetch with that project ID you're able to pull in that project verify its metadata verify that it's the project that you actually want to contribute to or use and then replicate all the source code and then when you have the source code you also have the remotes of the remotes and you have all of this other you know data the social

overlay of gets you know also pulled in to your local machine where then you can interact with it locally and offline and so in addition to solve the global seamlessness issues that we were talking about before you could introduce a global canonical layer that connects these projects and these users the peers in the network with each other and so to do that you know we propose a totally ordered consensus back to registry which holds that canonical project metadata that I said that you initialize when you start a project and you publish it to the network and it links them together so this gives us the ability to have this global layer that for example

github has the network effects right if I'm able to you know you need this global layer because if I'm not able to search or discover a project that I want to contribute to then what is the value of the network in the first place and so this global canonical layer solves that for this code collaboration use-case by connecting projects to each other but also gives people the ability to have an audible and transparent way to verify again that maintainer ship project histories and also like different contributions so this adds this other security layer on top of the cryptographic peer identities and so these components together really make up radical and so radical is a secure

peer-to-peer network for code collaboration and so we call it radical because tis the root of code club you know radical is the foundation just as git is the foundation for current code collaboration we want to keep building on that be the first route that sprouts from the seed for peer to peer code Club and so as I said really the stack is building on top of git and so we're really just extending it with peer-to-peer application so we don't replace a storage layer we introduce a new networking layer on top and it's a peer-to-peer replication layer that works just like SSB designed a little bit differently but has the same like social you know replication paradigm where we disseminate projects through a

network and replicate them through remote server motes and it's all tied together with a global project registry which connects the canonical projects metadata and gives us the ability to search and discover projects just like we would on github and so why is this cool so this doesn't only first it's peer-to-peer right so like inherently it's solving a lot of the issues with hosted platforms because it's giving us the ability to collaborate on code without trusting third parties or intermediaries and we don't have to set up our own hosting right so it's kind of this out-of-the-box ability to share and network with other developers but there's also like a lot of really new nice things that having appeared this

peer-to-peer solution offers us right and so that's the ability to have everything be offline so this radical developer experience is inherently offline and local so not just my source code like my source code is not just the only thing that's local anymore right this gives us the ability to have our issues our comments our reviews these collaboration layers use new means of collaboration that hosted platforms have put in place we take them and now we're able to make them local again and we're able to interact with them local so we make everything available so if github is down it's not like you can't access your PR is or your issues right you have them and it's available offline without trusting or relying on a hosted platform and then as I said it also introduces

this new layer of security so we have these security ride entities which are linked to projects and linked to code but also all of these artifacts getting these issues and these reviews and these PRS are all now cryptographically signs so you can introduce security at different layers and introduced it at the different collaboration primitives so it's not just like your commit signing anymore you know what I mean their commit signing but you're also signing issues you're also signing reviews and so you're introducing this layer of verification that again hosted platforms can access right now because you're relying on them to show you the green check or to verify it for you and as I said it's just peer-to-peer so now getting into the project registry

which again is this global seamless layer built on top of the peer-to-peer replication layer so why do we need a totally ordered consensus backed registry okay a blockchain protocol and so there's like three reasons one of them is that it solves it provides like strong consistency so again it like layers in with the peer-to-peer network and gives us the ability to ensure like more predictable and consistent availability of objects across the network because you're linking them in a global way not just in a peer-to-peer way and then the second it provides a global seamlessness layer the peer-to-peer networks are lacking right now it introduces discoverability and

canonicity in a way that's necessary for code collaboration because we need to have these open borders we need to have these ways of accessing other people's code in other people's projects and so this global seamlessness layer is really important but it also gives us the ability to like verify at a global level and so third which I'm going to focus on is this concept of like persistent and resilient sovereign identity so if we talk about our identity on github so like what is our developer identity this is one of our founders github profiles our identity is really defined by like UI components right it's our profile it's our avatar picture it's our handle

it's our follows it's how many stars we have it's a projects that were associated with it's like how many contributions we have in the last year like I don't know like this is like such a means of like defining you know someone's experience or their work but this identity again is all defined by a UI a user interface a clothes garden that's taking in our data delivering us you know a reputation and that is now our identity and more and more this is becoming such an important part of our like real identity like meatspace identity you know this is how we communicate this is how we work together this is how we judge other people's work this is how you know we find other people to work with and so this idea idea of having like the free and open

source developer identity hosted on a central hosted centrally hosted platform a little bit unsustainable and so what this global project registry does is gives us the ability to have a sovereign developer identity to have something that is persistent and exists outside of a platform outside of a user experience and it's completely managed controlled and owned by me as my user I'm able to retain and have that autonomy to manage how my data is spread across you know a network and so building up from that identity you can introduce some really cool attributes by introducing like sovereign collectives so if you build

organizational functionality on top of these trustless sovereign identities you can start creating new ways to manage access control in ownership you know which are becoming as we see free and open source projects becoming less of just you know a code repository and more of a movement and more of a community you know the ways that we interact as an organization or community are so important and becoming extremely like the concept of governance and like transferring ownership and like how we actually manage a project as a maintainer is becoming like very core to how we you know thrive in a free and open-source ecosystem and so these sovereign collectives gives us new means

to manage and new means to control these aspects of you know free and open-source communities that people kind of are struggling to do right now and then third it introduces this sustainable layer right so like like having a global project registry enables interoperable value exchange and so this completely opens up you know what it means to introduce sustainability into free and open source into the freedom of and source ecosystem like we know that this is an issue that free and open-source maintainer x' and contributors are underfunded and unsustained because the volunteer model that we have in place right now just like doesn't vibe with like capitalistic incentives right and

so by introducing value exchange like direct borderless peer-to-peer value exchange you completely like open up what it means to create like cut what it means to support commons-based peer production in free and open source software development and that's like a really really strong value prop that we're like heavily exploring with the introduction of this global project registry is like what does sustainability look like in radical and how can we support it so this is what we have now and it's technically really great where we're you know we want and believe that radical provides a robust usable and functional alternative to code collaboration but it doesn't really stop there so like what

we're doing right now is focusing on building a usable alternative to code collaborations so this is a screenshot and I was gonna do a demo for you guys today but I didn't think I'd have enough time it provides us the ability to have like a front-end client that we can use that we can use just like we use github and so this is what we're building it's called upstream and we're building this for radical and for the radical Network giving developers the ability to actually you know collaborate like they do now on github just in a peer-to-peer way so I'm probably I think I'm over time already so that's all I have but this is the link to our discourse we actually host most of our development

conversations in the open so you can actually hop in there to see what our developers are working on day to day and get more updates on what we're building and what why we're coming out with like more of a public timeline very soon so be sure to follow us on Twitter which is at radical underscore XYZ and I have to hop into another talk right now that I'm late for and that you can follow me on Twitter or ping me on telegram if you want to chat more thank you [Applause]