WEBVTT 00:00.000 --> 00:10.360 Okay, ladies and gentlemen, welcome to the first session on Sunday. 00:10.360 --> 00:18.520 I think the title is self-explanatory, but Margaret is going to take us into detail on who 00:18.520 --> 00:19.520 really pays. 00:19.520 --> 00:20.520 Thank you. 00:20.520 --> 00:21.520 Thank you. 00:21.520 --> 00:31.680 All right, thank you, and good morning, everybody. 00:31.680 --> 00:35.840 Thank you for waking up early for being here. 00:35.840 --> 00:40.480 I think we are actually early, at least according to my watch, we are early. 00:40.480 --> 00:46.120 So I expect more people will arrive that they are just cutting it very close. 00:46.120 --> 00:52.640 Or anybody that is watching on the livestream or later on on recording, it's okay 00:52.640 --> 00:56.120 that you didn't wake up early, it's very early, I know. 00:56.120 --> 01:00.200 But yeah, thank you for being here, those that are here. 01:00.200 --> 01:07.920 As it's very early, and I know many of you have not really woken up, let's start with some... 01:07.920 --> 01:11.160 This is not working. 01:11.160 --> 01:13.160 To turn it on. 01:14.160 --> 01:16.840 Yeah, yeah, now it's working. 01:16.840 --> 01:17.840 Yes. 01:17.840 --> 01:20.640 So let's start with a show of funds. 01:20.640 --> 01:28.360 So who here has ever contributed to open source bead with like a bug report, a batch, or a 01:28.360 --> 01:31.800 feature, or maintaining a whole project? 01:31.800 --> 01:32.800 Nice. 01:32.800 --> 01:36.800 Yes, we are at first and we expect open source contributors here. 01:36.800 --> 01:44.240 Now, who here is a core developer, maintainer, cometer, like you are really in charge 01:44.240 --> 01:47.120 of an open source project? 01:47.120 --> 01:49.840 Okay, yes, quite a few. 01:49.840 --> 01:56.960 Now of those that just raise your hands, who of you does this on your spare time, your 01:56.960 --> 02:01.480 nights and weekends, your time that is not paid by an employer? 02:01.480 --> 02:04.360 Okay, quite a few again. 02:04.360 --> 02:11.360 And if you just raise your hand, would you like to be paid for that work? 02:11.360 --> 02:17.040 Okay, yeah, most of you, at some note, don't, and that's fine. 02:17.040 --> 02:21.600 It's fine to want to be paid, and it's also fine to not want to be paid, right? 02:21.600 --> 02:23.800 Like both, both things are fine. 02:23.800 --> 02:29.200 All right, so before we jump into the talk a little bit about me, I'm Marga. 02:29.200 --> 02:34.320 I'm originally from Argentina, but I've been living in Germany for almost 14 years now. 02:35.160 --> 02:39.560 As a sign note, one of the reasons that I moved to Germany was so that I would be able 02:39.560 --> 02:46.440 to attend for them, because coming from Argentina for just two days seemed like way too 02:46.440 --> 02:49.240 much trouble at the time. 02:49.240 --> 02:54.080 And I did attend for them quite a few times since then, but this is actually my first 02:54.080 --> 03:00.360 talk at first and so, yeah, you know, it took me 14 years to make it on stage. 03:00.360 --> 03:05.360 So, thank you. 03:05.360 --> 03:10.520 Anyway, I've been using free and open source software for 25 years. 03:10.520 --> 03:17.240 I've been contributing for almost all of those 25 years, and I've been a deviant developer 03:17.240 --> 03:22.400 for over 20 years, so that's also very much part of my identity. 03:22.400 --> 03:30.400 Professionally, I started work at a company in Argentina, where I was a system administrator 03:30.400 --> 03:33.600 software engineer doing a little bit of everything. 03:33.600 --> 03:36.920 And then, as I said, I moved to Germany to work for Google. 03:36.920 --> 03:41.680 I worked for Google in the team that maintains the internal line of distribution that 03:41.680 --> 03:44.280 is used by Google engineers. 03:44.280 --> 03:50.600 And I worked there for eight years when I finally got tired of the corporate culture, I left. 03:50.600 --> 03:56.520 I showed in a small open source consultancy called Kingfuck, where I worked on flat-care 03:56.520 --> 03:58.520 container Linux. 03:58.520 --> 04:03.640 It was really nice because it was a company that was all about open source, but unfortunately 04:03.640 --> 04:09.680 it got acquired by Microsoft, so that was like the darkest year in my life, where I 04:09.680 --> 04:15.840 worked for Microsoft for a year, but I didn't like it, it really was not for me. 04:15.840 --> 04:22.840 So I eventually left Microsoft and joined ISOValent, which was an open core company in 04:22.840 --> 04:29.600 the Kubernetes space, doing products like Silium and Tetragon, and I worked there until 04:29.600 --> 04:32.200 they got acquired by Cisco. 04:32.200 --> 04:36.080 And I was like, I'm not joining another corporate again. 04:36.080 --> 04:43.560 So fortunately, I landed Egalia, which is also an open source consultancy, but in this case, 04:43.560 --> 04:50.360 it's workers on open source consultancy, so yeah, no risk of being acquired. 04:50.360 --> 04:55.960 All right, so why did I just go over my resume for you? 04:55.960 --> 05:02.320 It's mostly just to give you an idea of the different things that I have experience. 05:02.320 --> 05:08.440 I worked for big companies, small companies, I worked for product companies, for consultancies. 05:08.440 --> 05:13.320 So I kind of seen a bit of everything, of course I haven't seen absolutely everything 05:13.320 --> 05:18.800 that is out there, but I know how it is to work in open source, in different spaces. 05:18.800 --> 05:24.360 So this is kind of like the basis that I'm using for the rest of the talk. 05:24.360 --> 05:29.600 All right, now that's it about me, let's now talk about the problem. 05:29.600 --> 05:34.040 Why did I think this was an interesting topic to talk about? 05:34.040 --> 05:40.120 The problem or not problem starts 40 plus years ago. 05:40.120 --> 05:46.760 So when Richard Stanna created the GNU project, he was creating a project for a small group 05:46.760 --> 05:48.120 of people. 05:48.120 --> 05:54.960 There were about a thousand nodes connected to the internet, the people that were participating 05:54.960 --> 06:02.400 in the movement where sending software via tapes or disks. 06:02.400 --> 06:07.840 They were not downloading it over the internet, and everybody that was participating 06:07.840 --> 06:12.280 was very much an enthusiast. 06:12.280 --> 06:19.600 By 1991, when the GNU scanner was published, there were already around a million nodes connected 06:19.600 --> 06:20.600 to the internet. 06:20.600 --> 06:29.920 So a lot more people were able to participate, but these people were still using use net forums 06:30.000 --> 06:33.800 and downloading software via FTP. 06:33.800 --> 06:37.480 So it was still very much hacker culture, right? 06:37.480 --> 06:42.640 There were geeks, there were people that either knew what they were doing or at least they 06:42.640 --> 06:47.360 were willing to learn to be able to know what they were doing. 06:47.360 --> 06:54.080 And the expectation was that if you were using free software, you were also contributing 06:54.080 --> 06:55.080 to free software. 06:55.080 --> 06:56.080 It was a given, right? 06:56.080 --> 07:00.440 Like, a user and a contributor were the same thing. 07:00.440 --> 07:05.560 Now when we reach current times, this has gone up a lot. 07:05.560 --> 07:11.920 And actually this graph is from Wikipedia, and it's weird that it cups at like 1 billion 07:11.920 --> 07:12.920 nodes. 07:12.920 --> 07:14.560 I've been pondering why it cups there. 07:14.560 --> 07:19.400 I guess it's like most of the nodes are no longer connected to the public internet and 07:19.400 --> 07:25.280 are actually behind proxies and so on and so that's why this doesn't keep going up 07:25.280 --> 07:30.040 because we know that there's more and more devices connected to the internet today. 07:30.040 --> 07:31.240 But that's not relevant. 07:31.240 --> 07:37.960 The relevant part is that open source, unfree software has reached basically everybody, right? 07:37.960 --> 07:40.080 It has expanded together with the internet. 07:40.080 --> 07:47.040 And today you can download thousands and thousands of open source projects from the internet 07:47.120 --> 07:53.600 without ever having to interact with the containers, with the developers of the software, 07:53.600 --> 07:54.600 right? 07:54.600 --> 08:05.040 And there was actually a code audit recently that found that 97% of the software that was 08:05.040 --> 08:10.400 audited has at least one open source dependency. 08:10.400 --> 08:13.120 Basically open source is everywhere, right? 08:13.200 --> 08:17.480 We could say that open source has one, right? 08:17.480 --> 08:18.480 Woo! 08:18.480 --> 08:26.840 Yes, we have one because we are everywhere. 08:26.840 --> 08:36.040 But has it really one when like all modern digital infrastructure depends on the projects 08:36.080 --> 08:41.920 and random person in the vraska is thankfully maintaining in 2003? 08:41.920 --> 08:44.480 Probably not. 08:44.480 --> 08:50.320 And yeah, so no talk about open source sustainability is complete without this awesome comic 08:50.320 --> 08:52.320 from its KCD. 08:52.320 --> 08:59.320 But I do want to know that this comic only shows like one problematic piece and in reality 08:59.320 --> 09:02.800 we have a lot more, right? 09:02.800 --> 09:08.080 This is actually better than our reality. 09:08.080 --> 09:14.000 So open source has one in the objective of being everywhere. 09:14.000 --> 09:25.280 But we have not one in making this a sustainable model that can survive for the future. 09:25.280 --> 09:30.880 This is the part that we need to rethink. 09:31.880 --> 09:34.080 So I've been thinking about this for a while. 09:34.080 --> 09:37.320 Why is it that we are in this conundrum? 09:37.320 --> 09:44.720 And one of the things that I've been thinking about is this illusion of free on the internet 09:44.720 --> 09:48.360 today we get a lot of things for free. 09:48.360 --> 09:55.440 We get a web-made client for free, we get a search engine for free, we get social media for free. 09:55.440 --> 09:57.440 And why is it between quotes? 09:57.480 --> 10:01.480 It's been three quotes because we all know that this is not really for free. 10:01.480 --> 10:13.680 We pay for this by having to see ads and by having our data scraped and sold and bad things 10:13.680 --> 10:16.800 happening, right, without data. 10:16.800 --> 10:22.040 So it feels like it's for free, although we all know that it's not for free. 10:22.360 --> 10:29.240 Because we know it's not for free, we actually demand quality out of this product, right? 10:29.240 --> 10:36.840 That we get for free because we know that companies like Google and Meta are making money out of this product. 10:36.840 --> 10:40.840 And if they are making money, they should publish good products, right? 10:40.840 --> 10:43.560 This is how our minds operate. 10:43.560 --> 10:51.960 And this that makes sense for Google and Meta and other big companies does not make sense. 10:52.360 --> 10:57.560 For a maintainer that is doing this during their nights and weekends, right? 10:57.560 --> 11:03.000 And so this what we get for free is not always the same. 11:03.000 --> 11:09.160 But it's not easy to differentiate between one and the other. 11:09.160 --> 11:14.200 And you all that are here today probably can't tell the difference. 11:14.200 --> 11:18.760 But people that are not involved in the freeze-over movement, 11:18.760 --> 11:21.880 it's a lot harder for them to tell the difference. 11:21.880 --> 11:30.280 And so what happens is that people get in title and they feel like they have the right 11:30.280 --> 11:39.080 to demand quality out of these free things and they are not all the same. 11:39.080 --> 11:50.360 So then we come to the problem of users becoming consumers 11:50.360 --> 11:55.080 that demand quality, the expect service level agreements, 11:55.080 --> 12:00.840 accountability, security fixes, without giving anything back, right? 12:00.840 --> 12:08.760 And I'm not talking about any of you today here that woke up very early to be a twosom on a Sunday. 12:08.760 --> 12:17.480 I'm talking about the companies that are getting rich out of the immense value that has been created 12:17.480 --> 12:21.320 by the open source movement and don't give anything back, right? 12:21.320 --> 12:27.720 Companies that just make money, but don't give back with contributions, 12:27.720 --> 12:34.040 don't give back with donations, don't give back with employing free software containers. 12:35.240 --> 12:45.480 So this all taking and not giving back is creating the problem that I'm talking about today. 12:48.040 --> 12:53.400 And the problem that it's creating is the burnout of containers. 12:54.280 --> 12:57.720 So there's this quote that I found on an article. 12:58.920 --> 13:03.080 I found this article after I have been thinking about this for a long time and I found this article 13:03.080 --> 13:09.080 that talks about the same things. So if you're interested, I would really recommend reading it. 13:10.280 --> 13:15.160 I'll read this quote aloud. The risk of burnout increases when there is an imbalance 13:15.240 --> 13:20.840 between the energy or work demands of us and the resources are available to replenish it. 13:21.480 --> 13:28.760 So burnout is not just having a lot of work to do because if you are motivated and if you feel 13:28.760 --> 13:33.800 like you are replenishing your energy, you can do a lot of work and it's fine. 13:34.520 --> 13:44.280 The problem is when that work that you're doing does not then translate into satisfaction, right? 13:44.280 --> 13:51.560 If you're satisfied with what you're doing, if you have the time and the space to replenish your 13:51.560 --> 13:58.280 energy, it's fine. But when you don't, that's when burnout occurs. So this article, 13:58.280 --> 14:02.440 it was written by a psychologist or sociologist, sorry, something like that. 14:03.480 --> 14:09.320 Identified two major contributors to burnout in open source developers. 14:10.200 --> 14:16.440 The first is developers struggling to make a living and the second one is the toxic behavior 14:16.440 --> 14:24.360 from the community members wearing developers down. So to me, these two reasons, basically, 14:25.880 --> 14:33.160 are based on the same root cause that I was mentioning earlier. The members of the community 14:34.120 --> 14:40.600 are the users, the consumers, that are demanding too much from the developers without giving 14:40.600 --> 14:47.160 luck, without even thinking the developers, right? Like, lots of developers feel like they need 14:47.160 --> 14:52.920 to keep doing the work, but they don't feel like this work that they are doing is appreciated 14:52.920 --> 14:57.800 by the people that are actually receiving all of this value. 15:04.040 --> 15:09.640 And one thing I'd like to highlight is the fact that you can download the piece of software 15:09.640 --> 15:15.400 for free doesn't mean that it doesn't have value, right? And I know all of you probably know this, 15:15.400 --> 15:23.720 but it's also easy to forget. It's like, you download something, it's free and it's easy to forget. 15:23.720 --> 15:30.120 How much value is in that? All the hours, all the effort that was put into creating and 15:30.120 --> 15:36.200 maintaining the piece of software has a lot of value. And it has that value, even if it's 15:36.200 --> 15:41.640 like completely open source, even if the person that did the work did it all in their spare time, 15:42.840 --> 15:50.600 it still has a lot of value, right? It's a valuable piece of the world. And again, 15:50.600 --> 15:58.920 I know we all know this, but it's important not to forget it. And so the next step is to stop 15:59.320 --> 16:07.400 framing maintenance as more of it to you, as like, oh, it's the right thing to do, right? 16:07.400 --> 16:13.640 You created this piece of software and it's now used by hundreds of thousands of people. 16:14.840 --> 16:21.080 And yeah, it's the right thing to do. You should continue doing it for free and you shouldn't 16:21.080 --> 16:26.920 complain because you know, like you want to make the world a better place. So yeah, if you 16:27.000 --> 16:32.200 need to stay up all night fixing that security bag and yeah, 16:32.200 --> 16:36.600 nobody's going to pay you for that, but it's the right thing to do, right? So you should keep doing it. 16:36.600 --> 16:43.560 So that's not healthy. It's fine if this is something that you enjoy. If this is your passion, 16:43.560 --> 16:50.360 if this is something that you'd like to do, yes, that's great. But when it becomes a 16:50.360 --> 16:56.040 nobligation that you have to do out of guilt, then it's not rating more. And then it's when 16:56.040 --> 17:03.160 it leads to burnout, right? And so we need to stop thinking about this like the right thing to do, 17:03.720 --> 17:11.240 and thinking about this as work. And it's great if you like to do this work, but it's still work. 17:12.200 --> 17:24.680 And another factor is the maintainers are treated as vendors. And this means that companies feel 17:24.680 --> 17:32.600 entitled to send them an email saying like, oh, we need to now have compliance with our procurement 17:32.600 --> 17:38.680 of software and we need to know like all the dependencies, what security measures they are taking. 17:38.680 --> 17:43.960 So why don't you tell us what are the security measures that you're taking? These are emails that 17:43.960 --> 17:51.240 people have actually received like free software containers that are paid 0 cents by these companies, 17:51.240 --> 17:56.840 and they get emails from companies asking them to like do some kind of security audit. 17:59.080 --> 18:04.040 Who gives you any right to send me these emails, but they send the emails, right? And 18:05.000 --> 18:09.560 the maintainers don't have any kind of protections. And we all know that the licenses say, 18:09.560 --> 18:13.640 well, these comes with no warranty, right? Yeah, sure. It comes with no warranty. 18:14.360 --> 18:19.480 But companies can still send you these kind of emails, and it's not nice when you are under this 18:19.480 --> 18:24.760 kind of scrutiny. In particular, if there is some kind of security vulnerability in your software, 18:24.760 --> 18:32.120 it can become a lot of pressure from some of these super-entiled companies. 18:33.080 --> 18:39.000 So what happens if we don't fix it? Well, if we don't fix it, there's likely a lot of people that 18:39.000 --> 18:46.440 will learn out, a lot of people have already burned out. And what happens then is that this critical 18:46.440 --> 18:55.400 dependencies, this piece is that we talked about in the XKCD comment, get unmanned. And anything 18:55.400 --> 19:00.920 that can happen, right? Like there can be security vulnerabilities, there can be all kinds of problems, 19:00.920 --> 19:08.280 incompatibilities, and if we don't take care of our maintainers, they will not be able to take care 19:08.280 --> 19:17.480 of the software. All right, so and the other part of the talk is money. So how do we make money with 19:17.480 --> 19:26.120 free software? This was actually one of the first questions I had when I first learned about free software 19:26.200 --> 19:32.600 back in the year 2000. I was like, huh, and how do you make money out of it if you give 19:32.600 --> 19:38.920 this software for free, right? And the answer to that was, well, you charge for features and support. 19:39.640 --> 19:47.560 And surprisingly, or maybe not surprisingly, that's mostly the answer today. Still, 19:49.400 --> 19:54.360 25 years later, 26 years later, that's still the answer today. 19:56.600 --> 20:02.760 And so yes, as I said, like, I work at the consultancy, and this is what a consultancy does. So yes, 20:02.760 --> 20:09.640 you can charge for features and support. It is a thing, but it has a bunch of problems. 20:11.560 --> 20:15.240 One of the problems is that someone has to pay the development cost upfront, 20:16.360 --> 20:23.480 because most customers are not willing to say, oh, well, we want this piece of software. 20:24.440 --> 20:28.760 Yeah, you can develop it from scratch, and we will pay you to develop it from scratch. 20:28.760 --> 20:35.320 That it's very hard for that to happen. Instead, what happens is that someone, it can be you, 20:35.320 --> 20:40.600 or it can be someone else, right? Like, you can give support and develop features for software 20:40.600 --> 20:46.040 that someone has developed. It's fine. That's part of the open source model. And so someone 20:46.040 --> 20:50.680 has to have paid that development cost first. And when I say pay the cost, 20:51.400 --> 20:55.880 like, if you are developing this software in your nights and weekends, you are paying the cost, 20:55.880 --> 21:00.360 right? You're paying the cost with your spare time and your spare energy that you could be 21:00.360 --> 21:05.320 using for something completely different. But you're using it to develop this piece of software. 21:05.320 --> 21:11.480 That eventually, you will be able to sell support and new features and so on. 21:12.360 --> 21:18.680 But you're paying that cost. So someone needs to pay that cost upfront. And it's part of the 21:18.680 --> 21:23.720 this charge from features on support. It's part of the model. Yeah. Well, first you pay that 21:23.720 --> 21:29.800 cost upfront. And then you can charge for features on support. Then the next problem is that 21:29.800 --> 21:36.040 it's hard to get steady income. So if you're charging for features, once you develop the feature 21:37.240 --> 21:41.960 that feature is finished, and you need to find the customer willing to pay for the next feature. 21:42.760 --> 21:49.720 If you are charging for support, that's better, right? But you need to get enough customers that 21:49.720 --> 21:57.400 are add up to the containers that are supporting this software. Because customers are 21:58.360 --> 22:07.480 unlikely to want to pay for the full cost of developers, all the developers needed to support 22:07.560 --> 22:15.320 and keep maintaining the software. And finally, it's one of the first things that get cut. 22:15.320 --> 22:22.120 So if the customers for any reason are in a less good financial situation, 22:22.120 --> 22:27.800 it's very easy for them to say, oh, yeah, we have this support contract for this free software piece. 22:27.800 --> 22:33.560 Well, yeah, it's free software. Let's just not keep paying this support contract anymore. 22:34.520 --> 22:43.320 So yeah, that's like this works, but it's not easy. So of course, this is like a lot of people 22:43.320 --> 22:49.160 have realized that it's not easy. And a lot of people have tried to do business around free software 22:49.160 --> 22:56.760 and they found that it was a lot of work. So instead of doing that, they have come up with another 22:56.760 --> 23:05.320 model, which is called open core. So I haven't let you alone for a long time. So let's now do a show 23:05.320 --> 23:13.480 of hands. Who knows what is open core? All right. So most of you, but not everybody, I'm surprised. 23:13.480 --> 23:19.480 I thought it would be more. Anyway, so open core in brief is publishing some pieces of software 23:19.480 --> 23:26.280 as open source, while keeping some other pieces as close source and charging licensing cost for that 23:26.280 --> 23:34.120 other piece. There is a spectrum of open core. Not all open core software is the same. I heard 23:34.120 --> 23:42.040 someone say shit. And yeah, I mean, open core is not open source. There is a spectrum. Some, 23:42.920 --> 23:51.240 some open core is kind of like weight, right, that you're like hooking your customers into 23:51.240 --> 24:01.480 the open source piece and then reeling them into the enterprise version. But other open core 24:01.480 --> 24:06.840 is a lot nicer, where you could say, well, the open source version is kind of like the free 24:06.840 --> 24:13.560 for all weighting area and airport where the enterprise version is the VIP lounge where you have 24:13.560 --> 24:21.400 amenities and it's nicer, right. And yeah, and there's a lot of things in between. But the point 24:21.400 --> 24:31.240 is that the closed source version is not open source. And the company is paying for the open source 24:31.240 --> 24:39.160 development with the licensing fees that it's getting from the known open source. But. 24:42.360 --> 24:48.760 So I thought it would be interesting to look at the projects that are part of the CNCF 24:48.760 --> 24:55.480 landscape for those of you that are not in the Kubernetes space. The CNCF is a foundation cloud 24:55.560 --> 25:03.480 native computing foundation is a foundation inside the Linux foundation. And it has this like 25:03.480 --> 25:11.160 category of projects and you can be like incubating and and sandbox and graduated. So these ones 25:11.160 --> 25:20.200 are like the incubating and graduating graduated project of the CNCF landscape. And so I went through 25:20.200 --> 25:28.280 all of them to find out which ones were open core and which ones weren't. It took me a long time 25:28.280 --> 25:36.760 to do this. The whole train ride from Munich to Brussels. Anyway, so I found out that this ones are 25:36.760 --> 25:43.080 the ones that are open core. At least I could find that they were open core. It's not easy to find 25:43.080 --> 25:48.360 this information. They it's not like they say, hey, we are open core very big. You have to like 25:48.360 --> 25:55.560 dig in. Anyway, I thought it would be more, but still quite a number of them are open core. 25:56.280 --> 26:06.200 And then this ones were donated by companies that were using them for something. It could be that 26:06.200 --> 26:12.440 they did it like as part of their infrastructure or they did it together with some other 26:12.440 --> 26:18.200 customer and so they did some project and they created this thing and then they donated it 26:18.200 --> 26:24.040 to the CNCF. And these are like a bunch of very different companies. It includes companies like 26:24.040 --> 26:34.120 Google and Red Hat, but it also includes like Spotify or Lyft or Alibaba. So one. And then the others, 26:34.120 --> 26:40.760 the remaining ones are the ones that were developed by the community. So I think this shows a little 26:41.080 --> 26:47.000 bit of the who pays the up from cost, right? Like even the blue ones, the ones that were donated 26:47.000 --> 26:54.280 by companies, the up from post was paid by companies. So it's hard for the community to develop 26:54.280 --> 26:59.800 things. And I know there might be errors here. I did my best, but it's not super easy to find like 26:59.800 --> 27:07.880 all the information. Anyway, this was kind of anic data. And now for a different kind of anic data, 27:08.760 --> 27:14.760 there's a hashic or terraformed snafu. So show of hands who knows what this is about. 27:16.680 --> 27:24.360 Okay, not that many. Less than who knew about open core. All right. So on August 2023, 27:25.400 --> 27:31.240 hashic or announced that they were moving all of their projects to this business source license. 27:31.640 --> 27:38.360 For those that don't know, hashic or you used to have a bunch of open source projects, 27:40.200 --> 27:48.120 like terraformed and vault and a bunch of others. And at least to me, hashic or was an example 27:48.120 --> 27:53.560 of a successful open source company. They were doing open source and they were successful. 27:54.280 --> 28:02.040 Until they paid this. They switched to this bustle licenses that is not open source, right? 28:02.040 --> 28:07.240 It was a license that if you were an individual using terraformed, you were fine. But if you 28:07.240 --> 28:14.840 were building products on top of terraformed, you had to pay them. And of course, they were going 28:14.840 --> 28:22.520 after the big fish or we don't know. But we expect that they were going after the big fish, 28:22.600 --> 28:28.520 but in their going after the big fish, they also went after the small fish as well, right? So 28:28.520 --> 28:36.680 they were affecting everyone that was using their product. And this was done after, oh sorry, 28:37.480 --> 28:45.320 this was done after 10 years of having terraformed be open source. So a lot of things had been built 28:45.320 --> 28:50.360 on top of terraformed. A lot of companies were building products on top of terraformed. 28:51.080 --> 28:59.160 So what happened was that this company said, well, no, we are not going to pay to this bustle 28:59.160 --> 29:07.560 license. And so they got together and they created a fork that they first called it OpenTF, 29:07.560 --> 29:15.720 then they called it Open Tofu. Then they created this open fork. Not only is it open source, 29:15.880 --> 29:23.560 but it's also open governance. So it's really a community-based project that is not controlled 29:23.560 --> 29:30.360 by one company, but like many companies are contributing, many people from different places are 29:30.360 --> 29:41.000 contributing. And what I find the most interesting from this story is that when they started 29:41.560 --> 29:48.520 several companies committed to donating the time of their employees to this project, right? 29:48.520 --> 29:55.800 Like they committed to five years of five employees or three years of three employees. And so on. 29:55.800 --> 30:02.680 So they, they were willing to spend a lot of money because like five years of five employees, 30:02.680 --> 30:09.240 it's a lot of money. They were willing to spend a lot of money on keeping this project open source. 30:09.880 --> 30:17.400 They were not willing to pay Hashikorp for this. But so that's one interesting point. 30:17.400 --> 30:26.520 And then the other is they had not done this before, right? So when this minutes of the project 30:26.520 --> 30:35.080 going away appeared, then they were, yes, we are contributing our employees, but not before when 30:35.080 --> 30:41.160 it was owned by Hashikorp. So I don't know, I think all of this is kind of interesting. 30:43.400 --> 30:51.080 Okay, so we talked about the traditional, the open core, and now I'll just cover a few other 30:51.080 --> 30:58.360 options. There's a bunch of them. So one option is to turn to donations. So there's a bunch of ways 30:58.360 --> 31:05.800 you can ask for donations. One option is the GitHub sponsor thing. So if you are on GitHub, 31:05.800 --> 31:13.640 you can just say that you want to receive donations. And then if, as a dependency, so someone 31:13.640 --> 31:21.400 that is using your project goes to GitHub, they can check all the dependencies that are open to the 31:21.400 --> 31:25.560 donations. And they can say, oh, well, okay, so I have these 10 dependencies that are open to the 31:25.560 --> 31:34.680 nations, I will donate one URL per month to each or whatever amount, right? So that's one thing. 31:34.680 --> 31:40.440 And GitHub claims that they have distributed $40 million using this. So that's good. 31:42.200 --> 31:50.040 Then another option to ask for donations is through open collective, which is a platform that has 31:50.120 --> 31:56.760 different hosts that have collectives. It's hard to get all the terminology right, but yeah. 31:57.720 --> 32:06.120 So the open source collective, which is one of the hosts, has 2692 collectives. So a lot of collectives. 32:06.120 --> 32:12.760 And then the open collective Europe, which has open source and other things, not just open source 32:12.760 --> 32:18.840 projects, has like 400 more collectives. So all in all, it's a lot of collectives. What are these 32:18.920 --> 32:24.840 collectives? They are projects, right? Different projects. Like for example, oh, yes, the video 32:24.840 --> 32:31.800 system is there. And you can say, okay, we want to develop this feature and you can participate 32:31.800 --> 32:37.800 in the crowdfunding campaign to get enough money to develop that feature. Or you can say, well, 32:37.800 --> 32:43.800 subscribe to us. And then I'd like five euros per month. And then, like, if enough people subscribe, 32:43.800 --> 32:50.680 then that adds up to someone's salary, right? And it's great, like seeing the list of projects 32:50.680 --> 32:55.800 that are available and seeing the projects that have received donations, that's great, it's like, 32:58.200 --> 33:03.800 I think, no, yeah. It shows that a lot of people are actually donating and helping this projects 33:04.600 --> 33:11.080 be successful and helping employ developers like this projects when they reach enough money, 33:11.160 --> 33:17.480 they can employ people to work on this. And that's great. But also, it requires you to be 33:18.440 --> 33:25.160 asking for donations and constantly saying, hey, we have this page, please donate. And we have 33:25.160 --> 33:31.000 this campaign, please donate. So, like, it feels a little bit like begging, but it's working, 33:31.000 --> 33:36.680 at least for some projects. Another option is to go the route of government funding. 33:36.680 --> 33:43.960 So, NLNet is one of the organizations, not the only one. So, one of the organizations that 33:43.960 --> 33:51.320 distributes money from the European Union's next generation internet program. So, that's like a 33:51.320 --> 33:57.800 big pot of money. And you have to go through a lot of bureaucracy to get that money. So, NLNet 33:57.800 --> 34:05.960 and the other organizations that do the same facilitate this. They do most of the paperwork and they 34:05.960 --> 34:14.600 select projects that are going to receive the money. And then they give them the money and 34:14.600 --> 34:22.680 they take care of most of the bureaucracy. According to the page, the last time I saw it, like 34:23.480 --> 34:31.160 1,500 projects, 1,500 plus projects have received money through this way. And these are 34:31.160 --> 34:36.760 some of our large projects and well-known and some are not well-known projects and small projects 34:36.760 --> 34:46.040 that have proposed to do something that was deemed useful for the world. And so, they received 34:46.040 --> 34:52.520 this grant. The grants are usually not very big, but they can help a project that is struggling. 34:52.520 --> 34:59.400 They can help a project to get something done and get more features. In general, it's about 34:59.400 --> 35:05.080 developing new features or even new projects. It's not about maintaining existing software. 35:08.120 --> 35:15.080 Another government funded way of getting money is the summer intake fund, which is part of the 35:15.080 --> 35:20.520 summer intake agency. It's from the Sherman government, but actually they are helping maintain 35:20.520 --> 35:26.360 the digital infrastructure around the world. So, it's not just for German projects or German 35:26.440 --> 35:33.080 developers, but rather for the projects that make up the digital infrastructure of the world. 35:33.080 --> 35:40.440 So, they are set on making this situation that we are talking about better. And in the case of the 35:40.440 --> 35:46.760 summer intake fund, they're actually paying for maintenance. So, not for new features, but for 35:46.760 --> 35:53.240 keeping the projects well maintained. And in general, they are paying to larger 35:56.680 --> 36:10.200 projects. Yes, okay, that three was over something. All right. So, I was saying, in general, they pay 36:10.200 --> 36:18.920 larger projects. Okay. And finally, the last alternative that I want to mention is the open source 36:18.920 --> 36:26.280 pledge. This is something that started a couple of years ago and I find this initiative very interesting. 36:27.000 --> 36:36.280 So, the pledge is basically saying, we as a company, like, comment to the 19 2000 dollars per 36:36.280 --> 36:44.920 developer per year, two open source. And the companies can choose who they donate to and how they 36:44.920 --> 36:53.000 donate to. And the only thing that they need to do is donate at least 2000 dollars per developer 36:53.000 --> 37:01.160 per year per year. Some companies choose a bunch of projects that they depend on and they donate 37:01.160 --> 37:08.040 a little bit to each project. And some other companies, I've worked like the project of the month 37:08.040 --> 37:14.040 and they donate like a big amount of money to a project each month, depending which one was the 37:14.040 --> 37:23.480 project selected that month. But yeah, in the end, it has to add up to 2000 dollars per developer 37:23.480 --> 37:30.520 working at the company per year. The goal of the open source pledge is to establish a new social 37:30.520 --> 37:35.880 norm in the tech industry of company Spain, open source containers so that burnout and related 37:35.880 --> 37:42.040 security issues can become a thing of the past. So, in the abstract of my talk, it talked about 37:42.040 --> 37:49.800 acid time for an open source tax. And this is sort of like an open source tax, except it's voluntary 37:50.440 --> 37:56.440 and the company is the site who they donate to. So, it's not really how a tax would work because 37:57.640 --> 38:03.960 if it's a tax, everyone pays the tax and someone decides where the money goes. Well, this is kind of 38:03.960 --> 38:16.920 like you decide your own tax. All right, and finally, I feel like probably the most common way 38:16.920 --> 38:24.840 of making money out of free software is to be employed by a company that does something else 38:25.560 --> 38:31.720 but pays for open source developers. And I know a lot of people that are in this situation that 38:32.600 --> 38:39.160 they are working for another company for a company that is a for-profit company, but that some 38:39.160 --> 38:48.920 of their time is allocated to working on open source. It could be 20%, 30%, 50%, but to the company, 38:48.920 --> 38:55.800 it's a win-win situation because they are paying for the maintenance of this software that could be 38:55.800 --> 39:01.160 like a dependency or tool. Of course, I'm not talking about like the main core business of the company, 39:01.160 --> 39:07.000 the company should pay the developers working on its main core business, right? But like 39:07.000 --> 39:14.200 a dependency or a tool that the company is using and they give the developer the time to work 39:14.200 --> 39:20.520 on this tool and at the same time, they are getting the expertise of this person that can help them 39:20.600 --> 39:27.400 with whatever it is that the company does. So this is also a very common model of making money 39:27.400 --> 39:34.120 if you are a free software developer. All right, so we've seen a lot of different ways of making money. 39:34.840 --> 39:43.480 Is this enough? Is this enough? Not really, not really, it's not enough. It's not enough. 39:44.440 --> 39:52.200 All right, why is it not enough? No, no, you, I'm not asking you to answer. So it's not enough. 39:52.200 --> 39:56.840 It's not enough for many reasons. One of the reasons is like it's really complicated, right? 39:56.840 --> 40:04.600 Like you need to set up a collective and ask for money or you need to like run the grant application 40:04.600 --> 40:10.600 and when it runs south, like you need to correct the new grant application, which is kind of like academia. 40:10.680 --> 40:25.000 And it's, it's still complicated. But also, it's not steady, right? Like it's, it's really hard to 40:25.000 --> 40:31.560 make this. Let's say, okay, when you have a job, you have a job and you work on it and you get steady 40:31.560 --> 40:36.920 income. But all of these that we have been describing, it's, it's really hard to make it steady. 40:37.720 --> 40:47.720 But that's not the only reason it's not enough. And I want to propose here, uh, perhaps a controversial thing. 40:48.760 --> 40:56.840 Which is that this, everything that I described is about paying senior developers that are already 40:56.840 --> 41:04.120 maintaining core pieces of infrastructure. And of course, we should pay that people, those people, right? Like, 41:04.600 --> 41:12.200 for sure, I'm not saying we shouldn't. I'm saying that's not enough. Because if you only pay senior 41:12.200 --> 41:20.600 developers that are already maintaining core pieces of infrastructure, you're not giving space for 41:20.600 --> 41:29.720 people that cannot afford to work on open source in their nights and weekends, right? Being able to do 41:29.720 --> 41:37.160 a second show for free during your nights and weekends is a privilege. And it's a privilege 41:37.160 --> 41:46.280 that a lot of people don't have. In particular, women in the world, still today, are disproportionately 41:46.280 --> 41:54.760 in care, in, in charge of care taking responsibilities. So that's like, for many women, that's their 41:54.840 --> 42:06.280 second show, right? And thus, when you only pay senior developers, you're not making space for 42:06.280 --> 42:11.800 all of these people. So I know this is controversial, but I just wanted to plant the seed there 42:11.800 --> 42:21.800 of this idea. But now, I want to imagine a better world. And I want to imagine a world where we 42:22.200 --> 42:31.720 have teams of people that are paid to work on open source of work. Some are senior engineers 42:32.760 --> 42:37.880 and some are junior engineers that are coming up to speed and helping the senior engineers 42:38.920 --> 42:44.680 helping them with, like, bug reports or documentation or whatever tasks the senior engineers don't 42:44.680 --> 42:51.000 feel like doing. And so that eventually they can take over, they can create their own 42:51.960 --> 43:00.680 projects and be maintainers and so on. But also, like, all of them are paid a steady salary if they 43:00.680 --> 43:09.400 want to, right? This is the utopia that I like all of you to think about and think that this is 43:09.400 --> 43:16.360 possible, right? And you can say, well, but that's too much of a utopia and like, no, that's not really 43:17.240 --> 43:22.760 possible. But like, think back to, like, when in 1984's Dalman created a new project 43:23.800 --> 43:31.800 and the situation where we are today that we should say, like, 97% of software depends on open source, 43:31.800 --> 43:39.800 right? Well, today, we would already be in utopia, like, if you were back in 1984. So it's good 43:39.800 --> 43:45.560 to have a goal. And I think that's the goal that we should have that everybody that works in open 43:45.560 --> 43:53.320 source should be compensated if they want to, right? And I think they're still a great place for 43:53.320 --> 43:58.840 volunteer work. I'm not saying, oh, no, there shouldn't be volunteers. No, volunteers are super necessary. 43:59.800 --> 44:06.440 But if someone wants to do it as their full-time show, they should be able to do it. And that's the 44:07.240 --> 44:13.480 that's the utopia that I want to to plant. So some ideas to make that possible. I'm first 44:13.480 --> 44:20.840 on foot and then some ideas. First, remember that open source is already making the world better 44:20.840 --> 44:29.240 today that people working in open source are adding a lot of value today. And so people working 44:29.240 --> 44:35.480 in open source should be able to do it as their full-time show if they want to, right? Because we 44:35.480 --> 44:42.840 are adding a lot of value. It's not that we are begging for scraps. It's that we are asking for 44:42.840 --> 44:49.560 people to be fairly compensated for all of the value that is being added. All right. And now, 44:49.560 --> 44:59.640 some practically, yes, perhaps not so utopic. First is about companies giving back. I think the 44:59.640 --> 45:07.560 open source pledge is a great idea. This thing about it being sort of attacks, but with total freedom 45:07.560 --> 45:17.080 of who you are donating to. So if you work at a company that uses open source and hasn't signed 45:17.080 --> 45:24.120 the pledge, I invite you to talk to the management of your company and ask them to consider signing 45:24.120 --> 45:32.440 the pledge. And donating this $2,000 per developer per year. And if they say no, well, maybe 45:32.520 --> 45:37.880 donate a little bit less, right? $2,000 per developer per year may be too much. If you are 45:37.880 --> 45:43.480 in a country where that's a lot of money or if the company is not making that much money, 45:43.480 --> 45:50.680 whatever, but something. It's good to start with something. You will not get the batch of like we 45:50.680 --> 45:56.680 are in the pledge, but you can still post a blog about the donations that the company is doing for 45:56.760 --> 46:05.800 open source. And that's still good. And if possible, I would advocate for donating a steady 46:05.800 --> 46:14.440 amount every month, rather than big lumps of money to different projects. I getting steady income 46:14.440 --> 46:21.000 is more important even if it's less. Now, the next one is a little bit controversial, and it's 46:21.000 --> 46:26.040 about foundations employing developers. And it depends on the foundations, some foundations are 46:26.120 --> 46:33.080 already doing it. And it might not be controversial, but other foundations are not doing it. And so 46:33.080 --> 46:41.000 it might be controversial. But I want to invite all of you if you are part of a group that is not 46:41.000 --> 46:46.920 employing developers. And it's receiving donations, right? Like if you are in a foundation that 46:46.920 --> 46:53.320 receives donations and it's not employing developers, start to explore this subject and start to think 46:53.400 --> 47:02.920 about how could we make it work? How can we employ people and pay them for their job without 47:02.920 --> 47:07.960 discouraging volunteers? Because that's the problem, right? Like if you pay someone like are all the 47:07.960 --> 47:14.200 other volunteers going to say, oh, I'm not going to do it anymore because he's paid and I'm not paid. 47:14.200 --> 47:21.240 So yeah, I'm out. Okay, we don't want that. So we need to start thinking about this and start thinking 47:21.320 --> 47:28.760 how we can make it work. I don't have the answer, but if we stop from having the conversation 47:28.760 --> 47:34.680 because we don't have the answer, then we are never going to reach the answer, right? So please start 47:34.680 --> 47:43.080 this conversation at whatever organization you are in. And yeah, finally I have this other idea, 47:43.080 --> 47:50.120 kind of crazy, or maybe not so crazy, but there's the open source pledge. So I'm proposing the open source 47:50.280 --> 47:56.120 employment pledge, which is, well, if you are not willing to demand money, maybe you are willing 47:56.120 --> 48:06.200 to donate a time of your employees. So every 20 developers in your company, 50% of one person's 48:06.200 --> 48:15.000 time goes to them developing open source. And that 50% is like completely free of company influence, 48:15.000 --> 48:20.840 right? It's not like, oh, you work on this project, but we tell you what to do. Now, it's 50% of 48:20.840 --> 48:29.960 their time for maintaining the project, however they see fit. And so, and if the company is not willing 48:29.960 --> 48:36.200 to pay money, but they are willing to do this, that's an option. I have not created the website. 48:36.200 --> 48:41.000 You are all, like, whoever here feels the initiative to create the website, you can go and create 48:41.000 --> 48:48.920 the website. You have my blessing. I'm finally, I'm a big advocate for teams. So I feel like, 48:48.920 --> 48:56.680 if you are a solo maintainer and you are feeling a lot of pressure from keeping up with your 48:56.680 --> 49:04.360 project and you can't do it anymore, it's time to show up with other people and like help each other, 49:04.360 --> 49:09.880 right? When you need to take some time off, someone else can take over and when that someone 49:09.960 --> 49:14.440 else needs to take some time off, you can take over. So I want to invite you, if you are a 49:14.440 --> 49:22.120 solo maintainer, to try to find other people in your space and form a team. I'm a big believer in 49:22.120 --> 49:28.040 teams. All right. And with that, I'm done with this talk. Thank you very much. 49:28.120 --> 49:41.400 Thank you. I think we have, like, two, maybe three minutes for questions. 49:42.520 --> 49:44.680 So if there are some questions. 49:50.840 --> 49:55.880 So good morning, everybody. And thanks for this marvelous talk. I think for us, 49:55.880 --> 50:04.040 there needs this kind of talk some more than many others. So, thank you for your presentation. 50:04.040 --> 50:08.680 And I don't have to be brief, since I've plenty of notes about this. So first of all, 50:09.560 --> 50:15.880 voluntary text is there is not such a voluntary text. We need some unvoluntary text, 50:17.400 --> 50:25.800 particularly for big companies, since the positive externalities are so high. And this is a critical 50:26.200 --> 50:33.160 infrastructure, I keep it on time. It's a critical, such a for humanity, okay. We cannot 50:33.160 --> 50:41.320 flee for money, like road maintenance do not, do not back, okay, for money, okay. It's like 50:41.320 --> 50:47.240 road maintenance. So we need something more effective. And I think it's high time, not for reasonable 50:47.240 --> 50:55.400 proposals, but for unreasonable concrete proposals, like the part of Texas. So this is 50:55.720 --> 51:02.840 like irreseeable, defective company should believe or for this kind of stuff, but I think 51:03.560 --> 51:10.200 you want to go in the direction. So this is the first point, second point. One second, one point, 51:10.200 --> 51:18.680 at least one point. To me, your talk resonates so much with feminist theory, for instance, 51:18.680 --> 51:24.520 the difference between productive work and liberty work. And we need as developers to organize 51:24.520 --> 51:35.160 sorry, do you have a question? One second, one second. If it's more of a statement than a 51:35.160 --> 51:44.760 question, stop with your own talk. Any actual questions, please? No, no, sorry. All right, 51:44.760 --> 51:51.080 sorry, we can keep discussing, somewhere else, where should we go to keep discussing? 51:51.080 --> 51:58.440 Thank you, Mark, a great talk. Carry on the conversation after it.