 Så välkomna, everyone, sorry for my bed throat, but that's the way it is sometimes in the winter season. So I'm Björn Lundell, I'm a researcher in this area, the Fosch area. Walk out from that speaker there, I think, yeah. So together here with Mathias, he's also a researcher. And then we have two practitioners, Andrew Katz and Pam Shastik, with a lot of experiences in this field. And in this panel there are these that we will look into contemporary challenges that may inform practice. And also the other way around, in other words, what are the design of proper and really good and useful research studies. So we want input and we want to chair first the lessons we have learned by a few position statements before we open up for discussion. That's basically the idea. And you can see the affiliations there and the slides will be available afterwards. And the other panelists will introduce themselves in connection with their position statement afterwards. But first Andrew will say a few words on the idea here with the panel. Yeah, thanks very much Björn. I feel immensely privileged because as a practitioner it's not very often that you get an opportunity to get involved in the world of academic research. And for the last couple of years I've been working with Björn and with his colleague Gjörnas, who's in the audience here at the University of Huvdon, Sweden. And it's been the perfect mixture of, you know, I get all the good bits out of doing academe and none of the bad bits. So the good bits, they pay me, I get to research interesting things, I get to hang out with really bright people and talk about cool stuff. And I get to avoid the bad stuff, I don't get involved in any of the politics and having to do all the applications of research grants and that sort of thing. So I feel immensely privileged that I've been able to do that. And during the course of the research that we've done a few aspects have emerged from that, which really makes me think of the interface between what I'm doing in my day to day life as a practicing lawyer and what I'm doing in terms of research. And I think there's a huge opportunity for more interrelationship between those two particular worlds. And what I've been impressed by is that Björn and Gjörnas have always been looking on the practical side as well as the pure academic research side. I mean, you know, there is obviously definitely a place for blue sky research. But the idea is that the research that we're talking about needs to be directed in a practical and useful manner, a way that there's actually going to lead to some sort of results that are making some sort of difference in the outside world. And we're starting to see that already with some of the research that we've been doing in terms of unit standards that we've been able to demonstrate fulfill criteria in the European interoperability framework. So they classified as open standards and also a sort of recent position paper that was issued by the Swedish government there. It was a competition authority, wasn't it? So, you know, I think the emphasis is we're not talking about pure academic research. We're talking about areas in which academic rigor, academic resources can be applied to things that all of us come across. Most people in this room are going to fall into categories of developers, lawyers or academics or some combination of those three. And I think that that's the sort of direction in which that sort of research needs to and which we want to direct in. OK, thanks, Andrew. So we go through the four four of us. So I will start and give a few thoughts on what we have done as a position statement. As Andrew mentioned, we have done a big study for the Swedish competition authority where we looked at a number of projects. And in those projects, we see that there is a desire to develop system. And when they do that, if you do like procurement, you sometimes express yourself when you put requirements in a way which really inhibits use of us. You really inhibit use of open and open attitude and open standard. So you see a lot of requirements for specific preparatory products, which definitely exclude for solutions. You see specific trademarks, which is controlled by a single company, for example. And these kind of things, when you also refer to standard, which is controlled in a way and govern in a way, which cannot be implemented in force. Then you run into problems. And as part of the research, we have also commented on the new version, the draft version of the European interoperability framework as part of this. The study we did for the competition authority was published last autumn and the director general for the Swedish competition authority. He was really concerned. And in the forward to that, he really expressed that very clearly. That's quite unusual for such such report that closed standards and reference to such is problematic. I should say that the focus of that study was not false as such, but it was on standards and the effect it may have on competition. And when we saw that you refer to and write requirements in such a way that you in practice inhibit false, then it's not good for competition. So there is quite a lot of lock in. It's quite clear in the Swedish field. How should you do? How should you go about this? Well, we have another government agency responsible responsible for building framework contracts. They have designed framework contracts for adopting fos in the public sector. They started more than 10 years ago, actually. It's the only country in Europe that has these kind of contracts. And that practice has helped many agencies to work with this. And now recently, when they developed their most recent framework contract in 2014 for software and services in general, not only for fos. But they have a requirement saying that if you put in a mandatory requirement in a specification for a project, that that requirement can only refer to standards which are open standards in order not to inhibit fos. That's an effect of that. So it's it's sort of not okay to use that framework agreement if you refer to a standard which does not conform to this definition. So it's a way of strategic thinking, trying to avoid inhibiting competition, trying to avoid use of fos. So you could see it as a passive way of doing this. We have seen that a lot of management issues in this area is really problematic. We have seen lots of examples and it's really difficult to find good examples, actually. So the last quote that comes from the English summary of this report, it's a 200 plus page report for the competition authority. We have presented this in the European Commission at various meetings. And it's quite clear that there are few studies of this kind from different countries. So I think there is a need for more studies as part of this. So give you one example of what we have looked at in another study. We have tried to obtain clearance on a specific standard, which is an item standard. We try to contact all the organizations that control patents on a specific standard. And as you can see from this overview slide, it was not so very positive outcome. First, there are 18 organizations that have declared that they control patents on that specific standard. And we were able only to get in contact with three of them. The other ones we couldn't get access to, despite the fact that they have committed themselves under the ISO regulations to actually respond there. And the three that we were able to get in contact with, we were very unlucky or it was not so easy. So we couldn't clarify this as part of the study. And in the paper, we have quite a number of examples and quite a number of recommendations that could improve practice of standardization. And that is something that we have discussed now in various contexts. It's also clear that the content of these patent database is not so. It is not of very high quality. Five of the 16 organizations we couldn't even reach. And that's another problem. So that was my short introduction. So now over to Mattias. Thank you. So my name is Mattias Sturmer. I'm from Switzerland and studying and researching at the University of Burden. We are doing research like I wrote my master thesis and PhD thesis about open source software research. And now I'm still doing this because I think it's very interesting just to find out how communities work and how also governance works, for example. Today I would like to point out one particular research area we are focusing during the last two years. It's about ICT procurement of Swiss government agencies, but also any other organization in Switzerland, which is below the procurement law. So these are all the rail organizations, all the airports, all the electricity, power plants. Everything which is publicly funded is below the procurement law. And so they are obliged to actually make public tenders below the WTO agreement to the government procurement agreement, which is an international contract between many countries, about 50 countries, which agree that they want to procure publicly their goods. So what we did is actually we crawled the data of the Swiss national database, it's called CMAP.ch, it's the central platform where every agency publishes their procurements. And we actually were able to filter the data for IT procurement or actually ICT procurements depending on what codes you use. There is a CPV coding schema, common procurement vocabulary, so you can filter out all the IT procurements. And then we actually looked at which of the IT procurements were really going below the public tendering process and which actually had no bid contracts or direct awards without any tendering process. And we found out like with the non-IT area there's about like 15% of all the contracts are awarded without public tendering and within IT it's about 50 to 60% of all the contracts are without public tendering. So it's a very high amount of direct awards just to give you an example. For example, the Swiss federal government last year awarded 80 million Swiss francs, so it's about 70 million euro contract to Oracle without any public tendering. It's just regular license purchasement because they are dependent on Oracle, so that's why actually they say they are not able to find any alternative and that's why they give them a 70 million euro contract without public tendering. So this is the one category we found that there is no direct, there is no open procurement and then as Björn has said there is also a lot of procurements which say we want to buy Microsoft licenses. Who can offer us Microsoft licenses or who can offer us Adobe licenses or SAP licenses. So it's very product specific and very targeted so it's not really a non-functional vendor neutral procurement but it's actually very focused and this is the second category of products of problems we see. The third problem we see that they say for example we want to procure a new IT system. We allow open source or they don't say what they do but they don't really award the special qualities you have with open source. For example that you can copy it without any restrictions that you can look at the source code that you can contract other companies. And they don't really value the special properties of open source software. And the fourth problem, once they do actually procure open source in some or the other way which is good of course but they forget to actually ask if the product is just supported or is it used by only one company and developed. Does it have a community, how heterogeneous is the community, how active is development and so on. So these are the four issues we see in procurement. What could be improved for example, one first thing is that agencies should first of all publicly do open tendering process and not do direct awards. Second thing is of course don't specify any products or vendors but only say what functionality they need. A third point would be to actually say what of the additional award criterion of open source software is relevant for procurement process. Or actually that's what some agencies now start to do to actually require to provide the source code for the development. The city of burden for example procured a procurement software and they actually ask the company who develops the software to provide it as an open source software. And then actually the city is able to publish the software below an open source license again. And fifth to really measure and also demand certain quality of the open source software procured. So that there's not a software like developed by only one company but it's more broadly supported by companies. If you're interested in some of the data or most of the data we publish on a daily basis be schaffungsstatistik.ch, it's in German. I'm sorry but the platform publishes on a daily basis all the procurements linked together. So if you're interested look at the numbers. So thank you. Thanks Matias. So moving on to something slightly different. One of the papers that we were looking at involved analysis of the contractual documentation that students in Swedish schools are required to enter into when they are using the schools ICT equipment. And that also in many cases involves giving those pupils access to specific equipment. So they maybe lent laptops for example. Those laptops may contain free and open source software and one of the areas of research was analysing the inconsistencies between the terms contained in the licenses which the software is licensed under and the contracts that the students were asked to sign. But it became fairly clear that that really is one corner of a much more significant problem that I think is worthy of further investigation. So clearly we were only looking at this particular issue as far as Swedish law is concerned. And secondly we were looking at a fairly narrow use case in which the students were acquiring software that was preloaded onto a laptop that was lent to them or they were asked to download software at a later stage. Fundamental question is whether that amounts to distribution. As we all know the requirements contained in copy left licenses are triggered by distribution. And it's not straightforward in many cases to determine whether distribution has actually occurred in certain circumstances or not. So you can look at a pretty broad spectrum of this. At one extreme you could say that as has been argued in Germany before under German law, Mirren Bauhausen wrote a paper on this in 2014 to say that mere access to the functionality of a piece of software is amounts to distribution. So that's essentially saying that any version of the GPL is the AGPL. On the other extreme you could say that if I load a piece of GPL software onto some sort of device like a laptop or a set top box and I lend that to somebody whether that's on a short term, medium term or long term basis then no distribution takes place at all. So there's really a lot of scope for analysing with this lies, especially in terms of virtualisation. It's a very interesting question if somebody instantiates a virtual machine, they load the software into that machine and then they give somewhere else access to that machine. You know, we all have a gut idea that some sort of distribution must have taken place. But does that distribution take place at the point that I give the customer access to the machine? Or does it take effect at the point where the customer gets access to the root password? So I think that's something that's worthy of further investigation. Those analyses will probably differ from jurisdiction to jurisdiction. And from a practical perspective I spent quite a lot of time advising projects on what is an appropriate licence and clearly knowledge of what distribution actually means is pretty relevant in terms of being able to make that distinction. And the one thing that I will not be saying is distribution means distribution. So next slide. And then moving on to something completely different. So Björn has already touched on this particular issue in terms of the research that we have done in the context of standards. So what we are suggesting is that when a project wants to implement a piece of software that impacts on a standard, so makes these for standards and the standard essential patterns involved. Björn suggested earlier, there are major difficulties in terms of those projects being able to determine where the standard, what their particular implementation of it is actually impacting on any patterns, difficulties in terms of getting hold of appropriate licences for those patterns and also a potential incompatibility between the terms of any licence that they may receive and the licence of the software that they are employing, particular difficulty with the freedom or death clause in the various versions of the GPL. So at the moment in terms of the specific, if you are looking at organisations like ISO for example, the declarations are recorded in the database. But all of those declarations indicate is whether a particular organisation first of all claims to have a standard essential patent and secondly the terms on which they may or may not be prepared to licence that patent. They don't go into any details in terms of the licence other than in broad terms whether it's going to be royalty free or not for example. And they certainly don't go into any detail in terms of whether that licence is going to be compatible with specific licences or not. So one of the suggestions that we're proposing and something that also requires further research is in addition to being able to specify particular options on the declaration database, it's possible for the declarant to actually declare an actual licence on the database itself, which we're calling option zero. So the first part of this slide is the actual declaration as it exists at the moment. So if a declarant selects that particular option, then it's open for anyone wanting to implement that standard in software to approach the declarant and say can you let me have a licence, and then you're into a process of the declarant either responding or not, then providing a draft licence, then you have to analyse the licence and find out whether it's compatible, etc. The second part of the slide presents our suggested alternative, which would be possible for a declarant just to declare on the declaration form that yes, this is the licence that we're prepared to grant. It's a pretty broad patent licence that's intended to be specific to the implementation of that particular standard. And it also contains wording that's fairly familiar to anyone that has looked at the patent licence wording in licences like Apache, GPL v3, and so on. So the intention here is that if this was made an option available to declarants of standards, then it would really simplify the licence arrangements significantly in terms of projects wanting to implement those standards. And from the declarant's perspective, if they're large organisations that are used to licencing their software under licences such as Apache, GPL v3, then this is language that they will be fairly familiar with in any event. So the research is really intended to establish whether this is a viable option and whether it would be something that would help both the standards setting community and also any projects that want to implement those standards. If you want to read more about it, there is a paper on that. I'm Pam Chastick and as you can tell I'm an American lawyer. I came from a world before I worked in software. I was a lawyer for traditional companies including a proprietary software company. So I came to free software after having traditional training in the legal system and trademark and copyright in particular. And one of the things I've always been interested in is whether or not the legal system is actually performing the job that it was meant to perform in a particular context. So I've always had the sense that sometimes as lawyers we'd love to have the sort of doctrine that you go in and you can look up what the rule is. Particularly in the US where we have case law and we get to do research and we look it up and say, oh here's the rule. I have these four factors I need to apply and you sort of wind up. And sometimes it seems that you get into a system that is so far removed from what the actual intention was that you sort of have lost sight of what you were trying to do. And I'll bring that to start with the US constitution. I will move along here. I'll start with the constitution which as all the US lawyers have been trained. We have the patent and copyright clause which says which has this fundamental premise that any lawyer I talk to outside of free software just absolutely cannot rid themselves of. Is this concept that you have to incentivize people with money in order for them to be creative or to invent. So I've always sort of been skeptical of that. I think probably a lot of I would have a lot of people in the room who might also be skeptical of that. Then we have sort of another another event which was in the early 1980s. I got from the 1700s to 1980. There was an open question in the United States whether software should be subject matter of copyright. And there was a government study and the government study said yes indeed we believe that software should apply to copyright. And there are people who to this day strongly disagree with that legal conclusion. And I understand their rationale for disagreeing with it which is the highly functional nature of software. So as all of that is as a predicate what what I wonder is we wound up. We're sitting in these rooms having these legal discussions about these complicated licenses and and brilliant licenses. I don't I think it's a wonderful system. But I question I'd question everything about it. Whether the copyright system whether whether the legal system is performing the function it needs to perform in the free software world. So I'm going to start with some assumptions and I will even say you can disagree with these assumptions if you choose to. And I'd be happy to have people disagree with these assumptions. But these are assumptions that I started with in order to make a second slide. So the first slide was assumption. The adoption of free software is desirable and the legal system should provide incentives or at least not be an obstacle to it. Second assumption. We have in free software the legal system should incentivize it both the contributor and user level. So we have a lot of layers here where we can impart pressure to incentivize. And so for example it can be the individual contributor or it can be the company that's using it. But but there are there's multiple layers going on here assumption participation is vital to the success of free software. This I find interesting because this assumption is is entirely contrary to what copyright law has been built on. Is which this is which is the assumption of an individual. You know the artist in the Garrett who is creating this sort of soul work. And instead we have a hugely collaborative work. You know we there are other kinds of collaborative works movies and theater collaborative work also. But I don't think anything quite on the scale of free software. So those those of these with those assumptions. I will now ask the questions which so what the analysis was then. So here here the interests at play as I see them in free software. When we're talking about the traditional sort of quote unquote intellectual property and I've not included patent here because I don't think that's an incentive at all. But what we have we have economic rights. So we have the economic rights. We have moral rights and we have good will talk about each of those. So economic rights. There's right. I broke out. So you'll notice I didn't say copyright here because there are two really two interests that copyright covers. Different terminology in different countries. But we're talking the two aspects of an economic interest and a moral interest in the work. So in the terms of an economic interest normally I write a book and I sell the book and I get money for it. Well that's not. You can't you can't make money selling free software. Instead the economic interest comes from as much more indirect. You may be paid to write software so you get income from that. You may be paid by a software company that sells and distributes software as part of a larger project. It may not even be it may be more indirect. You may work for a bank that simply uses a software to facilitate the banking business and pays developers to write that. So it's a very indirect. There's also it's typically characterized as an economic incentive. But I think that we have a different kind of reward system here which is where one of the rewards might be that we get more free software out of it. That's the copy left license or simply we get improvements to it even on a permissive license. We're going to get improvements on it so. So there may be value in that economic system but that just gets realized a little a little differently. Moral rights. One of the things I always hear what's most important is developers is their attribution. Do not mess with the attribution. And then but I will say the moral rights. I don't know if it's on the next slide or not. I apologize if it is moral rights are not. There is no duty under US law at any rate to provide any attribution to the author. So this is an incentive. It's completely missing altogether from US copyright law and then goodwill. This is the concept that. So there's goodwill accruing to the project itself a very famous project. Of course Linux of all known project. It's a lot of goodwill in that name. There's a lot of affiliation. There's a lot of strength. There's a lot of value in that name. Don't tarnish my name. Don't tarnish my good name. Don't tarnish my good name. And that also then I think accrues to some degree to the contributors to it. As members of a project may sort of also get some of the gloss from that membership. So those are those are the incentives that I think exist in free software. So the question is what so am I right or not. I don't know we'd have to study it and find out just some thoughts sort of study open source companies. And the source of their revenue. How closely is it. Is it actually tied to copyright tied to that economic interest or not. Some ways I suggested we could study it. And then after we've done that. Are there changes to the laws that should be made that would benefit free software. And how do we do that. So for example. At this was my point. There's no copyright law that requires attribution. Should it should an employer be required to give its employees attribution. When they're working on free software. If it does economic reward player role. If it does. Is it working well. If it doesn't. Does it matter. Do we need to change it. Or is it something that. You know we can just ignore it because it doesn't. It doesn't have that much effect. And then the trademark. Does the trademark system do enough to protect this goodwill interest. That particularly young projects have and they're all there established in the name gets well known. That's great. But we have this sort of ambiguity about when that goodwill attaches to a project. With the project with the software name titles in some countries whether they protect them or not. So those are those are sort of the. So my take away was kind of like we have this really awful system for trying to incentivize what we want to incentivize. It doesn't. Yeah. We see. It seems to work. So there are better ways to do it. Is it to relieve it as it is. Is it work. Or should we investigate and see whether there are other ways that we can satisfy my basic assumptions. Which is more participation. More adoption. Those things. Questions. Are we on to questions now. Yeah. So now it's time for discussion. I will help my. I wonder if it might make most sense actually. So I'm going to move over here where there's not so much of a problem with feedback. Whether it might make most sense to deal with sort of Pam's section first. Because that's sort of freshest in our minds. And you know if we've got any thoughts about that that's probably the place to start and then work back. Was that make sense. Yeah. Okay. So I don't know that necessarily the best way to deal with it. First of all any questions from Pam's interesting talk. Yes. Judy the microphone. So my question I guess has to do with jurisdictions. One of the great things about Foss is that it's a global phenomenon. And yet the law is very specific to jurisdictions. And so the incentives that you talk about are going to be different across jurisdictions. Is there something that we can learn from Foss to bring it back to the legal profession to. If you will harmonize the incentives globally. Yeah that's I mean that's a really good question. That's sort of an even higher level than you're absolutely right. And that's kind of when I broke it out as economic and moral. It was recognizing that what we're looking for isn't is a normative situation. And the different countries laws each adopted differently. That's some sort of that the even follow on question is once we discover whether or not the system is working. You know how then do we implement it on a specific level in each country. That's a good point. So you mentioned several frameworks of law to try to incentivize these things and provide the goals that you want. To what extent is it worth considering whether exclusive rights in law like copyright are the right instrument to use versus. Going beyond those and just outright saying here's a legal contract as a license and the contract says exactly what we want and what we get. And whether or not it needs any basis in copyright or some other exclusive rights. We've traditionally shied away from that in open source licenses and said copyright defines the complete scope of the license. Could we get more out of not doing that. So the difficulty with trying to do it just on a contractual level is then you need privity between the the party. Someone has to accept the terms so I can I can in in the U.S. It varies from country to country but I can sort of impose just about anything I want in a contract. But I have to you have to agree agree to it. So that's where that's why copyright works is is what it is is me giving up some of my exclusivity. It's only if you stay within the boundaries of that that that keeps me from suing you for copyright infringements. But so a contract doesn't you need that sort of exclusive exclusive right founded in the law that you could then wave which gives people that incentivizes the behavior they're looking for. And I think Andrew could talk a lot about why it doesn't work in hardware. Because I mean to me that's actually another good point is sort of that we've created this ecosystem in software because of copyright and it works. And I think a lot of people have looked at it and tried to figure out how to make it work in a different kind of product. And we haven't really been able to find a way to do that. So we simply got a fantastic resource here of sort of bunch of lawyers from different jurisdictions. So my understanding is that certainly under the US law moral rights do not have a great deal of standing under UK copyright law. Moral rights exist but they're specifically excluded as far as software is concerned. So can any laws from other jurisdictions enlighten me as to whether in civil law jurisdictions in particular moral rights do extend to software. And actually just to amplify that to where they sort of serve as an incentive in software where they don't they wouldn't in the US. They wouldn't in the UK because they don't exist at all. There is no way to give up your moral rights even for softwares. In the I guess in the at least in the Italian and other countries in French system too. I'm quite scared about the US system because the author the developer could give away his own moral rights. Any attribution. Did I understand correctly the point. So I think we have a question at the back here. I'm going to be so fit. No sorry I need the exercise. Right. So who was asking the question. So in France we do have moral rights. It cannot be enforced. It has never been for software for free software to my knowledge. Nevertheless it's such a tradition that as a developer working for a company for contractor. If you ask that your name shows in the end product it will not be denied most of the time. So it's worth pursuing from the developer point of view and you have informal leverage to do that. Thanks. I think the other thing is that since copyright exists exclusive rights anyway then. So if you look at the notice provisions in the Apache 2.0 license for example they can become. They can act as a condition on the exercise of the other rights within the license. Sorry another question. Hello I'm Federico from Wikimedia Italy. I have one impression about the last presentation. Could you say that your presentation is mostly about avoiding the problem of the free rider. Because when I asked myself what sort of incentives we need for people I would say that the incentive should mostly not be legal. But simply because more people use a software and have an interest in developing it. And this is how usually free software improves. So this brings me to a question about the first presentation which is how much the restriction on procurement. Which are required to not if I understood correctly prohibits from mandating any specific technology in procurement. And unless it's included in the European interoperability framework and I did I understand correctly. And if so how well did it work because in Italy we have a similar law since 10 years ago. And it didn't have any practical effect. And this brings me to the to my point which is in my opinion it's not so important what the or it could be that it's not so important what the law says. What incentive is supposed to give to developers or software authors. As much as the problem that first individual developers rarely have the ability to actually enforce their rights. And this is the first problem. And second sometimes the law gives them some rights they do not want. Like the European Commission is now proposing three or five pseudo copyright rights out of the blue. Which could apply to software or some other free culture works and were not included in copyright licenses. So one would have to wave again all the new rights every time the legislator comes up with a new crazy law. And and you would have to retroactively relicence everything you have produced in the last few decades. And vice versa. If if. Yeah. Is there a question. The question is. Yeah okay. I'll answer the first one. Yeah so I think it just sort of summarized the question. On on the general topic which is yes absolutely one of the things I question is do we have this incredibly complicated system. That serves no end whatsoever that free software might exist just as happily if we didn't have this incredibly complicated legal system over overseeing it. And that is that is I said question the assumptions because I'm not sure that the assumptions are right or that it should exist at all. I think we have a second. We have a second answer here so so I completely agree with you which is why I mentioned the constitution was. I'm not sure that it's right but I will then there was a follow question. Yes it was about this framework contract. It was established 2014 and in the study we did for the Swedish Competition Authority. The projects we investigated was done just before that. Generally I can say that there is a limited understanding of the fundamental principles for public procurement. And it's quite clear that most of the issues that arises is due to a lack of understanding of those principles. It's short term win short term gain versus long term benefit as we see. I can run the mic. I understand that from from this talk about the moral obligation that it's only enforceable within a contract. And so I kind of understand the GPLCC etc. Doesn't fall under terminology of contract but under terminology agreement. So how can we turn a GPL agreement into contract so that they can enforce the moral obligations etc. The GPL is a license which provides exclusions which provides permissions under the various exclusive rights that you have. And the most important of those is the ability to copy the material to make modifications to the material and also to distribute the material. So if you don't comply with the obligations that relate to attribution then you don't have a license. So whether it's a contract or not is actually irrelevant. It means that whatever you're doing, whether you're distributing or whether you're copying or whether you're adapting. If you're not complying with those obligations to retain attribution then you no longer have a license. And as if you had no license at all you're in breach of copyright. So moral rights sort of sit on a different level depending on which jurisdiction you're looking at. I did just want to make a comment about it. Creative Commons licenses I think are a good example. So I can condition a license on being paid. I will not sue you if you pay me to use my software. But you can make that condition on anything. And so one of the conditions that's in Creative Commons licenses in every single Creative Commons license except the public domain is the attribution requirement. There is no right to attribution under the copyright law in the US. So I find that sort of a fascinating study in the fact that one of the biggest rewards, one of the biggest things that Creative Commons wanted to preserve was attribution. Which is not even something that we in the US provide for. And therefore you only get it because they make it an obligation under the contract. So I couldn't sue someone for not attributing me under US copyright law. So I think that's an interesting, very interesting point. Let's take the panel.