WEBVTT 00:00.000 --> 00:06.600 I will hold it, I guess. 00:06.600 --> 00:09.400 So I'm a product manager at denim, working on last feed 00:09.400 --> 00:13.520 docs, which is a note-taking, real-time note-taking, 00:13.520 --> 00:15.240 and documentation platform. 00:15.240 --> 00:19.200 Today we'll be talking about the role of libraries 00:19.200 --> 00:22.800 as infrastructure to build products. 00:22.800 --> 00:25.600 And we'll have three cores in this menu. 00:25.600 --> 00:28.520 First, we'll be a short last feed docs intro. 00:28.520 --> 00:31.480 Then a testimony by USF about blocknotes 00:31.480 --> 00:34.920 cooperation with us, denim, and then we'll 00:34.920 --> 00:40.320 propose a playbook for collaboration with first libraries. 00:40.320 --> 00:44.120 So going right into it, let's read docs. 00:44.120 --> 00:46.440 So what is it, it's this, it's the application 00:46.440 --> 00:49.960 you're seeing on the screen, and you might ask why 00:49.960 --> 00:51.560 and other not taking application. 00:51.560 --> 00:53.800 Well, we work in the government. 00:53.800 --> 00:57.080 There's a lot of world document out there. 00:57.080 --> 00:59.400 And I really like this screenshot in French 00:59.400 --> 01:02.640 from Reddit, who basically says, well, this 01:02.640 --> 01:08.360 is like a bad restaurant with a thousand mediocre options. 01:08.360 --> 01:10.440 I couldn't have phrased it better. 01:10.440 --> 01:13.400 So in docs, you have very few options. 01:13.400 --> 01:15.840 No, not much formatting. 01:15.840 --> 01:18.920 You come choose your phone, you come choose your spacing, 01:18.920 --> 01:22.080 you just write your contents, use some sort of cat, 01:22.080 --> 01:23.520 and you're done. 01:23.520 --> 01:26.680 So basically, this is what this docs, you write text, 01:26.680 --> 01:29.240 you format a little, you collaborate in real time. 01:29.240 --> 01:32.200 So you see the cursors of your friends. 01:32.200 --> 01:34.400 And what's nice is because it's web first. 01:34.400 --> 01:36.400 It's also suited for documentation. 01:36.400 --> 01:38.760 So you can put docs inside docs. 01:38.760 --> 01:43.400 You publish a link, and you have a documentation platform. 01:43.400 --> 01:45.200 Believe it or not, it was quite a success 01:45.200 --> 01:46.680 in the open source community. 01:46.680 --> 01:49.880 Someone published it on a hacker news. 01:49.880 --> 01:52.000 And it was a huge success. 01:52.000 --> 01:54.360 It wasn't the front page for two days. 01:54.360 --> 02:00.480 And now Docs is a 15K plus star project on GitHub. 02:00.480 --> 02:05.360 It's probably the most star repository in the French government. 02:05.360 --> 02:09.960 And you can see it was, this is Hacker News, boom. 02:09.960 --> 02:13.200 And I love because Hacker News is super drama, 02:13.200 --> 02:16.640 and you have like, Americans saying, like, it really feels 02:16.640 --> 02:18.680 like we are in a time where the EU countries 02:18.680 --> 02:21.480 are taking front and center on the world stage, 02:21.480 --> 02:24.240 with school, progressive, and unified stuff. 02:24.240 --> 02:29.200 And the US is in the ally, it's not in crack. 02:29.200 --> 02:29.840 Go Europe. 02:33.520 --> 02:39.200 So first and foremost, Docs is the proof that EU cooperation 02:39.200 --> 02:43.400 between member states can make great software. 02:43.400 --> 02:46.680 So government can code, and they should code, 02:46.680 --> 02:49.320 if they want to have real autonomy. 02:49.320 --> 02:53.320 And also, it's a great success story of direct cooperation 02:53.360 --> 02:58.080 with OSS libraries that make Doc possible. 02:58.080 --> 03:01.720 Docs at the government were not doing the hard stuff. 03:01.720 --> 03:04.560 We are just doing a jungle application, 03:04.560 --> 03:09.000 plugging OIGC doing some permissions on the docs, 03:09.000 --> 03:11.320 handling the subdocs and the UI. 03:11.320 --> 03:14.920 This guy is doing the hard stuff with a few other libraries. 03:14.920 --> 03:18.560 Basically, the hard stuff is what you're seeing here in the middle. 03:18.560 --> 03:20.560 Where people collaborate in real time, 03:20.560 --> 03:24.280 you have really metal mats going on down there, 03:24.280 --> 03:28.040 that are called CRDTs, config, conflict resolution, 03:28.040 --> 03:32.520 data types that allow people to work on different versions 03:32.520 --> 03:35.760 of a document and have the conflicts resolved in real time. 03:35.760 --> 03:38.280 This is pretty crazy that we have this in open source 03:38.280 --> 03:41.920 these days, 10 years ago, no way. 03:41.920 --> 03:43.840 And now we can just do this project. 03:43.840 --> 03:47.840 Like Docs has only one and a half years old. 03:47.840 --> 03:50.520 And it's already looking pretty sleek. 03:51.480 --> 03:54.160 So we also do a lot of stickers, 03:54.160 --> 03:57.480 because Docs is not only a French project. 03:57.480 --> 03:59.840 It's a German and Dutch project. 03:59.840 --> 04:02.360 That's why we have Baguette, Goudar, 04:02.360 --> 04:05.920 and the Bratzel behind the European computer. 04:05.920 --> 04:09.520 We believe that CRDT is so crucial to collaboration 04:09.520 --> 04:13.720 that we change the slogan of the French Republic 04:13.720 --> 04:17.720 to liberty, equality, serdity, of course. 04:17.720 --> 04:22.720 We welcome every word user as refugees inside Docs. 04:24.200 --> 04:26.800 We feel you, we feel for you of his users, 04:26.800 --> 04:29.160 you're living a nightmare, your document format, 04:29.160 --> 04:30.680 is a black box. 04:30.680 --> 04:33.000 Come to Docs, he will be much better. 04:34.280 --> 04:36.880 If you want, you can join me on this document 04:36.880 --> 04:38.320 by scanning this QR code. 04:39.960 --> 04:42.240 I'm gonna publish the document. 04:43.360 --> 04:45.480 It's already published. 04:46.280 --> 04:49.880 We should see people, oh sorry, the QR code. 04:49.880 --> 04:53.080 Yeah, if someone, I'm seeing some scans. 04:53.080 --> 04:55.840 Okay, and now people are messing with my document. 04:55.840 --> 04:57.600 This is the beauty of collaboration. 04:57.600 --> 05:04.320 One day you go on scratch pad, official scratch pad. 05:04.320 --> 05:08.600 We had a really bad experience of conspiracy theory 05:08.600 --> 05:09.520 and sexual content. 05:09.520 --> 05:12.520 So I'm not doing public documents anymore. 05:13.520 --> 05:16.560 But I have the perfect feature for that. 05:16.560 --> 05:20.760 We have a history version and I can just restore 05:20.760 --> 05:25.760 back my document to normal without the mess you just created 05:25.760 --> 05:29.320 and I will lock my documents because now you have the URL 05:29.320 --> 05:32.720 and you can actually make me give me a lot of trouble. 05:34.720 --> 05:39.120 And I will put it in private because why is this public as well? 05:39.120 --> 05:40.840 Okay, shit. 05:42.600 --> 05:47.520 Okay, okay, hang on, hang on, yes, here, the document. 05:47.520 --> 05:53.320 Oh no, all right, maybe the document above. 05:53.320 --> 05:55.520 Anyways, don't mess with medic. 05:55.520 --> 06:01.320 I don't remember, I don't remember where the master of the document is. 06:01.320 --> 06:07.320 Anyway, I will restore again and we are switching to UCF now. 06:07.320 --> 06:13.320 We've been collaborating for at least a year and a half now. 06:13.320 --> 06:16.320 We send this with denim and it has a presentation for you. 06:16.320 --> 06:19.320 Thank you. 06:19.320 --> 06:24.320 Not sure if I can top this. 06:24.320 --> 06:27.320 There you go. 06:27.320 --> 06:31.320 There's no people going to insert funny emojis in my slides. 06:31.320 --> 06:33.320 We need collaborative slides then as well. 06:33.320 --> 06:36.320 So maybe next year. 06:37.320 --> 06:41.320 Cool, so I'm a UCF of a long history in startups. 06:41.320 --> 06:44.320 A little bit shorter in open source. 06:44.320 --> 06:49.320 Actually, as part of this history, I built many products also on top of open platforms. 06:49.320 --> 06:53.320 Only to find out later after building cool stuff and after getting a lot of people. 06:53.320 --> 06:58.320 Using it, that those so-called open platforms run by US companies. 06:58.320 --> 07:02.320 Then suddenly shut down the platforms as soon as you come competitive. 07:02.320 --> 07:12.320 So I'm really excited here now to be working in the open source space to really show what's possible and that we can do things in a different way. 07:12.320 --> 07:14.320 So that brought us to our mission. 07:14.320 --> 07:23.320 Yeah, how can we bring the modern editing layer that Europe can rely on? 07:23.320 --> 07:25.320 Yeah, quick show guys. 07:25.320 --> 07:29.320 If what block now this, you're just in the editor, but it's kind of like the editing components. 07:29.320 --> 07:35.320 This is a demo with AI integration collaboration that you just played around with all the menus. 07:35.320 --> 07:42.320 So we try to make this simple and have all the things that you might need to build in. 07:42.320 --> 07:43.320 Why does this matter? 07:43.320 --> 07:45.320 A little bit to put it in context. 07:45.320 --> 07:50.320 Of course, everyone in the office, which you do all day, is collaborate on documents. 07:50.320 --> 07:52.320 And these kind of workflows. 07:52.320 --> 08:00.320 Traditionally, you do this in software dominated by Microsoft Google or VC back challenges, challenges like notion. 08:00.320 --> 08:08.320 And yeah, if we have good open source software that can power this, we can build our own things like dogs, like something you want to build yourself. 08:08.320 --> 08:13.320 And take back control, create independent alternatives. 08:13.320 --> 08:16.320 As I mentioned, we try to do this in a new way. 08:16.320 --> 08:24.320 So we held off outside investors and want to do this independently. 08:24.320 --> 08:29.320 And yeah, not as it started up chasing exit opportunities. 08:29.320 --> 08:33.320 Originally, I think they also deserve a shoutout. 08:33.320 --> 08:39.320 The initial R&D was based on another project and that was mainly found by NLNet. 08:39.320 --> 08:44.320 So without them, this wouldn't have been possible. 08:44.320 --> 08:56.320 Now of the last year or so, so I think 2025 for us was really a year where we grew something that kind of started as a side project to something that, yeah, now has a small team supporting it. 08:56.320 --> 09:00.320 So away back there. 09:00.320 --> 09:09.320 So this small team, like professionalizing the features, growing an organization around it, and really making it more sustainable. 09:09.320 --> 09:15.320 And also the stuff that we've been able to do has been funded by the collaboration with Tzendis and Dinem. 09:15.320 --> 09:20.320 So that's been really, really important for us in this phase. 09:20.320 --> 09:27.320 And then we also have business sponsors at C-Tratio have a little bit of diverse streams. 09:27.320 --> 09:31.320 To put this into context, because this talk is about funding libraries. 09:31.320 --> 09:34.320 And of course, with open source, you have open source applications. 09:34.320 --> 09:36.320 You have libraries, maybe have foundations. 09:36.320 --> 09:39.320 So you just saw Docs, which really sits at the application layer. 09:39.320 --> 09:41.320 So that's the thing that the end user uses. 09:41.320 --> 09:46.320 Docs is being built with block now that makes the integration for them easier. 09:46.320 --> 09:53.320 But actually, we also rely on some very important, very successful open source foundations. 09:53.320 --> 09:59.320 So we directly also collaborate with YGS, with Kevin, who, unfortunately, is unfortunately for him. 09:59.320 --> 10:03.320 He happens to be skiing and couldn't make it here. 10:03.320 --> 10:07.320 But also pros, mayor, for some of the foundational stuff. 10:07.320 --> 10:16.320 So this is really a collaboration of different open source things together for an ecosystem. 10:16.320 --> 10:24.320 What I thought might be interesting for this room is to, of course, we talk about the theory of the benefits of open source. 10:24.320 --> 10:31.320 To, with our use case, make this a little bit more tangible and really take that from theory to practice. 10:31.320 --> 10:40.320 I'll walk you through, I think, some of the advantages that we already see or the impacts that we have across these categories. 10:40.320 --> 10:42.320 So the first one is obvious. 10:42.320 --> 10:46.320 So the tools, yeah, your mission, Adina and other mission. 10:46.320 --> 10:50.320 And then this is to enable digital autonomy. 10:50.320 --> 10:53.320 So that was what we're doing with last suite. 10:53.320 --> 10:58.320 But then also we see other companies now who are building part of open desk, for example. 10:58.320 --> 11:04.320 So other open source companies like Open Project and XWC now because we're growing more mature, 11:04.320 --> 11:10.320 also adopting block nodes and YGS in their system. 11:10.320 --> 11:18.320 So, yeah, it becomes an even stronger ecosystem. 11:18.320 --> 11:21.320 I think if you talk about talent. 11:21.320 --> 11:28.320 So this, yeah, I don't think you can make, we know about the boss factor. 11:28.320 --> 11:33.320 You cannot make open source sustainable without just relying on one person. 11:33.320 --> 11:38.320 So this funding has been crucial for us to grow a small team. 11:38.320 --> 11:41.320 Without the need for outside funding. 11:41.320 --> 11:43.320 So for us, this has been really important. 11:43.320 --> 11:51.320 And I think it's key also if you want to encourage technical talent to keep investing in open source. 11:51.320 --> 11:56.320 And actually it can be a key benefit because if you look at this, this conference, 11:56.320 --> 11:59.320 I think it's the only conference that I know that happens on the weekend. 11:59.320 --> 12:01.320 It's because people are really passionate about this. 12:01.320 --> 12:06.320 And it's really exciting for engineers for talented people to be working on open source. 12:06.320 --> 12:12.320 Because you can impact so many other, you see other people building on top of what you build. 12:12.320 --> 12:15.320 So it's also like a really key selling point. 12:15.320 --> 12:23.320 I think there's many people telling the people locked inside of companies who would rather work on exciting open source projects. 12:23.320 --> 12:28.320 But I currently not able to do so because a lack of funding, etc. 12:28.320 --> 12:33.320 So I think that's a really, really key benefit. 12:34.320 --> 12:37.320 Then more of the boring stuff, economy. 12:37.320 --> 12:44.320 If we look at YGS, so there's this technology that's developed by Kevin and spirit famous, 12:44.320 --> 12:48.320 like tomorrow in the local first bedroom, we'll talk a lot about it. 12:48.320 --> 12:52.320 But there's like big companies building on this thing. 12:52.320 --> 12:55.320 But it's also a very foundational company. 12:55.320 --> 13:00.320 So if we look again at the layers that technology is built, 13:01.320 --> 13:04.320 you have YGS, but then there's also companies around it. 13:04.320 --> 13:07.320 But even these don't provide end user products. 13:07.320 --> 13:13.320 Now these in their turn, they provide like infrastructure or in our case and editor. 13:13.320 --> 13:19.320 That then enable like a gazillion of other companies that we can't even keep track of, right? 13:19.320 --> 13:20.320 So that's the benefit. 13:20.320 --> 13:23.320 If we talk about really investing in foundational stuff, 13:23.320 --> 13:28.320 yeah, you get like an exponential amount of people, 13:28.320 --> 13:32.320 developers, companies being able to build on this. 13:32.320 --> 13:36.320 So yeah, obvious power of open source, but still cool to see. 13:36.320 --> 13:40.320 Then for us, this is some of our clients using it. 13:40.320 --> 13:43.320 And I think that's also fun or interesting to highlight. 13:43.320 --> 13:48.320 We're talking about the global tech markets, the power dynamics there. 13:49.320 --> 13:53.320 We actually see, we launched, for example, AI integration. 13:53.320 --> 13:57.320 And we see lots of startups, including startups from Silicon Valley, 13:57.320 --> 14:02.320 actually now using our software to build their products instead of the other way around. 14:02.320 --> 14:07.320 So I think that's also a really positive impact. 14:07.320 --> 14:12.320 Yeah, this kind of sum of that. 14:12.320 --> 14:16.320 Then, of course, not all farm end games, there's some challenges. 14:17.320 --> 14:20.320 I think we're all figuring this out together, right? 14:20.320 --> 14:23.320 As a government, I think there's been a lot of talk in this room. 14:23.320 --> 14:26.320 Okay, how do you, including my Julian just now? 14:26.320 --> 14:29.320 Okay, how do you as a government invest in the open source? 14:29.320 --> 14:34.320 There's no, from both sides, there's not really a playbook for that yet. 14:34.320 --> 14:36.320 But it's important to figure this out. 14:36.320 --> 14:38.320 This is like the mandatory XKCD. 14:38.320 --> 14:41.320 I think everybody has seen 10 times, but still got to have maybe, 14:41.320 --> 14:44.320 maybe a hundred times, but still put it in here. 14:45.320 --> 14:47.320 Because that is the way it is. 14:47.320 --> 14:50.320 And only, I think this is key message. 14:50.320 --> 14:53.320 Like only sustainable, stable, revenue streams, 14:53.320 --> 14:55.320 allow us to really grow teams. 14:55.320 --> 15:00.320 And you need to be able to have teams that doesn't mean, like, 15:00.320 --> 15:04.320 20 people, but you need to have some people to have that share responsibility. 15:04.320 --> 15:09.320 Reduce that boss factor and also make sure that libraries can, 15:09.320 --> 15:12.320 can stay maintained, so not just focus on building features, 15:12.320 --> 15:16.320 but also keep them maintained in the long term. 15:16.320 --> 15:19.320 So for us, like, I think for an open source project, 15:19.320 --> 15:22.320 we're super, super fortunate to have this kind of collaboration. 15:22.320 --> 15:25.320 That is really enabled a lot of for us this year. 15:25.320 --> 15:28.320 But sometimes the funding is still unstable, 15:28.320 --> 15:33.320 and this does make it difficult to make that next step. 15:33.320 --> 15:37.320 I'm lucky, like, we have a little bit, sometimes I can, 15:37.320 --> 15:42.320 I can bridge these financial gaps, but not every contributor or 15:42.320 --> 15:45.320 maintainer is, yeah, has that possibility. 15:45.320 --> 15:52.320 So there's a lot of opportunity being left on the table here. 15:52.320 --> 15:54.320 Still challenging also to find different business models. 15:54.320 --> 15:59.320 Like, I wouldn't want to rely purely on government collaborations. 15:59.320 --> 16:03.320 You want diverse, right, revenue streams. 16:03.320 --> 16:05.320 And of course, in building open source, 16:05.320 --> 16:07.320 you're still giving a lot away for free. 16:07.320 --> 16:10.320 And it's still difficult to actually do that. 16:10.320 --> 16:13.320 I'm pretty confident we can, but it's still like, okay, 16:13.320 --> 16:18.320 we're still exploring this. 16:18.320 --> 16:20.320 And also, like, okay, this costs like now, 16:20.320 --> 16:22.320 we have this public private sector partnership. 16:22.320 --> 16:23.320 What does that mean? 16:23.320 --> 16:25.320 What do we expect from each other? 16:25.320 --> 16:27.320 So we're figuring out this playbook as we go, 16:27.320 --> 16:32.320 and I think Fugile wants to share a little bit about that. 16:32.320 --> 16:34.320 Yep. 16:34.320 --> 16:40.320 So we thought the main problem that we identify 16:40.320 --> 16:44.320 is that currently, like, 16:44.320 --> 16:49.320 if you think of contracting IT contracting in general as a cake, 16:49.320 --> 16:51.320 maintainers, they get the crops. 16:51.320 --> 16:55.320 Usually it's the end product that gets the money to do the product, 16:55.320 --> 16:58.320 sometimes financing what it needs, 16:58.320 --> 17:01.320 but you don't even, as a maintainer, 17:01.320 --> 17:03.320 you don't even get the piece of the cake. 17:03.320 --> 17:07.320 Well, you do the work that is supporting all the people 17:07.320 --> 17:09.320 that are above you. 17:09.320 --> 17:15.320 So we are actually lucky enough to be able to work directly 17:15.320 --> 17:17.320 with maintainers. 17:17.320 --> 17:21.320 This is actually new, like, we are directly funding the libraries. 17:21.320 --> 17:24.320 We are discussing with them, like, okay, what's your roadmap? 17:24.320 --> 17:26.320 What's my roadmap? 17:27.320 --> 17:30.320 If I finance this, like, is it worth it? 17:30.320 --> 17:31.320 Is it good for you? 17:31.320 --> 17:32.320 Is it good for me? 17:32.320 --> 17:34.320 And you are able to get into that discussion? 17:34.320 --> 17:36.320 And we are able to probably put, like, 17:36.320 --> 17:39.320 flags on releases of libraries and say, 17:39.320 --> 17:43.320 these friends finance comments. 17:43.320 --> 17:46.320 Friends and Germany finance comments in blocknotes. 17:46.320 --> 17:50.320 Friends and Germany finance attribution of edit in white days. 17:50.320 --> 17:54.320 And white days is downloaded, like, millions of times every week. 17:54.320 --> 17:59.320 So this is, I think, the piece that we need to figure out is, like, 17:59.320 --> 18:03.320 these guys are actually companies, like, blocknotes of company. 18:03.320 --> 18:04.320 White days is a company. 18:04.320 --> 18:08.320 White are the not part of the contracting ecosystem. 18:08.320 --> 18:12.320 White are not directly contracting these these guys, 18:12.320 --> 18:17.320 because, I mean, like, this is the best ROI even if I speak with the finance dude. 18:17.320 --> 18:22.320 I tell him, like, look at the ROI I have investing in libraries, 18:23.320 --> 18:26.320 versus investing, like, in end user products. 18:26.320 --> 18:30.320 Because libraries are actually used by all the end products. 18:30.320 --> 18:34.320 So with yourself, we actually, like, wrote this document. 18:34.320 --> 18:36.320 There are some comments also by Kevin. 18:36.320 --> 18:40.320 You can scan this. This is a GitHub page where you can put issues. 18:40.320 --> 18:42.320 You can put, pull requests. 18:42.320 --> 18:45.320 But essentially, the playbooks, the playbook looks like this. 18:45.320 --> 18:48.320 I just have one minute to go over a little bit more, five minutes. 18:48.320 --> 18:53.320 So we sat down with, with yourself and Kevin. 18:53.320 --> 18:57.320 And we say, okay, what is a good collaboration with maintainers? 18:57.320 --> 19:03.320 Okay, so maybe, like, don't ask for exotic stuff that only you will use. 19:03.320 --> 19:05.320 So first, I line on roadmap. 19:05.320 --> 19:09.320 Let's find out the stuff that already is there that I also need. 19:09.320 --> 19:14.320 A second one would be like, okay, it's cool that I invest with your libraries, 19:14.320 --> 19:16.320 but libraries are ecosystems. 19:17.320 --> 19:19.320 So they need to talk to each other. 19:19.320 --> 19:23.320 So it's good to also find a standardization. 19:23.320 --> 19:29.320 So maybe in the future, we might talk with yourself, like, hey, we are storing your weird formats. 19:29.320 --> 19:38.320 How about, you make a format that talks to other documentaries. 19:38.320 --> 19:43.320 Also, what we see is that people are very keen in investing features. 19:43.320 --> 19:49.320 That's normal, like, as I said, you get your brand into the release note that soft power. 19:49.320 --> 19:53.320 But once the future is out, the work just starts. 19:53.320 --> 19:55.320 You know, like, the future is out there. 19:55.320 --> 19:58.320 Then there are 10 versions where you need to fix a bunch of stuff. 19:58.320 --> 20:01.320 Then there is compatibility with the rest of the code. 20:01.320 --> 20:04.320 And the maintainer is basically taking on that all that costs on him. 20:04.320 --> 20:07.320 And it's a small team always. 20:07.320 --> 20:11.320 So like, I don't know, like, can we come up with some kind of rule where we say, like, 20:11.320 --> 20:19.320 if I invest one, the 10-year-old in a future can I invest, like, three-year-old in the maintenance? 20:19.320 --> 20:23.320 So I have some kind of balance because it's hard to put the price on maintenance, 20:23.320 --> 20:28.320 but, like, a percentage on my budget feature would seem fair in our opinion. 20:28.320 --> 20:34.320 Then, like, the problem is, like, how sustainable is defending? 20:34.320 --> 20:37.320 The worst thing that can happen to a project this day, 20:37.320 --> 20:41.320 I get 10-year-old, I do 50% of the project. 20:41.320 --> 20:45.320 And then, I don't get the funding for the rest of the project. 20:45.320 --> 20:47.320 My project is still, like, for six months. 20:47.320 --> 20:51.320 I take that code again, I can put to trash because it's already outdated, 20:51.320 --> 20:53.320 and I have to redo it again. 20:53.320 --> 20:58.320 So, last time for the maintainer, lost money for the contractor, nobody wins. 20:58.320 --> 21:01.320 Sustain funding, and actually, like, 21:01.320 --> 21:06.320 defending the whole package of the feature you want to do, is crucial, 21:06.320 --> 21:10.320 because otherwise, we are just wasting everybody's time and money. 21:10.320 --> 21:16.320 Also, it's not because your contractor, private or public, 21:16.320 --> 21:18.320 that you need to hide behind doors. 21:18.320 --> 21:20.320 Like, if you are contracting with maintenance, 21:20.320 --> 21:24.320 it's actually a good thing that you come inside the GitHub field, 21:24.320 --> 21:27.320 or the GitLab field, and you put your issues, 21:27.320 --> 21:29.320 and you put your plans out there, 21:29.320 --> 21:32.320 and you say, actually, what is your plan with this library? 21:32.320 --> 21:34.320 A citizen of open source, don't just say, 21:34.320 --> 21:37.320 I need this by tomorrow, actually, like, 21:37.320 --> 21:40.320 go in the issue, follow the issue, follow the updates, 21:40.320 --> 21:43.320 look at the releases, like, don't just sit there thinking, 21:43.320 --> 21:46.320 these guys are going to be done tomorrow, 21:46.320 --> 21:48.320 because it doesn't work this way. 21:48.320 --> 21:53.320 I have 10 dependencies, and maybe I'm relying on this guy to finish this, 21:53.320 --> 21:55.320 and this other guy to finish this, though. 21:55.320 --> 21:58.320 Not looking at what actually goes down in the libraries, 21:58.320 --> 22:01.320 and investing into, it's not productive. 22:01.320 --> 22:04.320 So you should be active, and if you are financing, 22:04.320 --> 22:07.320 you should have people involved in following this project, 22:07.320 --> 22:10.320 and that's how you will build it autonomy, 22:10.320 --> 22:13.320 and also, self-power, because if you are inside there, 22:13.320 --> 22:17.320 you can help make decisions for the future of the project. 22:17.320 --> 22:19.320 And then, well, we already talked about this, 22:19.320 --> 22:22.320 but every library has dependencies. 22:22.320 --> 22:27.320 Open source is people standing on everyone's shoulders. 22:27.320 --> 22:31.320 So you should definitely, like, we are working with blocknot. 22:31.320 --> 22:35.320 JS with YJS, but we are not working yet with prosmear, 22:35.320 --> 22:37.320 which is, like, a crucial dependency we have. 22:37.320 --> 22:40.320 So we should look at, like, financing this person, 22:40.320 --> 22:43.320 because, again, it's a solo maintainer. 22:43.320 --> 22:45.320 YJS, surprise solo maintainer. 22:45.320 --> 22:50.320 Only yourself has been able to grow a small team of three, four, four. 22:50.320 --> 22:56.320 So we should, like, these guys are being downloaded millions of times, 22:56.320 --> 22:59.320 every week, and it's a one-person job. 22:59.320 --> 23:03.320 So we should think, like, okay, we can contract with these people. 23:03.320 --> 23:08.320 We can give them a piece of the cake, and it's not that hard, 23:08.320 --> 23:11.320 just like, let's look at the libraries we depend on, 23:11.320 --> 23:17.320 because, as good as the libraries are, as good our product will be. 23:17.320 --> 23:21.320 So, comment, scan that code, 23:21.320 --> 23:24.320 then the effort on GitHub, and let's try to decide on a playbook, 23:24.320 --> 23:27.320 so that we can actually directly work with libraries. 23:27.320 --> 23:31.320 It will, I think, advance a lot, open source for everyone. 23:31.320 --> 23:33.320 Do you want to add, uh... 23:33.320 --> 23:35.320 No, thanks. Okay. 23:35.320 --> 23:36.320 Thank you.