WEBVTT 00:00.000 --> 00:21.000 All right, I think we're going to get started, maybe one or two more people coming in. 00:21.000 --> 00:26.000 I think there's a chair over there, no, okay. 00:26.000 --> 00:35.000 Round two of this fast-paced day of building Europe's digital public infrastructure, I give a sticker. 00:35.000 --> 00:49.000 I'm Ben Servi, I'm a co-founder and president of the Foundation for Public Code, which is a nonprofit that works to provide frameworks for collaboration between public administrations who are building digital public infrastructure. 00:49.000 --> 01:09.000 We've worked on a variety of projects, including Gavstack, we're founding member of also the digital public goods alliance, and also have worked with code bases like DCDM and Ex-Ro to turn into the types of entities that we're now trying to form a template for, which is what I'm going to talk about in this conversation. 01:09.000 --> 01:28.000 This sort of goes out also as a previsualization of the feedback that we're going to be giving to this process that the European Commission is now running that is about feedback on the open digital ecosystem package of legislation that they're going to be announcing soon. 01:28.000 --> 01:41.000 This is a critical opportunity that we have right now to think about new and very different ways that we could imagine how we make software together in the public administrations. 01:41.000 --> 02:03.000 Really what this is an exercise in speculative industrial strategy, which means we're sort of spiraling all the way up from talking about individual projects to talking about frameworks collaboration, talking about different types of legislation and then eventually an entire kind of world model or narrative universe design around which all of these things can interrelate. 02:03.000 --> 02:30.000 So what we really need to discuss here is how the open source community, which tends to be fairly diffuse and a fairly sort of unstructured in a lot of ways, becomes a thing that has representation in the world of policy, which means you need to have some kind of organizational there there in order to participate in this sort of ecosystem of funding and collaboration subsidy procurement, these types of things. 02:30.000 --> 02:59.000 So what we're proposing is that there should be a new type of NGO either as an actual construct within the local laws of the member states within the European Commission, European Union or potentially as an overlay like the B core where you have a set of sort of criteria that an existing typology of NGO can comply with in order to become a public product organization and sort of get the seal of approval to stamp. 02:59.000 --> 03:11.000 And so we're saying that the public is a typology of nonprofit that serves as a code based steward or a product steward. 03:11.000 --> 03:25.000 We're actually moving from talking about code bases to talking about products, because it turns out that a lot of what a successful implementation of digital public infrastructure looks like has a lot more characteristics to it than just the code base, which involve things like branding, 03:25.000 --> 03:30.000 outreach to other potential peer replicator organizations. 03:30.000 --> 03:39.000 The continuous, obviously, the core part of the continuous integration of open license components that are like traditional issues around software stewardship, 03:39.000 --> 03:51.000 but also specifically in this context governance, which is where we sort of get into the idea that there's a relationship between policy and software that could be broken in this organization. 03:52.000 --> 04:17.000 So part of what's interesting about this is it looks, it's almost like a branching narrative from the way that software at scale has been produced previously, which is through the extractive silicon value venture model, which is to the idea that sort of teleology, the goal state of having a silicon value funded venture is that you're going to have 100x, for instance, return to your investors. 04:17.000 --> 04:30.000 In order to do that, there are a number of limited strategies that these companies have found to do that, which is to sell their user information, so lock in their users and have all sorts of behavior that is sort of anti user, but pro shareholder. 04:30.000 --> 04:39.000 So it turns out that software might need to have a new set of priorities, and some of those priorities can be provided by recognizing that a confunction as. 04:40.000 --> 04:52.000 Interesting, there's something happened with the renderer there, so this is the governance model we're supposed to be talking about here, which includes at least two implementing public administrations. 04:52.000 --> 05:10.000 And they're local development vendors, so the idea around this is to immediately create a diversified community where you do not have a single stakeholder that can be hit by a bus or can be hired by meta or in other ways, removed from the process of providing public infrastructure. 05:10.000 --> 05:18.000 The governance model, though, really is about also creating a situation which you have sort of requirements led development. 05:18.000 --> 05:33.000 A lot of times in the open source community, you have a lot of people that are engineering led, there's someone has a very good idea about a very specific type of implementation of a technical problem, and so they make a code based, and then people are sort of, you know, can decide to work with that if they want to use. 05:33.000 --> 05:51.000 So this is not to suggest that all open source code bases should follow this type of model, but there should be a model that is sort of facing the public institutions in the sense that it helps them de-risk the idea of procuring this thing if they are participating in the model of governance. 05:51.000 --> 06:06.000 And that the software is upholding, but also it serves as a much more structured way of sort of choreographing and orchestrating the development and maintenance of the code and the product. 06:06.000 --> 06:22.000 We maintain a roadmap that is part of the governance model, so there is a mechanism by which the members is generally assembly and the directors decide how to prioritize the the work packages that come out of the roadmap. 06:22.000 --> 06:40.000 And also, super importantly, in terms of writing sustainable products is to have an outreach toward funding mechanisms, whether those are perfurs or members that are paying fees or participating in structured sustainable funding. 06:40.000 --> 07:03.000 So we've encountered a couple of patterns that we already have worked on that before we sort of formalize this template around how public products can be made and the first one is what we call the du-la pattern, which is, which is that a single public administration through either its own internal digital capacity or through its vendors. 07:03.000 --> 07:15.000 It goes out and fixes a problem that they have, maybe something in the way they're parking system works or they've decided they wanted to build a management system for their schools or they have some way of registering for public housing. 07:15.000 --> 07:30.000 So they have done that and this is one of the things that's happened in the past probably about the last 20 years of development in the public sector around open source is that a lot of times there's a real motivation to build something that is considered technologically sovereign. 07:30.000 --> 07:45.000 There was already in the 90s even the discourse of technological sovereignty and in Europe, but a lot of times what happened was that they would take with the mantle of an source and say, hey, we're going to go out there and build a source solution and they would implement and about six years later. 07:45.000 --> 08:02.000 There we go, what the hell is this? Where is our service level agreement that we can make it an 800 number call to support agency and there is not because those people were contractors hired by the public administration by the previous public administration and they have no idea what the hell is going on. 08:02.000 --> 08:25.000 And so that first wave of that process was sort of, you know, a bit of a panic by the time that that arc reached the end is that there was a fear that open source in the eyes of the public administration had been sort of spoiled by this, this inability to understand what to do with a code base in the long term. 08:25.000 --> 08:34.000 And so the first round of our work at the foundation for a public code was that we would help a lot of times it was almost like code base seen investigation. 08:34.000 --> 08:49.000 Go ahead and sort of see what's going on with the application that's deployed and figure out how we can build something where we go and find other replicators like we need to build a community of practice around this code base and say, okay, this thing is really useful actually. 08:49.000 --> 08:55.000 Let's see if we can sort of get people around it that are you know consultants that understand what's going on with it. 08:55.000 --> 09:09.000 SMEs that are working in the space and then help them come together to form an external entity that's outside of the original administration because this is the other issue that you come up against when you try and look for replicators across. 09:09.000 --> 09:25.000 And so what we're going to do is we're going to be able to build an administration is that they don't want to use software that's in control of someone else obviously so they're there so if they are participant in the new bodies governance model, then they can all feel like they have a toe in and they have some sort of control so that the risks for them as well. 09:25.000 --> 09:40.000 A lot of time this is what happened for instance with these themes that started it's a collaborative budgeting tool participatory budgeting tool built in Barcelona for this process called this making the super blocks which were a big. 09:40.000 --> 09:52.000 Urban planting intervention that allowed for the pedestrianized intersections and they allowed the citizens nearby to vote on how to furnish the square and it was a really good tool if they built. 09:52.000 --> 10:07.000 But it started to age and they realized that their budgetary they didn't really have the projected budget in order to take care of this thing that they built that they really enjoyed so they realized I mean Barcelona has always been very good at sort of marketing itself in a variety of. 10:07.000 --> 10:22.000 I had your friends and it turned out that in this case it was no different and they so they built a program to basically take the food base out of their own administration and put it into. 10:22.000 --> 10:38.000 public organization which is now the decid team foundation and it has a whole governance model is based on members assembly and that started to scale and it's scaled pretty rapidly and so that became one of the first examples of how something like a public product organization might work. 10:38.000 --> 10:58.000 So now in a sort of more mature environments there are a lot of other you know I should say in this sort of narrative universe in which the the public product organization exists a lot of new things are also starting to merge in other aspects of the of the sort of public ecosystem one of those things. 10:58.000 --> 11:17.000 is the open source program office inside of lots of public administrations and those those are the entities that are now facing into the prospect of making these types of code bases they're the ones who are there to talk to their other parts of the administrative public administration about the fact that there might be open alternatives. 11:17.000 --> 11:37.000 So there are a lot of other implementation if they're going to pretend or out because you know all the proprietary vendors bring the people to the golf game with the CTO city and so somebody internally has to say well actually there's this other alternative that's that's open and so now we have this prospect that you might come up against. 11:37.000 --> 11:50.000 multiple different public administrations realize they have the same need or that they're coming up against a contract with the vendor they don't like at the same time they're coming to end of a contract with the vendor didn't like at the same time and so they can. 11:50.000 --> 12:06.000 decide proactively to pull resources before the product even exists and say we're going to go out and basically have a seed round to build a public product collaboratively from the ground up and so that is essential essentially what's happening in a couple of cases now. 12:06.000 --> 12:17.000 I say that ex-road which is a co-product between Finland and Estonia as a it's a back in data exchange tool that works in the government of ministries. 12:17.000 --> 12:28.000 that that product serves as kind of a first template for this type of behavior so both there's these are two emergent patterns on how you might make a public product organization has long-term. 12:28.000 --> 12:39.000 this kind of maps really well to the idea of a you know this sort of classic layer cake technical diagram the the public product organization serves as a common core. 12:39.000 --> 12:47.000 for lots of different opponents they're useful across all the implementations and then the distributions are things that would exist in specific. 12:47.000 --> 13:00.000 legal or policy context that have constraints that are different from each other so maybe the French library system and the Spanish library system have different issues around privacy or even how long books are checked out whatever that is. 13:00.000 --> 13:10.000 you can build a distribution that's now a stable distribution across the different potential instances of libraries in cities and those relative domains. 13:10.000 --> 13:19.000 so the idea that that you have this mapping between the kind of political architecture and the technical architecture is one that's really interesting. 13:19.000 --> 13:34.000 and then you come to this idea of how you know I think one of the hardest things for open source code bases most often is their sustainability and support you know how do they how do they survive you know a lot of times I think there's another session here that's called free as in burnt out. 13:34.000 --> 13:42.000 that seems to be a real issue and so you know when working in the public sector there's this idea that there are multiple different strategies for. 13:42.000 --> 13:52.000 funding one of which would be to have a more traditional membership model where the people that are in the governance it have joined as members who are going to be implementing public administrations. 13:52.000 --> 14:09.000 pay a fee of membership fee which is part of a sort of core sustainable financial flow and then the product organization within the larger sort of ecosystem of these other emerging actors which I'll talk about in a minute including the European sovereign tech agency. 14:09.000 --> 14:29.000 the e-dick and the Ospo system various other horizon type of granting organizations because there is a coherent product organization has that has a long term to respond to these types of these types of public programs. 14:29.000 --> 14:45.000 long term. So probably most interesting to people in this room is that the model also sort of is about it's very different than the unicorn model right like this is the Silicon Valley model where sort of the winner takes all. 14:45.000 --> 14:57.000 is that it flattens the curve and you have a big public product which is available to everyone and within that situation there are lots of opportunities for SMEs like to. 14:57.000 --> 15:12.000 multi-ply and team as I call it here so you know you have the you have the people that are working on the implementations and the integrations which are at the front end of that you have the people that are developing common core roadmap work packages. 15:12.000 --> 15:41.000 And also contributing back from the distributions into the core based on some kind of political decision making of people decide I mean there's there lots of really interesting issues there in the way that policy and technology map to each other is like who decides what's in the common core and what's in it what's in a distribution you know there's like is that something that everyone needs or is that something in individual distribution needs and so having actual debate about that in a general assembly where people talk through that in a way exposes a type of policy. 15:41.000 --> 15:47.000 Transparency that doesn't exist you know outside of that world. 15:47.000 --> 15:55.000 One of the things that's really interesting that just happened is that the European digital interoperability consortium around. 15:55.000 --> 16:17.000 The digital commons was just formed and it's sort of trying to figure out what its remit is at the moment and so I'm sort of putting forward this this prospect that one of it's you know one of the possibilities there is that it helps scale these types of organizations big a lot of times procurement that happens because of the way that. 16:18.000 --> 16:31.000 Legislative boundaries work in in number states a lot of times it's easier for public administrations to procure from entities that exist in their own states which is you know there's this whole you in thing that's going on as well. 16:31.000 --> 16:38.000 People have been following along with that but it's an idea to create this like standardized entity for business across Europe so you can have. 16:38.000 --> 16:47.000 Procurement across these borders well the edict potentially serves as because it's essentially international treaty that all the number states that have joined it. 16:47.000 --> 16:55.000 have signed into it means that that entity exists across all of these number states and so potentially if you are. 16:55.000 --> 17:04.000 Public product organization that is being sort of stewarded inside of the edict that would mean that you could be procured by any of the number states and they were involved. 17:04.000 --> 17:13.000 So this is kind of the ecosystem that I was describing with a bunch of different public actors and sort of this like I said this kind of narrative universe design. 17:14.000 --> 17:17.000 of what the public product space looks like. 17:17.000 --> 17:28.000 There are agencies that are looking out for you know specific work package funding like the sovereign tech agency there are authors that are looking for connecting different ways of doing procurement. 17:28.000 --> 17:36.000 As groups potentially around some of these some of these projects are like I said before in the in the sort of burn raising capacity. 17:36.000 --> 17:42.000 going in together to form a new thing based on shared requirements that they've come up with in some other context of collaboration. 17:42.000 --> 17:53.000 The edict as a place that like large scale collaborative resources can be made and horizon of course and some other more familiar sort of European funding models to. 17:53.000 --> 17:59.000 Accelerate developments and innovation so one of the things I find really interesting about this also. 18:00.000 --> 18:11.000 In this sort of universe of open source there's also really potentially transformational effects that Europe could have on the process of open source which is that you know open source often suffers from not having. 18:11.000 --> 18:21.000 a design team that's as deep as the public as private products especially Silicon Valley products that have a ton of polish and design on them. 18:21.000 --> 18:37.000 Europe spends is one of the major entities that spends money on cultural development and cultural production and one of the problems here in Europe is that it does not recognize that. 18:38.000 --> 18:51.000 In a very conservative they're very technophobic essentially so I think I think that that if one works to introduce the idea that along with you know theater and dance and various other things. 18:51.000 --> 19:01.000 The software or something that was fun fundable as a as a cultural product is right now when you talk to your to the European Commission about how software is funded from the public sector. 19:01.000 --> 19:11.000 It is almost always an industrial product but this is part of the problem is that the reason why we all sort of live inside of American software jail with them. 19:11.000 --> 19:24.000 This is that is that they have spent a lot of time filling that gap that cultural products in software did not supply you know before so I think this is another huge turn. 19:24.000 --> 19:46.000 That's interesting to contemplate is that some of these SMEs that are working in the space of improving public products could be more culturally oriented especially if they have actors from the technical space that are working with them inside of this collaborative vehicle but I think having these public collaborative vehicles in which they can connect is one of the few places where you can find. 19:46.000 --> 20:04.000 You know a space in which the cultural community and the technical community have like a way to interact productively in a way that could be orchestrated by product people so and this is the thing that has been I think played the open source community for a long time and so the prospect of having this this type of. 20:04.000 --> 20:22.000 I think this is a very important part of the process would be a really interesting experiment but in general I think all of this is incredibly interesting to imagine just from a speculative perspective even if it doesn't get implemented because this is definitely a very different approach to how you could make software and scale. 20:22.000 --> 20:35.000 And then the current model that we're living in which is the Silicon Valley venture funded model so it's it's this is you know like I said something that is being offered up to the European Commission is part of feedback on their on their strategy. 20:35.000 --> 20:57.000 And so I think this is a really interesting moment to sort of consider you know how would you do this differently you know this is our proposal from having worked on these projects but it's just a it's a moment for consideration because we're obviously in a crisis right now the crisis gives us the opportunity to think about something that is different than what we inherited. 20:57.000 --> 21:04.000 And it's kind of an exciting time despite this slight terror of it all. Thank you very much. 21:04.000 --> 21:33.000 I would appreciate that you're thinking. Sorry. Yeah there are a number of different entities that are you know running kind of in parallel I think it's it's definitely part of the process of doing this type of stewardship at scale and I you know again there are lots of different I think. 21:33.000 --> 21:48.000 Part of the part of the next move in this in the introduction of it as a policy strategy is to bring these actors to the table that have traditionally been involved in this process and sort of see how we can build a fabric of a collaboration around around how to. 21:48.000 --> 21:56.000 And everyone has had a lot of experience around how these products are code bases sort of move into introduction and our maintains. 21:56.000 --> 22:06.000 And I think you know bring there's sort of a meta conversation to be had about collecting all the sort of wisdom in that and bringing that to the table. 22:06.000 --> 22:16.000 Yeah, I mean that. 22:16.000 --> 22:24.000 Oh, how is it just for the money without it being social games that is about nepotism and knowing the people and things like that. 22:24.000 --> 22:49.000 I mean that's that's always a risk and any organization including a proprietary one so I mean I think the idea with the governance model and the idea of having a members assembly is that there's some sort of peer policing and self-leasing it goes on in that situation but you know corruption is a thing that happens and in most political organizations so you have to be on lookout for it but but but I think it's you know when you have things that are. 22:50.000 --> 23:03.000 Driven by results I mean it's it's easier in this case if you have you know pull requests that are being reviewed and those work packages are tied to the pull requests that are that are accepted then it's a little bit more. 23:03.000 --> 23:16.000 Merit based I guess but you know there are a lot of other roles where you might be able to have some kind of you know more less less quantizably measurable. 23:16.000 --> 23:22.000 You know ways of you know getting rewarded for things so I'm I'm you know I don't we don't have the answer to that yet. 23:36.000 --> 23:41.000 I mean I think that oh sorry I'm sorry. 23:41.000 --> 23:49.000 I mentioned these are these are deemed as a success story what are the sort of criteria for a success in the way that I'm imagining that. 23:49.000 --> 24:02.000 Or evaluating that and I would say replication is you know the idea that has been successfully replicated to I don't have something like 600 municipalities globally now or something like that is a is a huge. 24:02.000 --> 24:28.000 You know when you know it's utility has been proved out I think and then the fact that it's it's process has been mostly sustainable I mean it's had some hiccups along the way but it's a it has you know developed funding strategy and also a governance model that seem to not be you know the hackable too much. 24:28.000 --> 24:31.000 Is that it one more. 24:41.000 --> 24:53.000 Well I think I mean the the idea is to I mean we've we've we've been working with municipalities to try and build these types of organizations just without any kind of official policy framing of that but I think that there is a. 24:53.000 --> 25:05.000 There's an accelerant to the whole ecosystem you know like I said these parts that are rising in like the Ospo and the sovereign tech agency and the e-dick those are all.