 Okay so I think we can then get started. Sorry for the delay. So good afternoon, good evening, good morning, a warm greeting to all of you and to my students and former students around the world included. I'm Simonetta Vitzoso, I'm a research and teacher at Trent University up in the Italian alps. So I sleep that night, a sleepless night, not with Stanley. I'm absolutely honoured and thrilled to be sharing this beautiful debate, interoperability and the new digital market sector. It is organised by the open forum academy and I'm a member of this academy while the ingredients are civil society, law, tech, economics, relatively gender balanced. This is the recipe of a panel and this is almost a perfect panel. Let me introduce the wonderful panadist to you. Well, first of all we have somebody representing the civil society, I think it's a very important present. So Katalina Szymieliewicz, she's the head of the Panopticum Foundation and the Panopticum Foundation is a non-governmental organisation based in Poland where among other things they imagine and test different strategies on how to curb digital surveillance. Then we have two lawyers and the first one is Lisa Lofdor Gormsen and she's the director of the Competition Law Forum and Senior Research Fellow in Competition Law, British Institute of International and Competitive Law. And then I'm seen in the chats that there are some, there are people with issues with audio, I hope everything is fine. Then we have Augustine Reina, he's director of legal economic affairs at the Bureau Europin des Unions de Consulateurs, the BEUC, which is the umbrella group for 45 independent consumers organisation from 32 countries. Then we have two technologists, Ian Brault, who is a computer scientist specialised in privacy, security and interoperability. He has many hats, but he prefers to be called, he calls himself independent researcher. Then Victorio Bertola, head of policy and innovation, open exchange, which is a German company, I found out, with offices also in Paris, Madrid, Turin and US, which focuses on open source solutions for companies. Then last but not least, we have with us an economist, or I should perhaps say the economist, Thomas Valetti, professor at Imperial College of London and former chief economist at the European Commission, DG Competition. So how are we going to stage this virtual debate? There will be a first round of questions from my side, and then we will open the floor to questions from you and the audience. And I will warn you that I will be shamelessly self-preference my students if they post questions, so if they recognise names, because this webinar actually replaces my lecture. So very briefly, what is the virtual panel about? Well, you'd be forgiven for having a feeling of deja vu, because we have been here before, not virtually, but physically in Brussels and exactly 10 years ago, I was there by the way. At a speech delivered to open forum Europe in 2010, Nini Cruz, who was then a competition commissioner in the long aftermath of the EU Microsoft case, said that complex anti-trust investigations followed by core procedures were perhaps not the only way to increase interoperability. She said that she intended to seriously explore all options to ensure that significant marketplace cannot choose to deny interoperability with their product, and she also said that it was certainly possible that the result of this exploration could be a legislative proposal, 2010, by a way of which to achieve interoperability, these words, on an ex-anti-basis, that is without having to conduct proceedings under competition. So 10 years later, where are we? Well, a tsunami of reports, studies, there's been a denouncement of the Digital Services Act, which should comprise asymmetric ex-anti-regulation of very large art platforms, acting as gatekeepers, but also new competition tool, which is a new market investigation regime, regime, and then quite a lot of proposals, quite concrete, is an amendment to German Competition Section 19A of the German Competition Act, which should be able to prohibit undertakings, so undertakings with permanent significance for competition across markets, they should be prohibited from hindering interoperability of products and services, and the possibility of data, so quite a lot of things going on. Okay, so in a nutshell, interoperability could be part of the package of the pro-competition interventions that they are designed to tackle the sources of tremendous market power held by digital platforms, by overcoming barriers to energy and expansion. Briefly, that was my introduction. My first question to Ian, I see you there, switch out your cover as well, so Ian, from a technical perspective, is interoperability a pipe dream of some competition economists, lawyers, and civil society groups, or is it practical, is it doable? Ian, are you there? Sorry, just jumping in from the off, I'm Sivan from Open Forum Europe. Ian is having some issues downloading Firefox right now. Okay, so never mind, I'll get back to him later, so big problem, okay, no problem. So what have we learned so far, because I mean one of the aspects of it would be standard setting, okay, open standards. So what have we learned so far about achieving solutions to interoperability problems through standard setting specifically in the ICT sector, because that was one of the ideas, nothing much was done, but the idea was to support the battles at European level. So what about the effectiveness of non-legislative measures, such as modern licenses for interoperability information guidelines, etc., as was suggested in 2013, staff working document analysis of measures that lead significant marketplace to license interoperability information. So, Victoria, thank you, the floor is yours. Oh, thank you. Well, the standard setting is an important part of the problem, but it is just a part of the problem. So, of course, if you want to enable interoperability, you need to have open standards, they need to be freely accessible, free of IPR restrictions, then we can discuss what it means, if it's standard or whatever. But you don't create interoperability just by making the standard or by licensing the standard. I mean, for example, if you take instant messaging, which is there, I mean, possibly the easiest case of application where we do not have interoperability and we would like to have it. For instant messaging, there is a lack of technical standards. There are plenty of them, at least a couple of them, which are open and good work immediately. But the problem is that the dominant platforms do not use these standards. Or if they use them, they use them, but they deploy them in closed ways. So the standard, the protocol of communication is open, but they refuse to exchange messages with anyone except from their own app and their own service. So we should stop seeing interoperability as an IPR issue. It is a competition issue, but it's a much broader issue. It's also an issue about consumer protection and indianity, even impacts the fundamental lives, not just the economy. So I think that both sides are important. So I mean, I say that the document from 2013 is showing at its age, because yeah, I mean, it's just, I think we can discuss about the positions. But I mean, the document says about, I mean, that we should force that cultural licensing is changing the spirit of open innovation, but dominant platforms don't really care about community spirit. They just care about their own interests. And also, I mean, the other point I think that I think it's worth waiting making very quickly is that open standard setting processes are customarily captured by the dominant companies, because exactly because they are open, which is good. But some companies can just send 100 people and just capture the process. So open doors are not enough. I mean, sometimes there are other industry sectors, so you know, for example, the European industry that show up, but cannot really get a voice because they're just a few people. I mean, even others, they call this non-technical masking. They can enter the room, but in the end, you do the culture to the fact that there are insiders, and we've been working on this for 20 years, they get ignored. Even the ITF, which is, I mean, basically, in a way, it's the good model, the one that has made the Intel grow. But the ITF also has a political agenda of its own, about the encryption, about the disempowering government or the internet. And I mean, whatever you think about it, I mean, you see this need that you have to reconcile these standard processes. We share public policy objectives that are really multi-stakeholder for multi-stakeholder settings, not just by the technical community or by the big, dominant platforms. So paradoxically, the last message I want to give, sometimes even the internet SDOs are not very multi-stakeholder. So that's also something we need to address. And perhaps just so the question then for it to be interoperability, if it's part of the digital markets, it would be then what type of provision and obligations and how would it be framed also in legal terms? And that would be my question to Augustine. So from the legal perspective, how would a truly effective interoperability requirement for digital gatekeepers within the digital market site possibly look like? So and the other thing I wanted to perhaps hear your thoughts about is the fact that, of course, competition authorities have experience with interoperabilities from some extent. There have been remedies in major cases. There have been a couple of cases also at national competition authorities. What have we learned from it? So thank you, Augustine, if you're in yours. Thank you very much, Simoneta. And thank you to the colleagues of Open Foreign Europe for the invitation. It's great to be discussing this as a kind of a break from the news across the Atlantic. Actually, it's true that we have been discussing interoperability issues in competition cases, particularly from the point of view of what we call vertical interoperability, while here we're more in the field of horizontally interoperability. But I think what is perhaps more interesting is to discuss what are the different regulatory models that we already have across sectors. And actually, I like what Victoria said at the beginning. It's not all about standards. It's about creating actually the obligations that will then develop such standards. And I think that in this discussion, in the context of the digital services act or the digital markets act or whatever we want to call it, are basically kind of three big models that we can explore. So the first model, which is based on a case-by-case assessment, for example, in the context of the new competition tool, which has some benefits because you don't need to define in advance which are the markets or the product categories to which you would apply interoperability. But just simply through a case-by-case assessment, you identify that are structural market problems. You can impose it as a remedy. But we have also more developed kind of schemes that relates to interoperability slash data access. Like, for example, in the context of electronic communication code or the payment services directive regarding open banking. And I think these are very interesting models and we can learn a lot from them. If you think of the electronic communication code, although these have not been yet put in practice, there is a possibility for an arrays to impose interoperability to OTTs over the top communication services. And what is interesting is that it can be implemented either as a general term for all the OTTs, but only also linked to S&P tests. So actually have the two possibilities. And the system in how it's going to be implemented is quite detailed regarding the different committees and the standardization process that needs to be involved. And then we have the payment services directive model, which basically PSD2 imposes an obligation to give access to the API. And then the whole process of implementation is left to the European Banking Authority. And what is interesting through our experience, because actually Berg is also part of that discussion, we also now start seeing the tensions between the banks, because for example, ideally you will have one dedicated interface, which is identical for all banks. So every single payment service provider can access that. But of course, the bank what we're trying to push is for their own dedicated interface, which means that the payment service providers will have to adapt to every single model, which technically is very difficult and extremely burdensome. And at the end, just place barriers to entry. So I think that we can learn actually from these experiences. We can discuss about having a regulatory mechanism or a legal obligation to grant interoperability, but we need to think very carefully about how is the model we're going to use for implemented with all the risks that are in tails. So maybe I leave it here and we can discuss more on the details afterwards. Thank you. Thank you so much, Agustin. And perhaps if Ian is back with us, I would perhaps link to what Agustin has been just saying about some issues with the FPSD2 and perhaps a more virtual example that respect would be open banking. And generally my question to you again is a sort of a technical perspective is interoperability a fight dream of some competition economist lawyers and city society groups, or is it really doable? I think open banking and PSD2, which underlies it is a really good example because Agustin was just explaining some of the difficulties that can arise if you set very high level obligations in law, but then don't have the implementation mechanisms to really make things work. So the UK has built on top of PSD2, created this program called Open Banking, used various other regulatory powers that exist in the UK financial sector, and really got our, I think it's nine largest banks to work together really closely to define common standards and APIs and processes. And even to the extent there's an organization called the Open Banking Implementation Entity, which actively monitors how well this is working and even on its website it publishes statistics every month about how many API calls have failed, for example, across the whole system. And I think that's a really good model for us to think about if we were thinking about making, you know, whether it's instant messaging, whether it's social media, which would be two obvious starting points, really work together in practice. I promise all the viewers, the problems we've had was not staged. It's a reminder. I had never used this platform before today. I had incorrectly assumed that because I use the second most commonly used web browser, Safari, that this platform would work with that. No, I've had to go and install Firefox while you were all talking to get it to work. I mean, we have, you know, just to go back to your pipe dream point, we take it for granted in many ways with many technologies that interoperability works really well. The internet is based around interoperability, you know, in particular, a standard called the internet protocol. That's the whole way the internet works, you know, which was standardized going back, you know, at least to 1981 to one of the most important recent standards. But going back further than that, the telephone networks all work together in this way because there are common technical standards and they've been required by governments in Europe and elsewhere to interconnect their networks. Going back to the 19th century telegram, the telegram system was made interoperable, was interconnected by what is now the ITU, one of the, you know, one of the world's first big technical standards bodies. So interoperability as a technical matter, and I'm a computer scientist, absolutely can work and absolutely can support functionality such as end-to-end encryption. But you do need the standard processes, which Victoria was talking about a bit as well, to work to do that. And I think open banking and this experience is a reminder as well, we would probably need something similar under the Digital Markets Act where the people that were creating the products and services could work together, overseen in the public interest. So not a small cartel of the biggest companies frustrating their competitors, you know, making a load of different standards, slightly incompatible standards, not paying quite enough attention to the user experience and so on. Those are important details to get right, but absolutely, I think it can be done. Thank you so much Ian for these inquiries. Perhaps it should also be reminded that the financing model of the open bank implementation entity paid by the banks themselves. So that's very handy of course, that's a matter of resources there. Thank you so much and then we stay in the UK and a little bit, Lisa is she there because I can't see the video. Well, I post the question and then we'll see. So in the case of Facebook, the CMA in the final report on digital platforms and online advertising discuss a particular cross posting and content interoperability. If you made it until NXW, the CMA also developed a very promising framework, I think, to be applied to interoperability more broadly, also thinking of impact on innovation incentive, etc. And so Lisa, if you are there, what's your take on it? Or more generally, when do you think that the case for mandated interoperability is greater in order to promote innovation and competition? But I can't see Lisa. Is there anybody from the organisation able to step in and tell me what happened? Well, let's imagine she will appear soon again. Sorry, sorry. I was just doing some troubleshooting. So I only had the back part of your statement, but I assume it's about Lisa. They were still trying to figure out the browser issues. It seems to be related. Okay, never mind. No problem. We perhaps then we move to to Maso. So the economist on the panel and what are what are the economists that we have it here. So from an economics perspective, of course, there are quite a lot of arguments which speak in favour of interoperability and a few ones also related to innovation incentives, homogeneous products that are against them. Then the question would be, is it now the framework of this Xante approach taking into consideration these platforms with those enormous powers they have a little bit different than our new normal discussions about interoperability. So we have learned a lot from these reports in the last two years or something. And the other thing is, okay, the argument would be, well, we did it for telecom. Okay, but then the argument from the other side would be, no, it's different, there are other markets, other issues there. So, what would be your take on that? That's right. And hello to everybody. So from an economics point of view, interoperability is related to rather strong economic concepts which are related to externalities. So we want to make sure that externalities are enjoyed because they bring some efficiencies and get different kind of externalities. You might search because you can use more data to refine the quality of the searches or you can see your friends. But let's assume these are all positive externalities. So that's great. This is bringing efficiency, but you also want to make sure that those efficiencies are actually going to be given to users and that they don't stay with the firms alone and you need competition for that. So the problem that we currently experiencing is that we do have externalities, but they're not enjoyed industry-wide. They are enjoyed by particular players within their own ecosystems. So you want to make sure that externalities are not accrued into a single player, but they can benefit other players because those players will bring competition and finally users will benefit. So that's fine. There are trade-offs, of course. Also these trade-offs are well understood, both in theory and in practice. And typically they go under the simple heading of, well, there is competition in the market and competition in the market is facilitated by interoperability, but there is also competition for the market. So let's make sure we don't destroy this dynamic phase of competition for the market. But this is not new. This is something we knew, for instance, as you alluded to in telecommunications. And we did with it in telecommunications. Those who remember Exanta, which we still have a little bit, Exanta regulation in telecommunications, the markets recommended for some Exanta regulation had to satisfy the so-called three criteria. And the three criteria tests were related to the fact that there should be high barriers to enter. And so think about search. There are high barriers to enter if you want to come into search engine. The market structure should not be tending towards effective competition. And again, if you think of search, there's been dominance for the past 12, 15, 15 years. And so the indicators are all there, that this is not just a transitory phase. And also the third test of the three criteria tests was that exposed competition policy would not be good enough to tackle the problems that I think we have enough experience so far with all the Google cases, the time it takes that competition policy to say the least is not a very good instrument to deal with that. So if I have to think about those three criteria tests that they would apply almost one for one in the cases of many of the other platforms that we are talking about, we also learned that the regulations that were in out to the asymmetric, they wouldn't apply to other platforms, but there was clearly a competition problem because the ultimate objective of interoperability and interconnection who the intercommunication was to, you know, invigorate or facilitate the so-called ladder of investment. So making sure that ultimately there would be some alternative operators that could challenge incumbents. And that's what we want. And this is going to come with differentiation. It's going to come with alternative products, increasing choice, increasing investment and so forth. So there's lots of similarities and also differences. The differences is that in telecoms at least interoperability and access regulation has been overpriced mainly. So we had prices and we had to find some numerical value for those prices. We also learned, however, and there is an analogy that precisely because access was regulated and hence profits or large profits could not be made any longer on access, incumbents started what we called back then sabotage. So doing non-price discrimination, making sure that access to the to the rival would be of good enough quality, not fixing faulty lines or being slow in responding. And so we had to deal a lot with non-price regulation too. And this is going to be incredibly important in many of these markets because there are no prices, especially when it comes to self-preferencing. And this means a couple of things to think that we really need to get our hands dirty to find solutions. Because if I think of functional separation of British telecom 15 years ago, it came with 236 undertaking. So commitment of BT who had to say exactly how open reach would not discriminate with other players that were not integrated in the BT group. In the case of Google, say we will need to have perhaps I'm not a technical expert access to the algorithm, which makes it very difficult to understand exactly what self-preferencing is going to mean. We have to be, there are solutions there, for instance, make sure that a product appears at the top of the list independently from the supplier. So the name, the identity of the supplier is something easy to do. But then there are proxies that even if you don't call it with a certain name, you don't call it Google, but you can imagine there are characteristics that will be associated with any particular solution. And therefore dealing with that is not going to be super easy. The other lesson that's going to be different here is that when we dealt with interoperability and interconnection, interconnections, somehow the moat around the consumers was not as large as we have it now. So essentially we had from the consumer side to make sure that, I don't know, the billing would not be affected, especially if there was dual play, triple play, or the number of portability would be ensured. So a few things that would make the consumer experience easier here instead is going to be much tougher because we learned a lot, but there is so much inertia and behavioral inertia. So how to unstuck these consumers, which are not particularly conducive to more searches. Even today, with very experienced people downloading a different browser, Ian managed, but Lisa didn't. So it's not that easy as you can see for the average person in the market. The thing which is definitely different, and perhaps Augustino and others can comment on it, has to do with privacy, which wasn't crucial in telecommunication. Sure, there is GDPR, but to be also honest, GDPR is still an abstract concept and fines are far too small. There is lots of evidence out there that are repetitive violations of GDPR and then comments don't care because the stakes are so large and they prefer to stay small fines. But the point here is that in the absence of GDPR, which is well understood and well enforced, you want to be careful with interoperability and interconnection because if there are leakages and there are many leakages as we speak, they can just get worse. The final observation, if I may, is that in the digital, we have some structural problems. It's a structure that leads to anti-competitive behavior. And rules and remedies, so remedies are exposed, but also rules exactly we are discussing now, often fail to address that structure. So it's a cat and mouse game that we have seen in the shopping and the Android cases, but I can expect to see the same cat and mouse game also when practical rules, example, will be discussed. And so this will be involving some further delays. Hence, one has to be thinking of alternatives to that. And I think the alternative, if there is a structural problem, it has to be breakups. This is something we don't talk a lot, but breakups to me must be on the table for two reasons. One is a reason of perhaps finding a final solution instead of delaying yet again with technical details. And the other one, and this is where I have, it is agreement with my former boss, commissioner of the stagger that says that this is ultimate, last resort type of thing. I think this is a bit strange because you use all the bargaining power if you're discussing example or expose, but you don't have on the table a realistic option of a breakup. Basically, your hands are not going to be very strong. The case of functional separation, I discussed earlier, which I think that on balance has been good to bring in additional competition in a system like fixed telecommunications in the UK. This functional separation occurred in the shadow of a potential and realistic structural separation that the parliament was actually ready to enact if the PT had not offered some serious commitment and undertaking. So it was in the shadow of that possibility that a good solution came out. But if you don't have that outside option, then also the solution you negotiate is going to be much weaker. Thank you. Thank you, Tomas. Tomas, I mentioned you, would you like to jump in on the issue of the GDPR, and perhaps also Ian, before we move to the other questions. Yeah, absolutely. And thank you, Tomas, for raising that point. Actually, this is not really an issue because we are discussing exactly, we were discussing exactly the same topic in the context of open banking. There had been very clear guidelines, for example, from the European Data Protection Board about how to do it and also the different methods in which you can reduce actually the risk to share data at all. We're talking about accessing to the consumer interface, but there are different models in which this could be done. And we have quite actually, I would say, sufficient experience in the context of open banking on exploring how the different methods actually work. There is one called Redirection Authentication Method, which is basically you through your bank account and your home banking, you can access or give access to a third party, to the credentials to provide a payment service. And actually, this proves that it's possible to make it safe and privacy-friendly and compliant. Of course, I think that they will come, those that don't have no interest, these two happen to say, ah, but there is a huge risk of data protection, a huge risk of privacy issues. But based on the experience that we have on open banking, that's not having really an issue so far. Ian, I don't know if you wanted to write something as an expert also, privacy, as a computer scientist specializing in privacy. I've been working for much longer on data protection than I have on competition. First of all, I think the GDPR does include many of the general rules and principles that we need to deal with this issue. It is a general purpose, horizontal instrument and the European Data Protection Board, which is the collection of the 27 member states, national supervisory authorities has already produced some guidance, not very specifically on interoperability as a whole, but it has looked at, you'll be aware, I'm sure, that the EU member states are working together on making their coronavirus contact tracing apps interoperable, so that if someone who's running one of those national apps moves to a different member state, ideally, you know, if they were later diagnosed with COVID, they can signal that through their own home app, but the warnings of exposure will also flow to the people in the foreign jurisdiction, where they have travelled to during the 14 days, roughly, period that those apps gather information for, so that the EDPB has published guidance on precisely that on the data protection implications of interoperability of contact tracing apps, and I think there's quite a bit even, it's a relatively short note, but I think there's quite a bit even in there addressing the issues that come up here. Someone asked in the chat, and it's quite a common question, so I'll very briefly mention it here as well. Isn't it easy to say telecoms was interoperable slash interconnected, but that's a much more homogenous set of products and services, whereas platforms do very diverse things and would be much more technically complicated to make interoperable. Not at the core of some of the functionality of some of those products, and I should say I've had this debate several times in the last few weeks. By interoperability, we're not saying that every search product should have every last feature of Google in this case. We're not saying that every instant messaging project should have every last feature of WhatsApp or Instagram Messenger or Facebook Messenger. We're saying that there are core elements of functionality in some of these different services, whether it's instant messaging, social media, which many different competitors, both the very large dominant platforms and several of their smaller platforms support. And if you look at the Bible, as Simonetta rightly calls it, the competition and markets authority, thousands of pages of analysis of this, you'll see there's a table where they do this for social media. And look, I think at the 10 most popular social media apps in use in the UK and identify a broad range of core functionality that's supported by all of those services. So I think when you have that kind of core functionality and standards which support that core functionality, and the World Wide Web Consortium had developed two standards called activity streams and activity pubs that support almost all of that social media functionality, and indeed over 4 million people use it in Mastodon, which is an open source competitor to Twitter. So yes, saying that we wanted to enable competitors to clone every last feature of Google would be incredibly complex and unjustified. But I think that I don't think anyone's arguing that I think we're saying four, you know, defined services that are widely supported, multiple products offering them, although often because of network effects, one winner has taken all in the market, but the market is tipped already to a significant degree, would be very difficult for competitors to get back in. Okay, thank you so much, Ian. And that's a very nice bridge to the thing I wanted to discuss with Lisa, because it was that I'm very happy to sit back, sorry for all those technical injustices. So the topic would be exactly the idea to make interoperability concrete, and that was done by the CMA in Annex W of the final report that we have been mentioning as well, the Bible. So please, Lisa, you're taking it. And generally, when you think that this type of mandates, interoperability mandates are most promising in terms of competition and innovation. Thank you very much. First of all, are you able to hear me and see me? Thank you very much. I had interoperability issues getting into this webinar. So I'm grateful that you had the patience to wait for me. So thank you very much. You're very late in the hour. So first of all, I wanted to convey that I think that the CMA did a good job. And I agree with some of the conclusions that they have set out in Annex W, which is the one you're asking me about. And let me just very, very quickly talk you through that, Annex, because you may not have read it or you may not remember what was in it. But basically, the CMA, they analyze four different types of intervention. They talk about the data project, they talk about accessing connections, cross posting and content interoperability. So these are the four things. And it recommends only accessing connections and cross posting connections. So it does not really want to, they don't recommend the data transfer project. And they don't recommend the content interoperability. And well, I do think that it's important to acknowledge that the CMA say that interoperability is not a black and white solution to these things. And while interoperability will promote competition, as more competitors are able to asset the market, there are also some intrusive measures. And that's why they don't think that they want to go into content interoperability. They talk about the data market unit that they will establish. And they're basically saying that we will allow that data market unit to look at these, if they think that this form of interoperability is the way forward, then they will allow them to look at it. So that's basically more or less what they are trying to convey in NXW. Whether you ask me another question as well, which is really, well, there's definitely sufficient evidence to mandate interoperability on social media. And this is for two reasons. Firstly, all the different competition reports you've seen all over the world, you've seen the A, triple C, you've seen the report from the Stiegler, you've seen the Furman report, you have seen that there's plenty of reports and studies, and now of course the market study that you referred to from the CMA. They have all arrived at similar findings, which is that in the sense that the strong network effects and the use of data makes this market, the social media market, kind of clone to monopoly. And the question is not if there is a pressing need for intervention, but rather an issue of remedy design. So social media is a segment highly dependent on innovation, and I think therefore mandating prescriptive rules with the idea of promoting competition could be counterproductive if it isn't done in a careful manner. This does not mean that highly intrusive measures should be disregarded, but instead due to consideration, it's required to I think strike a fair balance between the promotion of competition and protecting the incentive to innovate. But I do think that the CMA has done a fairly good job and also the fact that they leave it to the DMU to then further scrutinize these measures for future. Thank you so much for your reflections. If I can say something personal, yes, but like also the fact that they try to make the difference between the functionalities which are very innovative and then others which are not. So there's an innovation approach there, which I think is quite promising. But then of course we should need perhaps to develop or redevelop the innovation and balancing test as was proposed by the community. I would say though that the cross posting that they're suggesting is of course nothing new. Right, I mean the CMA has found that Facebook has degraded this functionality over time. So it's something that has already been there, but Facebook has degraded it and they have done that because they argue that this wouldn't be good for privacy and Facebook has argued that privacy concerns means that the cross posting is not really ideal at the moment. But other platforms such as Instagram have retained this functionality and users are able to post on Facebook and swivel through this. Lisa and then Kateshina, thank you for having waited 50 minutes for your turn. Because I think you're I think the fact that you are here I think is really quite exciting because we have talked so far about competition, even GDPR a little bit and then of course innovation. But so the question here would be about the practical value of interoperability for say normal users. Imagine scenarios that could be positively affected by this change. So not only pro-competition framing, but contribution to solving perhaps even more serious societal problems caused by large platforms like private censorship, monetizing this information, controlling valuable knowledge about humans that could be used also perhaps now during our pandemic. So what's your what's your thoughts on that, Katarina? Thank you Simanetta for full disclosure. I'm also just a lawyer and an expert, not a user representative. My organization is not even an association. We do what we think is best for users. We don't represent their voice. But indeed in this debate we try to push our best understanding of the user value or simply value for humans. I personally hate the framing users and I think we need to abandon it although we've been entangled in this language by large platforms that are our enemy precisely in this battle. So what would be the big promise? The big promise already transpired from what others said so far the federation of social networks or the opening the gates of world gardens promises new functions that are very intuitive like the ability for users to send messages across different networks, the ability to access news feeds and other content from friends or simply public sources in other networks without multi-helming, without using other services with all it takes including the burden of managing privacy there and the burden of accepting or being forced to accept terms and conditions. But indeed as also Ian showed in his interventions why it is simple conceptually it is not simple legally it is not simple technologically that would require adoption effective adoption of open protocols, some kind of standardisation across the EU. We would face or we will face hopefully issues to solve when it comes to data protection, how to protect users from spam, we will face data security issues. I agree with comments I've seen on the chat this is not exactly as simple as sending a text message to another mobile network so even if we see complications around this big promise there is a whole list of smaller but not less important promises that I see in introducing mandatory interoperability and these are honestly for me major reasons to fight for ex-ante rules for large platforms that will hopefully entail mandatory interoperability which are the smaller but not less important promises. First of all I see here a promise of full execution of the GDPR, the law that we have in place for the last two years but that wasn't really effective in forcing large platforms to comply. As long as data is effectively controlled by large social platforms we don't have authorities, we don't have practical tools to for example make data portability a reality or give users full access to their marketing profiles which is essential that they see, verify and correct these or we don't really have a clear idea how people using social platforms could force them to stop using certain categories of data that are sensitive or maybe even have been collected or generated as a result of observation without users consent. We don't have practical tools for this as long as that data sits on these company servers and legal authorities like the data protection authorities would have to go inside intervene in a very radical way to check what's happening on the server so for me introducing interoperability is in the first place a way to get more control over our own data. I imagine a situation where a user is able to send a signal, something like a D&T do not track me but a more granular more sophisticated signal using open protocol to the platform and on the other hand on the side of the transaction the platform server reacts without me requesting requesting compliance by nobody means of platforms controlled interface or I can imagine that there are clients independent clients to manage personal data available to users that will offer a much better value in terms of managing own data in terms of managing settings including advertising settings because of interoperability. Finally that's last but not least there is a promise of unbounding or functional separation of these large platforms which would open a space for very interesting services not only to manage own data but to modify core functionalities of the platform. If we go that far and I do hope we with the EU and the Exxon rules being on the table we go that far I imagine quite radical solutions happening within the platform where independent service providers start offering new ways of managing newsfeeds new ways of filtering out disinformation new ways of protecting users from abusive advertising so here is the answer to the societal problems that you see Manetta also brought in your question we clearly see there is a lot of evidence there's a lot of debate showing that there are real societal harms connected to the way large platforms use algorithms to control people's behavior to control the debate to monetize people's behavior we need to stop that vicious circle of power that feeds back to platforms and gives them every day more data and more power coming with this data and the only way to do it for me is that sort of functional separation where I can still choose to host my content on Facebook I can still choose to keep my profile on Facebook if I feel that that's something I have or I want to do but I can free myself from being subjected to algorithmic data processing at Facebook and more importantly I can free myself from being targeted by Facebook's controlled algorithms that are controlled in their own business interests so that I think that promise is the the most important one when we discuss societal problems and user or simply the protection of human beings online thank you so much for the really fascinating so our time is up it's 5 pm the reason also is my students have other classes to attend so that's it I can keep them so I don't know if you want to say one final word any of you you have 30 seconds it was much too short and we had some technical issues so we should I think repeat it is a wonderful conversation when I found it wonderful so is there anything you would like to say as final words no I take it as a no so I thank you so much I really enjoyed it I'm really I couldn't look at the chat I hope that Astor is making a copy of the discussions in the chat because I'm really very curious I've seen that Victoria has been very active but I'm not good at multitasking so thank you again to everybody take care and let's hope for the best thank you