1 00:00:25,780 --> 00:00:32,560 Hello and welcome to today's episode of SolarCast 2 00:00:32,560 --> 00:00:37,760 and with us today we have two developers 3 00:00:37,760 --> 00:00:40,360 and the core maintainers from Radical. 4 00:00:40,700 --> 00:00:46,140 Unfortunately in this episode today my main recording 5 00:00:46,140 --> 00:00:49,860 microphone broke down so what you'll be hearing 6 00:00:49,860 --> 00:00:54,080 is my backup microphone and luckily that was 7 00:00:54,080 --> 00:00:57,740 still going but the audio quality is maybe 8 00:00:57,740 --> 00:01:00,980 not up to par for the usual episode. 9 00:01:01,980 --> 00:01:05,280 Just a forewarning but I think it's still 10 00:01:05,280 --> 00:01:05,800 good enough. 11 00:01:05,920 --> 00:01:07,760 I did my best to edit it so 12 00:01:07,760 --> 00:01:10,220 it will be a pleasant enough conversation. 13 00:01:11,160 --> 00:01:14,740 And the conversation that we'll be hearing will 14 00:01:14,740 --> 00:01:17,160 be going through a lot of different aspects 15 00:01:17,160 --> 00:01:19,920 and you'll be kind of with me on 16 00:01:19,920 --> 00:01:23,860 the learning journey of what it means to 17 00:01:23,860 --> 00:01:28,320 do peer to peer and local first Git 18 00:01:28,320 --> 00:01:31,500 collaborations and forages. 19 00:01:32,120 --> 00:01:35,700 So if you are as curious as I 20 00:01:35,700 --> 00:01:40,800 am and was then tag along and let's 21 00:01:40,800 --> 00:01:42,060 dive in. 22 00:01:47,680 --> 00:01:50,440 Thank you so much for joining today for 23 00:01:50,440 --> 00:01:55,280 this episode of SolarCast and with us here 24 00:01:55,280 --> 00:01:57,800 today for those of you who cannot see 25 00:01:57,800 --> 00:02:02,820 us is Lawrence and Fintan from the Radical 26 00:02:02,820 --> 00:02:03,460 project. 27 00:02:04,660 --> 00:02:10,380 Radical.xyz. So to get to know y 28 00:02:10,380 --> 00:02:12,520 'all a little bit would it be okay 29 00:02:12,520 --> 00:02:14,860 that you tell us a little bit about 30 00:02:14,860 --> 00:02:17,640 how you entered Radical and what your backgrounds 31 00:02:17,640 --> 00:02:18,040 are? 32 00:02:18,420 --> 00:02:23,840 Yeah so mine was like a long winding 33 00:02:23,840 --> 00:02:30,720 journey of getting to work on like functional 34 00:02:30,720 --> 00:02:36,840 programming during college times, working remote for a 35 00:02:36,840 --> 00:02:38,560 company that actually used Haskell. 36 00:02:39,200 --> 00:02:41,300 So yeah I became enamored with Haskell in 37 00:02:41,300 --> 00:02:46,760 particular during my university days and that eventually 38 00:02:46,760 --> 00:02:48,560 led me to OSCoin. 39 00:02:50,380 --> 00:02:54,180 And that happens to be the kind of 40 00:02:54,180 --> 00:02:56,820 like roots of Radical so to speak. 41 00:02:58,440 --> 00:03:01,960 So we kept in touch and I was 42 00:03:01,960 --> 00:03:03,440 interested in the project. 43 00:03:04,180 --> 00:03:06,900 It was kind of the idea of OSCoin 44 00:03:06,900 --> 00:03:11,040 was to try and bring sustainability to open 45 00:03:11,040 --> 00:03:15,100 source projects through a kind of like weighted 46 00:03:15,100 --> 00:03:23,060 graph system, cryptocurrency and basically if you're dependent 47 00:03:23,060 --> 00:03:27,820 upon a lot then your project would theoretically 48 00:03:27,820 --> 00:03:29,800 get rewarded a lot. 49 00:03:30,500 --> 00:03:34,420 So there's this classic xkcd comic with the 50 00:03:34,420 --> 00:03:36,560 stack of blocks and we have the tiny 51 00:03:36,560 --> 00:03:39,220 little block holding up everything in the corner. 52 00:03:40,620 --> 00:03:43,120 So in theory that block would be getting 53 00:03:43,120 --> 00:03:44,840 a lot of funding to make sure that 54 00:03:44,840 --> 00:03:46,820 the whole stack doesn't collapse. 55 00:03:47,760 --> 00:03:51,520 They realized that not only is funding a 56 00:03:51,520 --> 00:03:56,340 problem but also the like collaboration on code 57 00:03:56,340 --> 00:03:59,300 artifacts and repositories seem to be a problem 58 00:03:59,300 --> 00:04:05,920 as well because it's rooted in centralized companies 59 00:04:05,920 --> 00:04:08,800 and centralized countries let's say. 60 00:04:09,200 --> 00:04:13,500 GitHub being based in America under Microsoft. 61 00:04:15,720 --> 00:04:19,279 So early indications of what we're seeing now 62 00:04:19,279 --> 00:04:24,760 in this like shift towards more sovereign solutions. 63 00:04:25,120 --> 00:04:27,180 So like trying to break away from the 64 00:04:27,180 --> 00:04:27,520 US. 65 00:04:30,540 --> 00:04:33,900 Yeah and it was right about then that 66 00:04:33,900 --> 00:04:34,440 I joined. 67 00:04:34,600 --> 00:04:36,360 That project was kind of like the first 68 00:04:36,360 --> 00:04:41,800 iteration and after that it was a Rust 69 00:04:41,800 --> 00:04:48,300 project and it was the second iteration of 70 00:04:48,300 --> 00:04:49,780 what we call Radical. 71 00:04:49,780 --> 00:04:53,400 And then the name was Radical Link and 72 00:04:53,400 --> 00:04:56,720 instead of a kind of Lisp machine it 73 00:04:56,720 --> 00:04:57,900 was kind of back to basics. 74 00:04:58,600 --> 00:05:01,120 We've got Git, why don't we build on 75 00:05:01,120 --> 00:05:01,620 top of it. 76 00:05:03,080 --> 00:05:06,460 So that's around when I joined about six 77 00:05:06,460 --> 00:05:07,600 and a half years ago. 78 00:05:09,380 --> 00:05:11,540 I've been working on things since then. 79 00:05:12,260 --> 00:05:16,480 I can get into the why a bit 80 00:05:16,480 --> 00:05:18,400 more but maybe I'll hand it over to 81 00:05:18,400 --> 00:05:21,340 Lorenz to kind of give his perspective of 82 00:05:21,340 --> 00:05:23,680 why he got into this. 83 00:05:24,380 --> 00:05:25,260 Thank you Fintan. 84 00:05:26,680 --> 00:05:28,660 Hello to all our listeners. 85 00:05:29,100 --> 00:05:32,740 I am Lorenz and I joined the project 86 00:05:32,740 --> 00:05:34,040 quite a bit later. 87 00:05:34,320 --> 00:05:36,800 So I was not around in this OS 88 00:05:36,800 --> 00:05:39,480 coin phase and the let's say the earlier 89 00:05:39,480 --> 00:05:43,520 explorations and iterations of the project. 90 00:05:43,760 --> 00:05:47,380 I joined quite a bit later just as 91 00:05:47,380 --> 00:05:49,540 a open source contributor. 92 00:05:51,320 --> 00:05:55,080 So I first started contributing just out of 93 00:05:55,080 --> 00:05:58,100 interest because I liked the idea. 94 00:05:58,400 --> 00:06:03,380 I was actually as GitHub got acquired by 95 00:06:03,380 --> 00:06:07,640 Microsoft a few years back I started wondering 96 00:06:07,640 --> 00:06:12,980 okay like where is this going and I 97 00:06:12,980 --> 00:06:17,680 figured why aren't there more projects that work 98 00:06:17,680 --> 00:06:21,140 on the decentralization aspect and making the decentralized 99 00:06:21,140 --> 00:06:25,800 components of Git which are really strong more 100 00:06:25,800 --> 00:06:27,360 usable and more user-friendly. 101 00:06:28,440 --> 00:06:32,300 Because the reason is that the underlying situation 102 00:06:32,300 --> 00:06:35,740 is that Git actually is very decentralized but 103 00:06:35,740 --> 00:06:39,520 the dynamics in the real world have unfolded 104 00:06:39,520 --> 00:06:41,220 in a way that we see lots of 105 00:06:41,220 --> 00:06:44,120 concentration so we see lots of people gravitating 106 00:06:44,120 --> 00:06:47,840 towards GitHub big part of the open source 107 00:06:47,840 --> 00:06:51,500 ecosystem and a few other forges but GitHub 108 00:06:51,500 --> 00:06:52,800 is the biggest by far. 109 00:06:54,360 --> 00:06:56,760 So at that time I didn't even really 110 00:06:56,760 --> 00:06:57,880 know about Radical. 111 00:06:59,080 --> 00:07:04,200 Blockchains were more prominent but I had a 112 00:07:04,200 --> 00:07:09,420 few questions and reservations there so I didn't 113 00:07:09,420 --> 00:07:14,260 really engage much in these projects but then 114 00:07:14,260 --> 00:07:18,040 at some point I learned that Radical actually 115 00:07:18,040 --> 00:07:22,100 switched the implementation to Rust and switched to 116 00:07:22,100 --> 00:07:24,880 one that is not dependent on the blockchain 117 00:07:24,880 --> 00:07:28,220 but is more open to like opt-in 118 00:07:28,220 --> 00:07:30,380 in the future or build on top but 119 00:07:30,380 --> 00:07:32,900 certainly not in the core of it as 120 00:07:32,900 --> 00:07:33,960 it was previously. 121 00:07:34,280 --> 00:07:37,620 In fact the situation that Fintan mentioned before 122 00:07:37,620 --> 00:07:41,100 about the graph to distribute the resources and 123 00:07:41,100 --> 00:07:43,940 funds among open source communities this is a 124 00:07:43,940 --> 00:07:46,080 project that spun out into so this still 125 00:07:46,080 --> 00:07:49,160 exists it's called DRIPS and we say it's 126 00:07:49,160 --> 00:07:52,600 our sister project but that kind of was 127 00:07:52,600 --> 00:07:54,820 realized that it might be better to split 128 00:07:54,820 --> 00:07:56,820 these two aspects and I do agree. 129 00:07:57,800 --> 00:08:01,160 So after that period I decided okay now 130 00:08:01,160 --> 00:08:04,340 is the time that I start contributing to 131 00:08:04,340 --> 00:08:06,060 one of these projects and then I started 132 00:08:06,060 --> 00:08:07,340 contributing to Radical. 133 00:08:07,820 --> 00:08:11,160 As usual you know you find those things 134 00:08:11,160 --> 00:08:14,380 that you think could be improved you start 135 00:08:14,380 --> 00:08:17,980 with smaller ones then you somehow understand the 136 00:08:17,980 --> 00:08:20,700 code base you find a few bigger ones 137 00:08:20,700 --> 00:08:24,840 and that's how I grew into a more 138 00:08:24,840 --> 00:08:29,520 senior contributor to the project and then at 139 00:08:29,520 --> 00:08:32,679 some point I was asked whether I would 140 00:08:32,679 --> 00:08:36,299 like to maintain the project together with Fintan 141 00:08:36,299 --> 00:08:40,840 and Adrian and yeah that's what I'm doing 142 00:08:40,840 --> 00:08:41,240 now. 143 00:08:41,960 --> 00:08:45,320 You mentioned something there as usual and then 144 00:08:45,320 --> 00:08:48,020 you describe this process of like finding these 145 00:08:48,020 --> 00:08:49,940 different projects that you want to contribute to. 146 00:08:50,160 --> 00:08:53,020 When you say as usual who are you 147 00:08:53,020 --> 00:08:55,440 talking about here like as usual? 148 00:08:55,680 --> 00:08:58,400 Yeah it's probably colored by my own view. 149 00:08:59,220 --> 00:09:02,080 I would say as usual for really an 150 00:09:02,080 --> 00:09:05,200 open source enthusiast you know the kind of 151 00:09:05,200 --> 00:09:07,780 people that use open source software and if 152 00:09:07,780 --> 00:09:10,540 they see an issue or something that could 153 00:09:10,540 --> 00:09:13,340 be improved they really go clone the code 154 00:09:13,340 --> 00:09:16,600 figure it out do the fix and so 155 00:09:16,600 --> 00:09:20,380 on and yeah I have done that in 156 00:09:20,380 --> 00:09:22,920 like a handful projects those that I use 157 00:09:22,920 --> 00:09:25,220 personally and I like to do that when 158 00:09:25,220 --> 00:09:27,940 possible but of course it's not always possible 159 00:09:27,940 --> 00:09:30,480 to start with when the software is not 160 00:09:30,480 --> 00:09:30,780 free. 161 00:09:31,180 --> 00:09:34,600 So I've been having a lot of conversations 162 00:09:34,600 --> 00:09:38,580 with some friends recently who are new in 163 00:09:38,580 --> 00:09:40,140 the open source space and they want to 164 00:09:40,140 --> 00:09:43,640 start exploring and I think one of the 165 00:09:43,640 --> 00:09:45,380 challenges is kind of what you touched upon 166 00:09:45,380 --> 00:09:47,780 here is like finding what project to contribute 167 00:09:47,780 --> 00:09:51,980 to and so we touched on it or 168 00:09:51,980 --> 00:09:56,220 so why did you find that Radical was 169 00:09:56,220 --> 00:09:58,660 the project worthwhile to contribute? 170 00:09:58,860 --> 00:10:02,700 I think that Radical is pretty far out 171 00:10:02,700 --> 00:10:06,000 there in terms of ambition and also in 172 00:10:06,000 --> 00:10:12,380 terms of challenge in bringing Git back closer 173 00:10:12,380 --> 00:10:14,620 a bit to its roots or at least 174 00:10:14,620 --> 00:10:18,520 how we see it bringing this decentralization aspect 175 00:10:18,520 --> 00:10:18,920 back. 176 00:10:19,140 --> 00:10:22,580 So there are quite a few open source 177 00:10:22,580 --> 00:10:25,460 projects out there in terms of different forges 178 00:10:25,460 --> 00:10:29,840 for example many many will know Codeburg and 179 00:10:29,840 --> 00:10:34,900 that they use ForgeJoe others might know SourceHUD 180 00:10:37,500 --> 00:10:42,060 but as far as I know Codeburg is 181 00:10:42,060 --> 00:10:45,200 an association so they are structured as an 182 00:10:45,200 --> 00:10:48,740 association registered in Germany they invite new members 183 00:10:48,740 --> 00:10:51,980 to join and the core offering is that 184 00:10:51,980 --> 00:10:55,760 they host infrastructure for you to collaborate on 185 00:10:55,760 --> 00:10:58,620 open source code open source software free software. 186 00:10:59,300 --> 00:11:01,160 And they're kind of like GitHub but open 187 00:11:01,160 --> 00:11:01,520 source? 188 00:11:02,360 --> 00:11:06,160 I mean I wouldn't go that far I 189 00:11:06,160 --> 00:11:10,460 wouldn't say they are like GitHub because GitHub 190 00:11:10,460 --> 00:11:13,740 is yeah is a bit of a different 191 00:11:13,740 --> 00:11:17,880 topic I guess but Fintan yeah please take 192 00:11:17,880 --> 00:11:18,280 it away. 193 00:11:18,680 --> 00:11:21,360 Maybe it's good to just clarify what like 194 00:11:21,360 --> 00:11:27,360 a GitForge means in general so most people 195 00:11:27,360 --> 00:11:30,540 my favorite comment is like Git it's already 196 00:11:30,540 --> 00:11:35,060 decentralized when we talk about Radical and we 197 00:11:35,060 --> 00:11:37,620 fully recognize that of course it is it's 198 00:11:37,620 --> 00:11:44,120 a decentralized tool and how developers started using 199 00:11:44,120 --> 00:11:47,160 it is through a decentralized workflow called email 200 00:11:47,160 --> 00:11:50,680 so the Git core team and the Linux 201 00:11:50,680 --> 00:11:56,440 team use email lists for collaborating through Git 202 00:11:57,160 --> 00:12:01,920 and so email can be one Forge and 203 00:12:01,920 --> 00:12:06,380 that the like step after that was SourceHUD 204 00:12:06,380 --> 00:12:09,820 which is an email based Forge where you 205 00:12:09,820 --> 00:12:12,860 get a website where things can be hosted 206 00:12:12,860 --> 00:12:15,060 and you can see the emails coming in 207 00:12:15,060 --> 00:12:15,740 and discussions. 208 00:12:17,700 --> 00:12:22,840 GitHub is another GitForge which is also a 209 00:12:22,840 --> 00:12:27,640 platform and that's what people are most familiar 210 00:12:27,640 --> 00:12:29,080 with as a Forge. 211 00:12:29,980 --> 00:12:34,600 Other Forges now exist such as Codeburg and 212 00:12:34,600 --> 00:12:38,820 also Tangled on the Approto protocol if people 213 00:12:38,820 --> 00:12:41,220 are familiar with Approto and Newsguy. 214 00:12:43,440 --> 00:12:48,520 GitLab another Forge yeah so okay so I'm 215 00:12:48,520 --> 00:12:50,140 coming from the perspective of like peer-to 216 00:12:50,140 --> 00:12:52,100 -peer space and there's tons of different people 217 00:12:52,100 --> 00:12:53,900 coming from a lot of different spaces can 218 00:12:53,900 --> 00:12:55,660 we go back like a little bit to 219 00:12:55,660 --> 00:12:57,840 the basics I think anyone working with code 220 00:12:57,840 --> 00:13:02,940 has encountered like Git has encountered repositories and 221 00:13:02,940 --> 00:13:05,700 have heard about Forges but what actually is 222 00:13:05,700 --> 00:13:06,300 a Forge? 223 00:13:06,660 --> 00:13:09,780 Yeah yeah exactly so it's I guess it's 224 00:13:09,780 --> 00:13:14,820 a home for collaboration so Git gives you 225 00:13:14,820 --> 00:13:20,340 the commits and files so blobs, treats, commits 226 00:13:20,340 --> 00:13:25,220 that allows you to version control your code 227 00:13:25,220 --> 00:13:29,660 over time but as humans we have lots 228 00:13:29,660 --> 00:13:33,120 of other ways that we collaborate and that's 229 00:13:33,120 --> 00:13:36,940 where we see things like issues or task 230 00:13:36,940 --> 00:13:44,500 trackers or discussions on changes identifying as someone 231 00:13:44,500 --> 00:13:50,540 who collaborates on these Git repositories that's I 232 00:13:50,540 --> 00:13:55,120 think that's where Forges bring or like yeah 233 00:13:55,120 --> 00:13:57,280 yeah they paint the picture a little bit 234 00:13:57,280 --> 00:13:59,200 more on top of Git. 235 00:14:00,680 --> 00:14:03,460 And in comparison to like something like CRDTs 236 00:14:03,460 --> 00:14:07,720 like we have this space that's become really 237 00:14:07,720 --> 00:14:11,280 really popular lately with Local First and all 238 00:14:11,280 --> 00:14:13,700 of that and CRDTs are a common means 239 00:14:13,700 --> 00:14:17,480 of facilitating collaboration like what's the differentiation there 240 00:14:17,480 --> 00:14:19,160 between Forges and CRDTs? 241 00:14:20,480 --> 00:14:22,120 I'll let Lorenz take it. 242 00:14:23,340 --> 00:14:28,440 So I think that CRDTs are a technology 243 00:14:28,440 --> 00:14:33,040 and a definition of how a data structure 244 00:14:33,040 --> 00:14:35,520 should work and what features a data structure 245 00:14:35,520 --> 00:14:40,820 should have and you can build something like 246 00:14:40,820 --> 00:14:43,580 a Forge which is a little more abstract 247 00:14:43,580 --> 00:14:46,320 concept like a feature set of things that 248 00:14:46,320 --> 00:14:49,020 people want to do so you approach you 249 00:14:49,020 --> 00:14:50,820 can approach it from two sides you can 250 00:14:50,820 --> 00:14:52,320 say okay what does the user want to 251 00:14:52,320 --> 00:14:53,980 do okay they want to do code review 252 00:14:53,980 --> 00:14:56,400 they want to discuss the changes and then 253 00:14:56,400 --> 00:14:58,960 the CRDT is a bit on is a 254 00:14:58,960 --> 00:15:02,440 bit of technology that can enable this and 255 00:15:02,440 --> 00:15:05,900 then yes what Radical actually does is it 256 00:15:05,900 --> 00:15:09,840 implements a Forge based on top of CRDTs 257 00:15:09,840 --> 00:15:12,740 so many of the things in Radical that 258 00:15:12,740 --> 00:15:15,800 we talk about like patches and issues are 259 00:15:15,800 --> 00:15:19,500 implemented as CRDTs so the underlying structure still 260 00:15:19,500 --> 00:15:23,040 is a CRDT but it is not critical 261 00:15:23,040 --> 00:15:25,840 for the user to know that when they 262 00:15:25,840 --> 00:15:27,660 open a patch there will be an operation 263 00:15:27,660 --> 00:15:30,360 on the CRDT that's abstracted away. 264 00:15:31,840 --> 00:15:36,200 Yeah to analogize it would be like asking 265 00:15:36,200 --> 00:15:40,340 what's the difference between a page rank and 266 00:15:40,340 --> 00:15:43,800 a web search or like a search engine 267 00:15:43,800 --> 00:15:47,480 so page rank is just an algorithm and 268 00:15:47,480 --> 00:15:50,180 a search engine can like bring other features 269 00:15:50,180 --> 00:15:50,960 to the table. 270 00:15:52,280 --> 00:15:55,620 Yeah so we were on the topic of 271 00:15:55,620 --> 00:16:01,140 like why and so the why enters into 272 00:16:01,140 --> 00:16:03,600 the space now and then we kind of 273 00:16:03,600 --> 00:16:06,640 like moved into what actually is Radical. 274 00:16:08,740 --> 00:16:11,640 I have an idea of bringing us back 275 00:16:11,640 --> 00:16:13,340 a bit here to where we were before 276 00:16:13,340 --> 00:16:17,140 talking about the different Forges and other solutions 277 00:16:17,140 --> 00:16:20,440 in this space and we talked about a 278 00:16:20,440 --> 00:16:23,180 bunch of those and you can you can 279 00:16:23,180 --> 00:16:25,080 place them on a line and you can 280 00:16:25,080 --> 00:16:28,880 think about okay how centralized are they and 281 00:16:28,880 --> 00:16:32,580 I think we have like peak centralization with 282 00:16:32,580 --> 00:16:36,500 GitHub and then you can place the others 283 00:16:36,500 --> 00:16:39,500 right and then we have this concept of 284 00:16:39,500 --> 00:16:44,900 federation where as there are a few different 285 00:16:44,900 --> 00:16:47,260 ways of how to achieve federation but usually 286 00:16:47,260 --> 00:16:50,420 you have multiple servers and then the user 287 00:16:50,420 --> 00:16:52,560 logs into one of these servers and then 288 00:16:52,560 --> 00:16:55,820 there's data being federated between the servers so 289 00:16:55,820 --> 00:17:01,120 all of these solutions which includes Tangled on 290 00:17:01,120 --> 00:17:04,760 AT Proto and which includes the work towards 291 00:17:04,760 --> 00:17:08,260 federation in Forge Joe I would put them 292 00:17:08,260 --> 00:17:09,079 in the middle. 293 00:17:11,160 --> 00:17:13,800 Email also yes you're very right to point 294 00:17:13,800 --> 00:17:16,099 that out email is also of that kind 295 00:17:16,099 --> 00:17:19,420 you communicate your email server and then the 296 00:17:19,420 --> 00:17:24,040 emails are replicated and then we have another 297 00:17:24,040 --> 00:17:28,640 category where Radical falls into which is basically 298 00:17:28,640 --> 00:17:31,300 where there are no servers there are just 299 00:17:31,300 --> 00:17:33,660 nodes in the network and they sync in 300 00:17:33,660 --> 00:17:37,960 a peer-to-peer manner yeah and there 301 00:17:37,960 --> 00:17:41,520 aren't really any other competitors in a sense 302 00:17:41,520 --> 00:17:45,620 or any other projects that are currently ambitious 303 00:17:45,620 --> 00:17:48,780 and as active as we are to develop 304 00:17:48,780 --> 00:17:52,140 a really peer-to-peer Git layer. 305 00:17:53,860 --> 00:17:56,420 And I have so many questions in this 306 00:17:56,420 --> 00:17:59,160 regard I'm so curious and I've been hearing 307 00:17:59,160 --> 00:18:01,000 ever since I heard about Radical which was 308 00:18:01,000 --> 00:18:04,840 about two years ago now I've been wanting 309 00:18:04,840 --> 00:18:08,000 to figure out a little bit more to 310 00:18:08,000 --> 00:18:10,600 start from the perspective of why is it 311 00:18:10,600 --> 00:18:12,320 important to do it peer-to-peer in 312 00:18:12,320 --> 00:18:15,880 this setting like what makes the big difference 313 00:18:15,880 --> 00:18:18,160 here and then how do you enable those 314 00:18:18,160 --> 00:18:19,100 peer-to-peer connections? 315 00:18:19,500 --> 00:18:23,300 So I think it's important in general to 316 00:18:23,300 --> 00:18:27,380 have a technology that works in this local 317 00:18:27,380 --> 00:18:30,020 first and peer-to-peer manner because it 318 00:18:30,020 --> 00:18:34,200 puts all the power into the user or 319 00:18:34,200 --> 00:18:36,340 at least very very close to the user 320 00:18:36,340 --> 00:18:39,560 they control their own hardware there is no 321 00:18:39,560 --> 00:18:42,880 central authority that tells them when to log 322 00:18:42,880 --> 00:18:45,580 in or whether their account is blocked or 323 00:18:45,580 --> 00:18:49,420 not and it is a technology that enables 324 00:18:49,420 --> 00:18:54,120 this completely decentralized setting where you know I 325 00:18:54,120 --> 00:18:59,300 join Funkfeuer which is an association of people 326 00:18:59,300 --> 00:19:03,280 that put Wi-Fi antennas and on their 327 00:19:03,280 --> 00:19:06,320 roofs and then they might make direct peer 328 00:19:06,320 --> 00:19:10,240 -to-peer connections if you need internet access 329 00:19:10,240 --> 00:19:13,220 to let's say a data center in order 330 00:19:13,220 --> 00:19:16,800 to use your forge then well you're tied 331 00:19:16,800 --> 00:19:19,780 to this access but with Radical it will 332 00:19:19,780 --> 00:19:24,700 just work in such a completely partitioned part 333 00:19:24,700 --> 00:19:27,700 of the of the internet just as well 334 00:19:27,700 --> 00:19:31,620 for example if Fintan and me work in 335 00:19:31,620 --> 00:19:33,860 a room and are logged into the same 336 00:19:33,860 --> 00:19:38,280 Wi-Fi then we can just directly exchange 337 00:19:38,280 --> 00:19:40,740 information over Radical and we don't need to 338 00:19:40,740 --> 00:19:43,820 go out to a data center and so 339 00:19:43,820 --> 00:19:46,440 on so we are immediately much more resilient 340 00:19:46,440 --> 00:19:49,860 and much more sovereign than with these other 341 00:19:49,860 --> 00:19:50,480 technologies. 342 00:19:50,920 --> 00:19:54,500 It enables this routing agnostic perspective where you 343 00:19:54,500 --> 00:19:57,200 can route over different transport layers it sounds 344 00:19:57,200 --> 00:19:59,920 like could you use mesh, Wi-Fi, Bluetooth, 345 00:20:00,860 --> 00:20:02,500 sneaker net whatever kind of here? 346 00:20:02,740 --> 00:20:04,960 Yeah I have to admit that we are 347 00:20:04,960 --> 00:20:09,400 currently using TCP transport so we are a 348 00:20:09,400 --> 00:20:12,540 bit limited in that regard but we could 349 00:20:12,540 --> 00:20:16,240 extend so that's a limitation of the current 350 00:20:16,240 --> 00:20:19,560 implementation but not out of the protocol design 351 00:20:19,560 --> 00:20:22,720 so the data structures that arise from the 352 00:20:22,720 --> 00:20:26,340 sync process they work even if you you 353 00:20:26,340 --> 00:20:30,840 know transfer via sneaker net or whatever but 354 00:20:30,840 --> 00:20:34,020 the current implementation expects to talk IP and 355 00:20:34,020 --> 00:20:34,400 TCP. 356 00:20:35,700 --> 00:20:38,220 I had a previous conversation with IRO and 357 00:20:38,220 --> 00:20:40,180 we talked a lot about map traversal or 358 00:20:40,180 --> 00:20:43,560 hole punching to establish these peer-to-peer 359 00:20:43,560 --> 00:20:44,540 connections. 360 00:20:45,300 --> 00:20:49,980 How do you go about that? 361 00:20:49,980 --> 00:20:54,980 We know about IRO, we are actually in 362 00:20:54,980 --> 00:20:58,200 loose contact with them and we are really 363 00:20:58,200 --> 00:21:01,780 watching the awesome work they are doing and 364 00:21:01,780 --> 00:21:07,700 we are thinking of at least allowing connections 365 00:21:07,700 --> 00:21:12,020 via IRO to in a like a testing 366 00:21:12,020 --> 00:21:12,440 phase. 367 00:21:13,460 --> 00:21:17,620 It lines up really well on like the 368 00:21:17,620 --> 00:21:21,360 the cryptography of it so in radical nodes 369 00:21:21,360 --> 00:21:25,420 are identified by an edwards 25519 key and 370 00:21:25,420 --> 00:21:28,260 the same is true for IRO so we 371 00:21:28,900 --> 00:21:32,540 might get the direct correspondence between IRO endpoints 372 00:21:32,540 --> 00:21:36,840 and and a radical node ids so it 373 00:21:36,840 --> 00:21:38,780 all looks it all looks really nice and 374 00:21:38,780 --> 00:21:41,120 they would give us the hole punching capability 375 00:21:41,120 --> 00:21:43,020 which we don't have currently. 376 00:21:44,540 --> 00:21:50,420 But it's an ongoing process and we're definitely 377 00:21:50,420 --> 00:21:52,260 eager to experiment more. 378 00:21:52,940 --> 00:21:55,620 And this is actually we're getting into an 379 00:21:55,620 --> 00:21:57,680 area that has been a big part of 380 00:21:57,680 --> 00:22:01,120 my research in the past which is this 381 00:22:01,120 --> 00:22:03,720 what's starting to emerge is kind of an 382 00:22:03,720 --> 00:22:08,960 ecology of protocols from previous years like 10 383 00:22:08,960 --> 00:22:11,540 years ago a lot of projects were trying 384 00:22:11,540 --> 00:22:14,140 to build the entire solution on their own 385 00:22:14,140 --> 00:22:17,600 which meant rebuilding the same wheels over and 386 00:22:17,600 --> 00:22:21,700 over but now these kind of protocols are 387 00:22:21,700 --> 00:22:25,180 entering more of a modular form of design 388 00:22:25,180 --> 00:22:27,860 so it sounds like you're solving a quite 389 00:22:27,860 --> 00:22:32,460 a specific part of the puzzle piece of 390 00:22:32,460 --> 00:22:35,200 peer-to-peer networking and specifically for git. 391 00:22:36,520 --> 00:22:39,980 What is like where do you see yourself 392 00:22:39,980 --> 00:22:42,400 and what is your solution more concretely? 393 00:22:42,920 --> 00:22:46,480 Yeah that's a great question and I do 394 00:22:46,480 --> 00:22:49,760 hearken back to when I started a phrase 395 00:22:49,760 --> 00:22:52,240 that was thrown around was like protocols not 396 00:22:52,240 --> 00:22:52,840 platforms. 397 00:22:55,480 --> 00:23:00,000 So yeah I guess where we're trying to 398 00:23:00,000 --> 00:23:04,860 fit in is building the protocol on top 399 00:23:04,860 --> 00:23:11,820 of git for collaboration and as we spoke 400 00:23:11,820 --> 00:23:14,920 before we're talking about we're talking about it 401 00:23:14,920 --> 00:23:18,500 in a self-sovereign way we also fit 402 00:23:18,500 --> 00:23:24,720 into this local first even offline first scenario 403 00:23:24,720 --> 00:23:27,960 where you can hop on a plane still 404 00:23:27,960 --> 00:23:31,020 work on your collaboration not just your code 405 00:23:31,020 --> 00:23:35,100 so I can like reply to issues reply 406 00:23:35,100 --> 00:23:38,140 to patch comments and then come back online 407 00:23:38,140 --> 00:23:43,720 later and sync up with the network so 408 00:23:43,720 --> 00:23:48,060 yes putting that protocol first and hopefully making 409 00:23:48,060 --> 00:23:51,280 it repeatable so that people can come and 410 00:23:51,280 --> 00:23:55,560 boost the ecosystem with other implementations in the 411 00:23:55,560 --> 00:24:05,260 future I think so 412 00:24:05,260 --> 00:24:08,900 I'm also thinking like you mentioned in the 413 00:24:08,900 --> 00:24:10,800 beginning like a question that a lot of 414 00:24:10,800 --> 00:24:13,160 people have asked you is like okay so 415 00:24:13,160 --> 00:24:15,440 why are you doing this git is already 416 00:24:15,440 --> 00:24:19,660 decentralized like what do you say when people 417 00:24:19,660 --> 00:24:26,160 ask that like where do you because okay 418 00:24:26,160 --> 00:24:28,740 I think it boils down to decentralization be 419 00:24:28,740 --> 00:24:32,080 a very fuzzy term because there's so many 420 00:24:32,080 --> 00:24:35,380 forms of decentralization and federation and peer-to 421 00:24:35,380 --> 00:24:37,740 -peer are often dumped in as it clumps 422 00:24:37,740 --> 00:24:42,680 some under decentralization is it there in the 423 00:24:42,680 --> 00:24:49,100 distinction that by enabling what or by doing 424 00:24:49,100 --> 00:24:51,840 what you're doing with git you're kind of 425 00:24:51,840 --> 00:24:54,740 bringing it to a new level of decentralization 426 00:24:54,740 --> 00:24:59,340 my thoughts here are that we are we 427 00:24:59,340 --> 00:25:02,320 have realized and I think many people will 428 00:25:02,320 --> 00:25:07,020 agree that software engineering and managing a software 429 00:25:07,020 --> 00:25:12,500 project being open source or maybe even not 430 00:25:12,500 --> 00:25:17,960 is not just about git commits and and 431 00:25:17,960 --> 00:25:20,120 tags and so on there's a lot that 432 00:25:20,120 --> 00:25:24,480 works on top like code review and discussions 433 00:25:24,480 --> 00:25:27,100 about the code and tasks and so on 434 00:25:27,100 --> 00:25:32,840 and these are just fundamentally not part of 435 00:25:32,840 --> 00:25:35,360 git so here we compete in a sense 436 00:25:35,360 --> 00:25:38,620 with these other forges so git is decentralized 437 00:25:38,620 --> 00:25:42,620 in the aspect that you can you can 438 00:25:42,620 --> 00:25:45,180 put the repository on any computer and if 439 00:25:45,180 --> 00:25:48,180 you set those computers up well enough then 440 00:25:48,180 --> 00:25:51,000 you can have this network of git repositories 441 00:25:51,000 --> 00:25:55,140 and that is decentralized but you're decentralized only 442 00:25:55,140 --> 00:25:58,860 at the level of git features which aren't 443 00:25:58,860 --> 00:26:01,140 really enough to do all the things you 444 00:26:01,140 --> 00:26:04,180 want to do as alluded to before so 445 00:26:04,180 --> 00:26:07,660 this is one aspect you you need more 446 00:26:07,660 --> 00:26:10,140 than just git and you also want it 447 00:26:10,140 --> 00:26:12,840 to be decentralized that at least we do 448 00:26:12,840 --> 00:26:16,840 and then one other thing that comes to 449 00:26:16,840 --> 00:26:19,940 mind is that for using git in a 450 00:26:19,940 --> 00:26:24,360 decentralized way you have to make careful configuration 451 00:26:24,360 --> 00:26:27,220 like you have to share these urls of 452 00:26:27,220 --> 00:26:30,240 all the murals of your repository with everyone 453 00:26:30,240 --> 00:26:33,560 so if five people want to want to 454 00:26:33,560 --> 00:26:35,840 collaborate on a git repository in a truly 455 00:26:35,840 --> 00:26:38,940 decentralized way then each of these five persons 456 00:26:38,940 --> 00:26:42,820 has to note down the git url of 457 00:26:42,820 --> 00:26:45,520 all the four others and we make an 458 00:26:45,520 --> 00:26:50,820 improvement here because we have a globally consistent 459 00:26:50,820 --> 00:26:55,760 way to address git repositories and that's something 460 00:26:55,760 --> 00:26:57,920 that git itself doesn't have there is no 461 00:26:57,920 --> 00:27:01,860 standard way to say this repository which is 462 00:27:01,860 --> 00:27:06,360 agnostic over various locations of the repository that's 463 00:27:06,360 --> 00:27:08,180 something that we add and note that this 464 00:27:08,180 --> 00:27:10,620 is not centralized so there's not a global 465 00:27:10,620 --> 00:27:15,480 registry where you register the repository but this 466 00:27:15,480 --> 00:27:19,800 is done in a decentralized way are you 467 00:27:19,800 --> 00:27:24,380 using dhts like distributed hash tables no we're 468 00:27:24,380 --> 00:27:28,720 not using dhts currently we do node discovery 469 00:27:28,720 --> 00:27:33,560 via a gossip network so at first you 470 00:27:33,560 --> 00:27:36,260 join the network usually via what we call 471 00:27:36,260 --> 00:27:39,240 a process of bootstrapping so you first have 472 00:27:39,240 --> 00:27:42,120 to know some node in the network and 473 00:27:42,120 --> 00:27:45,480 then from then on the nodes will give 474 00:27:45,480 --> 00:27:47,600 out addresses of other nodes that they have 475 00:27:47,600 --> 00:27:49,780 found on the network and then very quickly 476 00:27:49,780 --> 00:27:51,840 you build an address book of other nodes 477 00:27:54,080 --> 00:27:57,560 awesome so i'm wondering now like you talked 478 00:27:57,560 --> 00:28:01,580 a bit about like the importance of enabling 479 00:28:01,580 --> 00:28:05,680 the collaboration aspects in connection to git and 480 00:28:05,680 --> 00:28:10,460 we also mentioned platforms and stuff like that 481 00:28:10,460 --> 00:28:16,240 where what is like dream scenario where do 482 00:28:16,240 --> 00:28:18,480 y'all want to go like what is 483 00:28:18,480 --> 00:28:21,420 the roadmap what is a future that you 484 00:28:21,420 --> 00:28:25,300 envision when radical is like a project that's 485 00:28:25,300 --> 00:28:27,580 complete fully because i guess it's also still 486 00:28:27,580 --> 00:28:32,380 in development yeah um what does that look 487 00:28:32,380 --> 00:28:34,520 like for you that's a it's an interesting 488 00:28:34,520 --> 00:28:38,960 question because there's a question of when is 489 00:28:38,960 --> 00:28:43,340 it a software project ever done and i 490 00:28:43,340 --> 00:28:45,660 actually it reminds me of a post by 491 00:28:45,660 --> 00:28:49,040 moxie um one of the founders of signal 492 00:28:51,440 --> 00:28:55,100 and in it they write about kind of 493 00:28:55,100 --> 00:28:58,660 like envy or jealousy of musicians where it's 494 00:28:58,660 --> 00:29:00,500 like you release a piece of music and 495 00:29:00,500 --> 00:29:02,140 then it's kind of just done and you're 496 00:29:02,140 --> 00:29:05,480 you're done i mean you can always like 497 00:29:05,480 --> 00:29:12,760 remaster edit remix whatever push software just like 498 00:29:12,760 --> 00:29:15,380 it feels infinite it can go on forever 499 00:29:15,380 --> 00:29:24,760 but yeah but then yeah is it mostly 500 00:29:24,760 --> 00:29:28,600 like scope creep or is it like is 501 00:29:28,600 --> 00:29:31,440 it because there's maintenance what is the main 502 00:29:31,440 --> 00:29:34,680 thing that makes it infinite do you think 503 00:29:34,680 --> 00:29:38,440 maintenance does make it feel infinite the most 504 00:29:38,440 --> 00:29:40,880 but of course i think what makes it 505 00:29:40,880 --> 00:29:43,140 feel infinite is just like the power of 506 00:29:43,140 --> 00:29:48,680 ideas and um yeah maybe scope creep is 507 00:29:48,680 --> 00:29:50,940 kind of like a way of describing that 508 00:29:50,940 --> 00:29:55,680 but i guess it's all about like constant 509 00:29:55,680 --> 00:29:58,480 improvements like git was