WEBVTT 00:00.000 --> 00:13.400 So, I'm presenting today on behalf of US that's a community driven project that I'm 00:13.400 --> 00:18.560 leading it's around for a year and I want to use the occasion here at first time to 00:18.560 --> 00:27.080 a little bit summarize where we are, what it is about, and what are the next steps. 00:27.080 --> 00:30.760 So, why are we doing US? 00:30.760 --> 00:35.940 So last year was emotionally very difficult for some of us it was a year of demigration from 00:35.940 --> 00:39.160 Windows 10 to Windows 11 for many of us. 00:39.160 --> 00:46.600 I also was immigrated against my wish and there are other questions of course that are 00:46.600 --> 00:49.080 relevant in this area. 00:49.080 --> 00:53.960 Some hardware didn't allow for such windows migration because they were end of life, they were 00:53.960 --> 00:56.400 not supported by Windows 11. 00:56.400 --> 01:01.120 And then in my previous rule I was working for data protection authority that also had 01:01.120 --> 01:08.880 a case with Microsoft and then using products for Microsoft at the same time feel a bit awkward. 01:08.880 --> 01:12.960 So it's also a matter of having control of your working environment and there are some 01:12.960 --> 01:15.160 areas that are more important than others. 01:15.160 --> 01:21.160 I would say this is the case for competition authorities, for data protection authorities, 01:21.160 --> 01:22.400 for courts. 01:22.400 --> 01:27.880 But there are also the military, the embassies, the critical infrastructure at large, where 01:27.880 --> 01:34.680 possibly with open source you would have more control and more resilience to carry out 01:34.680 --> 01:35.680 your work. 01:35.680 --> 01:40.520 And these are all the motives where I think there is a good case to use open source also 01:40.520 --> 01:44.560 as a driver for our digital workplace. 01:44.560 --> 01:50.040 Of course we see that there is rarely the case today and there are many reasons I suspect 01:50.080 --> 01:56.560 why we do not all benefit from Linux on our corporate devices. 01:56.560 --> 02:01.640 Immigration is, this can be some difficult to organize, you basically need for some time 02:01.640 --> 02:07.960 two teams at the same time, a Windows team, for the ongoing operations and then a Linux 02:07.960 --> 02:11.040 team to prepare the new set up. 02:11.040 --> 02:17.680 Immigration can be very expensive because the cost saving benefits will only emerge later 02:17.720 --> 02:24.320 and if many people are doing it at scale, then you may not have the know-how in-house 02:24.320 --> 02:29.020 and the ecosystem and comparison to the Windows world is quite small, so there is also 02:29.020 --> 02:35.960 shortage of expertise and in the absence of world models, but we all know where it worked 02:35.960 --> 02:41.160 out perfectly, there are many unknowns yet and there is of course a risk to fail and we 02:41.160 --> 02:47.560 have seen possibly in the Linux migration world more big cases that failed than that 02:47.600 --> 02:50.760 where a success which also makes us not so attractive. 02:50.760 --> 02:54.960 And of course there are the end users who are going to want to have any change at all 02:54.960 --> 03:01.480 and please don't bother them and this is of course also something that a migration 03:01.480 --> 03:07.960 would need to face and not all of these problems can be solved with something like 03:07.960 --> 03:13.560 a US project or a risk-release or some are more inherent to an organization and how 03:13.640 --> 03:18.400 human beings are in general, but I think some elements can be improved if we would collaborate 03:18.400 --> 03:20.400 to explore some energies. 03:20.400 --> 03:26.440 And I see that there are two areas for these energies, the first one is that if we have 03:26.440 --> 03:33.080 a common approach on how to organize such a Linux migration, we would benefit from sharing 03:33.080 --> 03:38.360 of knowledge, from sharing of the expertise that we have around and if we use the 03:38.360 --> 03:44.200 same building blocks that we know already from the cloud technology, from servers where 03:44.200 --> 03:51.560 we have been using Linux for years, we also explore these energies from these investments 03:51.560 --> 03:57.200 in the cloud area and also from the cloud engineers that many public sector organizations 03:57.200 --> 04:00.040 may already have in-house. 04:00.040 --> 04:02.800 So what is the US? 04:02.800 --> 04:08.320 It's not a project of the opinion, but it should be and some people are working active 04:08.560 --> 04:13.560 on changing that, we will see how it goes, it is not a new Linux distribution because 04:13.560 --> 04:18.400 indeed there are already many Linux distributions, but it is also not for private home users, 04:18.400 --> 04:21.920 they can perfectly use whatever Linux distribution they are already around, they don't need 04:21.920 --> 04:24.920 a managed system for themselves. 04:24.920 --> 04:30.560 And it's not a project that is producing original code, of course we propose upstream 04:30.560 --> 04:35.760 changes that would benefit our use case, but we reuse mostly existing codex explained 04:35.760 --> 04:37.800 people how they can use it together. 04:37.800 --> 04:42.440 So it's a proof of concept on the deployment of a fedora-based Linux OS with a KDE plus 04:42.440 --> 04:48.480 my environment and bootable container technology that we know from the server world and 04:48.480 --> 04:50.760 a typical public sector organization. 04:50.760 --> 04:55.440 And it's not limited to the public sector on principle, many companies or NGOs have 04:55.440 --> 04:59.200 merry-similar requirements and could also benefit from such a setup. 04:59.200 --> 05:04.560 So US is a community of people who want to bring Linux to the desktop and the corporate 05:04.560 --> 05:08.680 environment and it's another space with a website where we want to gather all the information 05:08.680 --> 05:11.320 necessary to do that. 05:11.320 --> 05:12.800 So how do we want to do this? 05:12.800 --> 05:17.440 So I have done some research where Linux is used in the public sector on the desktop 05:17.440 --> 05:21.360 and you see there are already quite some areas and it's a little bit growing. 05:21.360 --> 05:27.160 So last week I added this one here in Prague, let's see archiving, the National Archiving 05:27.160 --> 05:34.240 Center where they use fedora, bootable with bootable containers, very similar set up for 05:34.240 --> 05:40.080 there for the rack stations and we see that in Spain especially in the education sector 05:40.080 --> 05:45.280 there's quite some use of Linux but they're all work very isolated and all the setups 05:45.280 --> 05:51.480 are different and therefore we have every struggle again from scratch and the idea is to say 05:51.480 --> 05:56.160 yeah to make it at least easier for those that would like to get on the map that would 05:56.160 --> 05:59.680 like to adopt Linux, let's do a similar fashion. 05:59.680 --> 06:04.600 The proposition here and these are the only two technical slides I think I have is to start 06:04.600 --> 06:10.440 with the container technology where we have a base layer that could be fedora and central 06:10.440 --> 06:16.200 S because these luring mechanisms at the moment best supported by fedora possibly with 06:16.200 --> 06:21.240 KDE then we have maybe a common U.S. layer with things that every organization certainly 06:21.240 --> 06:28.600 needs but then of course every organization could add another Docker layer with their specifics 06:28.640 --> 06:35.520 they are drivers for their printing system, their VPN whatsoever then we package this with 06:35.520 --> 06:41.360 a tool that is core blue build it's open source we can do this in our existing get continuous 06:41.360 --> 06:49.320 integration system and then with his bootable C technology we can transfer this on a desktop 06:49.320 --> 06:56.640 without any hypervisor and with the form and server that is normally used to manage servers 06:56.680 --> 07:03.040 we can actually also manage these desktop devices so that they will get provisioned fully 07:03.040 --> 07:08.680 automatically and also keep up to date and then of course for the users we would have something 07:08.680 --> 07:17.640 like free e-pa where users then can authenticate using the central directory of users to the 07:17.640 --> 07:25.760 laptop and it also supports somewhat active directory to have intermediate hybrid setup and 07:25.800 --> 07:31.040 this I think what we will need to to face we will have first organizations with only windows 07:31.560 --> 07:35.240 that use X of directory then in the next step we want to add 07:35.920 --> 07:41.800 some working groups that work then with a Linux desktop that is authenticated 07:42.800 --> 07:50.280 and authenticated via free e-pa possibly still embedded and part of the X of directory and eventually we can take this array and 07:50.280 --> 07:53.600 we end up with an organization fully on open source 07:53.600 --> 08:04.600 so now let's boot up the system so where are we so far we have raised awareness on these matters 08:04.600 --> 08:11.600 with users through our personal network through social media and we have reached out and had discussions 08:11.600 --> 08:17.600 with the UPN data protection by the UPN Commission's digit department, connect department, 08:17.600 --> 08:24.080 HRDS department with a joint research center that also this already using Linux, 08:24.080 --> 08:31.600 e-rogers, duropole, the U-parlement courts, we have also spoken to international organizations such as the 08:31.600 --> 08:38.600 International Committee of the Red Cross, the United Nations, the International Criminal Court and the International Court of Justice 08:38.600 --> 08:48.600 and with a number of e-member states with denom, with senders, well-known and these four are here at Lissie Coldstein with the Ospo of the Netherlands 08:49.600 --> 08:57.600 with the people from the Ospo of Luxembourg and we had a demo where we showed that during the talk 08:57.600 --> 09:07.600 we can provision hands off a machine with the US and we will do again a demo most likely early in February in Madrid 09:07.600 --> 09:12.600 so I hope that we will also have some discussions with the Spanish 09:12.600 --> 09:19.600 and normally I would expect that this year we will also have a first feasibility study with the pilot and a public sector 09:19.600 --> 09:25.600 organization with 150 seats possibly but we can only talk about it more after it happens 09:25.600 --> 09:34.600 so what has been the feedback so far? What do people want? So when I talk to them they do not want to have anything innovative 09:35.600 --> 09:42.600 they want to have something that is already around for 20-30 years at best that works really well with thousands of seats 09:42.600 --> 09:47.600 and they don't want to be the first one to use it, they want to be the last one using it 09:47.600 --> 09:55.600 they want to have enterprise support, they want to have enterprise tools for life cycle management of everything, monitoring and so on 09:55.600 --> 10:04.600 and if we now consider that those that have an urgent need for digital sovereignty will be moved first 10:04.600 --> 10:12.600 then this will be called, it will be the military, it will be secret services and they require of course a compliance with many different rules 10:12.600 --> 10:18.600 so if you have the first one for a product to go through this this will be very hard 10:18.600 --> 10:27.600 so ideally we rely on tools that have been already certified before in various contexts to make this process easier 10:27.600 --> 10:37.600 then of course organizations often have already some sort of Linux use so they may have containers, they may have servers, VMs 10:37.600 --> 10:47.600 or they may have open theirs that also has internal Linux distributions and then the question is do they want to have another one for the desktop or do they want to benefit 10:47.600 --> 10:54.600 from doing the supply chain security under certification only once and use it for all of these use cases 10:54.600 --> 11:02.600 so ideally there is this idea to have consolidation to one or very few Linux distributions in their organization 11:02.600 --> 11:10.600 and then on the server side SAP is very important for many of these organizations and they don't want to change everything at once 11:10.600 --> 11:20.600 so whatever it is would be nice if it has support for SAP and then some of the tools like MATLAB or SPSS or whatever this organization is doing in substance 11:20.600 --> 11:26.600 then it should be of course supporting the sovereignty of the organization should be sustainable 11:26.600 --> 11:35.600 and I talk about this in a moment a little bit more and many public sector organizations really like supporting their local economy 11:35.600 --> 11:42.600 so if they can have a local champion that would be something that they would check out first this was my impression 11:42.600 --> 11:50.600 but then again with the consistency it is difficult because your local economy might not provide everything 11:50.600 --> 12:01.600 and then you will start importing other softwares from other governments and they have maybe the internal different dependencies 12:01.600 --> 12:12.600 and this is why I think if as Europe or even large on a global scale we can agree on a set of digital public goods that we use again and again 12:12.600 --> 12:20.600 we get closer to the goal of having consolidation with the tool and it will be much easier to support and maintain them in my view 12:20.600 --> 12:33.600 so what is the commission saying on the topic of digital sovereignty much has been said in speeches which is a difficult to process 12:33.600 --> 12:46.600 in this precise context part there is actually a project ongoing for cloud purchase system that allows institutions to get sovereignty in the cloud space 12:46.600 --> 12:56.600 so that they have prepared an annex with the sovereignty objectives that gives a few details on how to measure specifically digital sovereignty in the context of cloud 12:56.600 --> 13:02.600 so this framework does not apply for us because we are not doing cloud to redo the desktop operating system 13:02.600 --> 13:09.600 but there are still some elements that I think can give an idea of how these organizations are thinking about sovereignty 13:09.600 --> 13:21.600 highlight already two of the criteria there is this one of one strategic sovereignty it assesses ownership stability amongst many other things 13:21.600 --> 13:30.600 and then there is another criterion number five which is about supply chain security will also point out that the graphical origin plays role 13:30.600 --> 13:39.600 there is another talk about this specific document and what it means and how organizations can assess their own products 13:39.600 --> 13:46.600 offered by Emil Brook tomorrow from open so I all invite you to check that out I try to be there as well 13:46.600 --> 13:55.600 so what does it now mean for our question for the distribution that U.S. showed a build upon 13:55.600 --> 14:07.600 I have written a blog post was a number of points, tables and numbers to also get some feedback from the community and the community then normally answers 14:07.600 --> 14:19.600 but fedora is too much of the U.S. how come that we do not use outpind Linux and yeah I think if we do not do it was nix so as we are doing something one 14:19.600 --> 14:26.600 so I this is like 80% of the feedback that I have and I want to remind everything also here through this talk again 14:26.600 --> 14:36.600 it is very different if you want a corporate managed Linux fleet and an organization that is a court or if you are in your hacker space 14:36.600 --> 14:42.600 and you have also your own Linux computer so you have much more freedom in the second case 14:42.600 --> 14:55.600 so when we now look at those Linux distributions that I have been battle tested where there is a support network of at least one big player 14:55.600 --> 15:05.600 they are not so many Linux distributions so there is Debian which is not profit in the U.S. but also in France managed by volunteers 15:05.600 --> 15:13.600 but many organizations and companies work with stebian and can offer support there is Ubuntu which is a company in the UK 15:13.600 --> 15:26.600 200 million dollars apparently revenue there is open user which is more revenue more staff with a headquarter in Germany and it is registered in Luxembourg there is redhead 15:26.600 --> 15:41.600 which is again factor I don't know 10 more revenue more staff in the U.S. but also in the U. Interestingly there is the Armalinox Foundation which is a foundation in the U.S. 15:41.600 --> 15:47.600 backed by some big players and then there is the rocky Linux Foundation that is something very similar 15:47.600 --> 15:58.600 so what are the takeaways here? Fedora is factor 10 bigger and most of the developers in the U apparently work with this with Fedora 15:58.600 --> 16:05.600 and that also mirrors here in the room when I asked who is using open user Fedora there were a lot of hands on Fedora 16:05.600 --> 16:10.600 many people overestimate the market share that Ubuntu has on the Linux market 16:11.600 --> 16:18.600 I think Ubuntu really made Linux more known but it is a lot of private users it is not so much the corporate users 16:18.600 --> 16:29.600 and also the cloud business to drive the adoption of the Linux distributions and it seems that open user because it is certified for SAP products 16:29.600 --> 16:38.600 is used a lot in the cloud area as well and Ubuntu may have less paying users even though it is very well known brand 16:38.600 --> 16:49.600 so now the problem is that all these distributions are there good enough for us and this is why I think it was so nice that we had the previous two talks 16:49.600 --> 17:00.600 where we heard first about the public product organizations so we could now ask is anyone of these distributions that we have seen 17:00.600 --> 17:07.600 close to that or there was this idea of the digital public goods in the previous talk 17:07.600 --> 17:13.600 and I can tell you that the user Linux enterprise has been certified as a digital public good 17:13.600 --> 17:17.600 and also Fedora is a digital public good 17:17.600 --> 17:28.600 but when we look at the criteria that the European Commission put out for the cloud where they look for long term ownership stability 17:28.600 --> 17:38.600 I wonder how these mechanisms the digital public goods and the initiatives around it, the infrastructure around it are reflecting that well 17:38.600 --> 17:44.600 so I think that could be a gap in the certification mechanism because if you want to have very good ownership stability 17:44.600 --> 17:52.600 I think the first talk that we are the second talk in the track today on the digital public good 17:52.600 --> 18:01.600 organizations is pointing to the right direction we want to have some sort of a nonprofit foundation that is financially stable 18:01.600 --> 18:07.600 and when you would ask governments in the you they would prefer that they are of course based in the you 18:07.600 --> 18:14.600 or alternatively for governments they would own the product that would of course also give them a lot of ownership stability 18:14.600 --> 18:21.600 and when we now look through the distributions that I have had on the previous slide we see that red head as well as based 18:21.600 --> 18:27.600 so not good Fedora has no separate legal entity not good 18:27.600 --> 18:34.600 Suze enterprise is the u-base so that would tick the box but open Suze has no separate legal entity 18:34.600 --> 18:42.600 there are some people from Suze that point to the geico foundation but I do not understand how sustainable the geico foundation 18:42.600 --> 18:46.600 is at the moment and who is backing it apart of Suze itself 18:46.600 --> 18:52.600 there is the al-malino foundation that seems to work very well but it is u-s based and who knows 18:52.600 --> 19:01.600 loud red head how sustainable that can be there is Ubuntu in the UK so that is a company not a foundation 19:01.600 --> 19:08.600 d-b-n has several legal entities a foundation also in France but it does not have the innovation 19:08.600 --> 19:14.600 the technical features that we would need for our model of u-s that would help organizations 19:14.600 --> 19:19.600 to do the customizations on their end and share most part of the of the base and build it in their 19:19.600 --> 19:28.600 good art continuous integration service so this is why at the moment d-b-n is not ready to take the role of Fedora 19:28.600 --> 19:35.600 so what we would actually need is a better governance for these Linux distributions something 19:35.600 --> 19:42.600 that is really matching the requirements of the public sector something that is matching the public mandate 19:42.600 --> 19:48.600 that the sector has so what we do not want is that a company can go on with 19:49.600 --> 19:55.600 with institution rate until entire European public sector is using it and then try to 19:55.600 --> 20:01.600 change their licensing, stop the development and we have seen some of these products going 20:01.600 --> 20:09.600 that pass in the last year and of course it would be unclavered to now bet on the wrong underlying 20:09.600 --> 20:16.600 distribution that would then be prone to go a similar pass so easily d-b-n would now come up 20:16.600 --> 20:23.600 and also of a boot C support then it would be on the plan for u-s or almalinux could 20:23.600 --> 20:29.600 also incorporate a foundation outside of the u-s could be in the European Union the 20:29.600 --> 20:36.600 zoo foundation could also improve their at least their marketing and explain better what they 20:36.600 --> 20:42.600 can do what they are about that would all help a lot and now I'm coming to my last slide 20:43.600 --> 20:49.600 I was very much inspired by the Davos speech of the Prime Minister of Canada 20:49.600 --> 20:58.600 that we heard some ten days ago Marcani who was addressing the situation of sovereignty 20:58.600 --> 21:04.600 as well and he said risk management comes at a price but that coast of strategic autonomy 21:04.600 --> 21:09.600 of sovereignty can also be shared and I think this is not only true for the economies 21:09.600 --> 21:16.600 it's also true for the operating system on the in the public sector collective investments 21:16.600 --> 21:21.600 in resilience are cheaper than everyone building their own fortresses shared standard 21:21.600 --> 21:26.600 reduced recommendations complementaries are positive sum and the question for middle powers 21:26.600 --> 21:32.600 like Canada is not whether to adapt to the new reality we must the question is whether 21:32.600 --> 21:37.600 we adapt by simply building higher walls or whether we can do something more ambitious 21:38.600 --> 21:44.600 something together and this is why also for us it is of course a goal to not bring 21:44.600 --> 21:51.600 possibly only two three u countries behind the common goal but at best a much broader community 21:51.600 --> 21:57.600 and therefore maybe the next u-s and inside was a little bit too narrow it could have been 21:57.600 --> 22:03.600 something more ambitious because I would be happy when we can also share the goals that 22:03.600 --> 22:10.600 you with u-s with other middle powers like Canada like Brazil like India like Mexico or 22:10.600 --> 22:16.600 Australia and then I wonder if we want to reach out to these communities and say let's 22:16.600 --> 22:22.600 share the the common investment and build the strategic autonomy together what would 22:22.600 --> 22:30.600 be their take on the Linux distribution would there also be a fan of next OS would they 22:30.600 --> 22:37.600 do they know already open to them can we can we forge an alliance on on one of these and this 22:37.600 --> 22:43.600 is why I feel for the u-s project we will certainly try to aim for Linux distribution 22:43.600 --> 22:49.600 that does not only help us with the consensus in a few member states possibly in the u but 22:49.600 --> 22:58.600 at best also on a much more global scale and here by I want to close my talk I'm happy 22:58.600 --> 23:03.600 to have any questions you may have I think we have five minutes left three okay 23:03.600 --> 23:08.600 two and a half then it has to be very very good questions on the point please 23:29.600 --> 23:35.600 so what would it say the question is now how can how much how closely are to mitigate the risks 23:35.600 --> 23:43.600 so there is someone who is estimating creating a foundation and building the distribution 23:43.600 --> 23:50.600 independently from a company would possibly take three for people you do it over three years 23:50.600 --> 23:59.600 for the beginning it's five million euros so and more users you have the more clearer 23:59.600 --> 24:06.600 is the use case I'm in talks with open user and with al Malinox to tell them about the difficulties 24:06.600 --> 24:11.600 that I see and if they can improve their their stands we will see how it goes some have 24:11.600 --> 24:18.600 already yeah shown some interest I would say and I'm quite quite optimistic that we can get a 24:18.600 --> 24:25.600 step further this year one more question 24:25.600 --> 24:40.600 so the question is if I had on my radar to just use the enterprise linux for us 24:40.600 --> 24:55.600 so the question is if I had on my radar to just use the enterprise linux for us 24:55.600 --> 25:08.600 I'm sorry I'm unable to answer that question any other question 25:08.600 --> 25:23.600 okay that's one suggestion 25:23.600 --> 25:29.600 the question is why not reach out to private companies and help build a building 25:30.600 --> 25:39.600 of course totally true so my vision is that there's a foundation that has a lot of companies 25:39.600 --> 25:45.600 big companies that are supporting it and it could be through the defense sector it could be the banking sector 25:45.600 --> 25:52.600 it would be those private sectors that also work in critical infrastructure who have an interest to have an alternative 25:53.600 --> 26:01.600 who would then also back such a foundation and yeah we're trying also to find context and this 26:01.600 --> 26:08.600 cases but it is not very straightforward but yet on our radar so I think we at the end of the 26:08.600 --> 26:18.600 talk thank you so much 26:18.600 --> 26:24.600 and please set me an email for any follow-up