 But I've been with the organization since 2003. And I've been a Debian developer, actually, in my free, as in spare time, since 2010. Thanks to Zach for being my account manager back then. Coincidentally, I started the FSF in 2003. I also first started using Debian in 2003. So the two have always been connected for me, even though we have some kinds of separation. So I'm here representing the FSF, but I am also hoping to do some Debian work. I currently maintain two packages. I contribute very little. I maintain a planner for Emacs, which is a piece of productivity software. And then I maintain another piece of software to destroy your productivity, which is called XWORD. That's for doing crossword puzzles. Small contributions, but I always love being here and project like. So I appreciate the chance to talk about our relationship between the Debian and the FSF here today. And I want to start off by saying thank you. Too often, this gets lost in discussions about disagreements. But I want to make a point of thanking everybody here for all of the work that you do to deliver free software to users, to promote free software at the end. But even in the absence of that kind of official endorsement or promotion, we need to remember that we have so much in common. And the FSF never criticizes anybody for running Debian main on their computer. That's all free software. The issue is solely about the ways to promote and get new people involved in free software while steering them away from non-free software to the extent that we can. So I'm going to move through this somewhat quickly, because I want to make sure we have a lot of time for discussion at the end. And then open it up. And of course, I'll be here all week as well. So I'd really like to meet with anybody for the rest of the week who's interested in talking about the issues I raise or has any other ideas or wants to have any other conversations with the FSF. So I think Zach and Bill have talked yesterday but really demonstrated the need for us to be talking about this. The free software movement I think has been growing. I think there's more free software users than ever before. I think there's more people who have some idea about the issues involved than there ever have been before. But we also have many new kinds of challenges that we have to address and threats. The issues of services and mobile computing or the growing importance of technology and software as an actual medium for protests and activism. There aren't that many people, especially not doing it for the reasons that we do it. And I think both those talks yesterday to open the conference really pointed to the need to revisit some of the things that have kept us from working together as well. We can already mention that Debian is an actual FSF endorsed distribution, but also independent of that, things that Debian and the FSF can work together on regardless of whether the first thing is true or not. So the FSF doesn't have a single official distribution. We have a list of distributions that we think are suitable for recommending to the public for the goal of promoting and advancing actually conveying the ideas along with that and growing the movement as a social movement concerned with ethics. So the two most relevant parts of the criteria that we have for distributions to uphold those values are first, the commitment to remove all non free software and second, a commitment to not steer users toward non free software outside of the distribution. So now I'm gonna say a few things that are going to maybe sound critical, but I want everybody to know that this all comes from a place of deep respect for what Debian does. I'm a Debian developer and I'm not interested in standing up and criticizing myself, disrespecting myself even at the behalf of my employer. But I think what we want here is for Debian to focus more on the awesome free software work that it does and possibly less on some other things that have been part of it historically. So Debian is the distribution that's the most commonly used distribution that is the closest to meeting the FSF's criteria. And that's because of the huge step that was taken before the release of the squeeze back in 2011 to remove all the non free firmware from the kernel. And because we appreciated that and recognized how difficult that is to do and what a commitment that is, we published an announcement and did our best to praise Debian for taking that step. And this is an example of the kind of thing I want to talk about later of this is related to the endorsed distribution issue but it's also related to the idea of small cooperation on specific issues that we can find reasons to promote each other's work. So by now a list of the things that we see as problematic for that are true of Debian now or seem true of Debian now as it relates to our criteria. And I can't promise that this is comprehensive but it's indicative I think this will cover all the types of issues that would come up. So the biggest one of course is the relationship of non free and contrib sections to the projects. Debian has an idea that those things are not part of Debian and that's stated very clearly in different parts of the project documentation. But in other parts the distinction is not so clear. So we can start with the wording of the commitment to non free software, support for non free software and the social contract itself. And these are the three spot explicit about this that will support people who create or use both free and non free works on Debian. We will not object to non free works that are intended to be used on Debian systems. And we acknowledge that some of our users require non free software so we will support their use and provide infrastructure for them. Those things supporting is not the same word that we use which is a steering. So the question is when does supporting non free software like that become actually steering users toward non free software? I think one confusing thing about the wording of the social contract is the use of the word Debian actually because Debian in the social contract sort of refers to both in the same document the project and the distribution. And I think that distinction is important when we start looking at this closely because the criteria pertain to the distribution and what a user experiences when they install and run the software when they visit the home for the software on the web to learn more about it and that sort of thing. Those aren't necessarily the same thing as what other projects the Debian community might be taking on. So supporting and steering are not issues for the first criteria which is actually shipping non free software. Debian already is meeting that in Maine. There's no dispute about that. So this is really what we have to focus on. Now this idea of supporting and steering gets a lot of pushback because some people see it as kind of censorship or trying to hide things from users inevitably. Obviously anybody can install whatever they want on a free software system that's kind of the point. So we have to, but I think that it's not censorship and we can see this from looking at Debian which itself has kinds of policies that set limits on how far the project will go to steer users toward non free software. Contrib exists, seems to exist for largely that reason that the program, even though it's free if it's only purposes to install a non free program then it doesn't belong in Maine. And then I think it's a very similar idea to what the FSF is talking about that if something exists in a distribution to you very directly lead users to non free software that's not something that fits in a distribution we want to promote as an example of the free. And there's limits on ways that packages in Maine can refer to packages in Contrib or non free. What, you know, where those package names can appear in which fields in the control file. So I think what this shows is that, you know broad generalizations about trying to filter what users see aren't gonna help here. We have to talk about, you know we're already doing that in Debian and we need to talk about how far those things should go and what the actual concerns are. So I have a. So the requirements here seem to be a bit inconsistent. You can recommend packages in non free but you can suggest packages in non free. Well, if anybody user sees this that's a suggestion to install the non free program. And we know that technically well what that really means is the package manager behaves a certain way if a certain option is activated. But if this option is, if this information is displayed to a user suggest me and suggest. That means that our search by default will turn to results there in both non free and Contrib which sort of, you know, to a user who's learning about Debian for the first time certainly give the impression that these things are part of Debian. There's texts on the page which does make the point that they are not. But then you do the search and a list of applications come up some from Non Free, some from Contrib and they're labeled a little bit but it gives an impression of unification I think. The installer, while the non free firmware was removed from the kernel we, I know that people have been very, very present but still wanting to direct users toward the non free firmware in order to have specific hardware on their system work. And I know that's a really tricky problem but I think in this case it has a lot to do with the manner in which it's presented and the message that's delivered. But you can imagine sort of defeats the whole purpose if you just really encourage users to go install that software. How do we actually turn that problem into an opportunity for the free software movement to where we can together apply pressure to get those issues fixed rather than kind of release that pressure by continuing to lead users to the non free software to make the components work. Documentation is an issue in some places. So places like the Debian Wiki have everything all together. If you go through the instructions for doing different things installing the link from software on the front page of the Wiki, it takes you to a page that lists different things you can install and flash and non free things are listed there right together with everything else. You know, it's a Wiki, I could... Okay, thanks. You know, it's a Wiki, I could have just edited it before the talk. But I think it's a principle to think about. What's appropriate for official documentation about official Debian? You know, to what extent should that direct users toward non free applications, especially when things like flash probably aren't really there necessary when people want to do certain things but in the grand scheme of things is Debian going to try to push the envelope there or just provide the technical instructions. This is one that gets very complicated but I think is also extremely important. We'll just make up a name for it today upstream loop holes, programs that are in main but install extensions. There's a lot of them. And if those extensions are non free, is that a problem? And I think sometimes the answer is yes. Sometimes no. And I think one of the cases where it clearly is a problem is the case where the add on extension interface is directly incorporated in the program. And you see this now with Ice Weasel. If you click on add on, so then you get a in the browser. Not, you know, it's course loading over the web but it looks like part of the browser interface. Display add ons that you can choose from. And when I checked this earlier today, the very first add on that appears in the list is one called ever note clearly, which is proprietary. And in fact, it's just not really related but the license is funny. You are free to use ever note clearly with any of the applications or services offered by Mozilla. All rights reserved. So that's a short license to give a credit but that's not free. Worse than that, the list, the application, the add on does not have, there's no indication that it's proprietary in the Ice Weasel window. So I only found. Ice Weasel is not offered by Mozilla, right? So this is, yeah, but I think if you click yes on that, why are they doing the license? Yeah, the, you know, oh, what a tangled web we weave when you start incorporating little proprietary things. And I'm pretty sure the other two on the top row were also proprietary but I didn't click through to check it. So this is the case where I think, this is essentially violating the principle in addition to this specific problem is violating the principle behind contrib which is you have a program which is a free program but is leading to installation of non free software. And of course add ons are an encouraged and suggested part of using the browser. So that's something that I think causes problems for user freedom. So we're working with Trisco, one of the distributions that we endorse on providing an extension repository for Mozilla based browsers that lists only the free extensions from the official Mozilla location. There's been some conversation with Mozilla in the past trying to get them to change that. And if you go to the Mozilla add on pay library in your browser as, you know, type in the URL you will, it will show you what the license is. Weirdly that information is not brought in to the ice weasel embedded view or the Firefox embedded view. Also it's not that helpful because this license was called custom license. No, that's doesn't really say it's non free. And I think this is an example to close these upstream doop holes. One thing that I would really like to happen out of cooperation with Debian is for us to work together on fixing these problems upstream. You know, I think that regardless of whether we end up on the same exact page for what constitutes steering and supporting user stored non free software. This is, you know, Debian I think people would largely agree that this problem that would be better if we didn't have it and if we didn't have to worry about it. And I think that coalitions of us working together can be effective at achieving these things and getting them fixed. Now I know, you know, Ice Weasel is a web browser. So you can stop the extension library from showing non free software but you're not going to stop people from finding non free software on the web. People could even use Debian to write non free software. So it's not about blocking all possible access to non free software. We're not trying to impose restrictions on users but we're trying to do is draw clear lines about what we, what we directly enable, what we directly encourage, steer and support users to do. And I think, you know, this is one example of a way to draw a line is, is that interface directly embedded in the program? If you had to go to the Mozilla site to install a proprietary add-on, that's a different thing. That's just like going to any other webpage. That's an argument we would have with Mozilla and not with Debian. Other programs do do better and Debian main Libre Office their extension repository is actually all free software. So this type of thing is possible in widely used popular programs. And it's something that I think we should work together to get changed. So that's kind of the quick list of types of issues. And I think they all are going to take a lot of thinking to go through. We set up a mailing lists for this quite a while ago and I completely my responsibility, my fault for dropping the ball and moving the discussion forward then. But I think it's important that we start it back up again and we have that mailing list still there for that purpose. And I wanna make clear that we're not just asking Debian to change. I think that I wanna engage in this dialogue because I want to see ways the Office of Criteria can be made better as well. And I think that conversation about the issues can lead to that. These aren't, these criteria have been around for a while and we have distributions that are meeting them but we're always seeking ways to improve them, make them make more sense as a strategy for promoting what we wanna promote. Then a few big questions here just to think about. Is there any kind of victory condition within Debian for removing non-free software from the social contract? What would be, in what world would you, as a Debian contributor, user, think, would make non-free software unnecessary? It's a replacement for the rating that currently exists in non-fee. If you had to think about it. You can't get the answer. Yeah, so I'm gonna, we'll answer these questions together here. But non-free is not very big right now. All right, it's quite small actually. So what is the point? And I think that's kind of a thought experiment to see how intrinsic the notion of supporting non-free software is to Debian and Debian's goals. And related to that, you know, does Debian need to support users who have non-free software requirements and the way that it did when the social contract was conceived given that there are now so many other people happy to take Debian and build non-free stuff on top of it? You know, that wasn't the case back when Debian was started. What does it mean for the Debian trademark policy? If non-free and contrived are not part of Debian, then is it okay for a company to sell a server with the Debian trademark in the compatibility section when operating that server requires installing software from non-free, contrived, or third-party package repository with proprietary software? I think, you know, I did read the trademark policy, it doesn't address this question, and I'm not actually recommending a change to it right now, but I think thinking about that, because that's often a point of disjunct having this conversation is what is official Debian? We have our internal definition and we have how it's perceived, and the trademark policy is kind of a definition of what official Debian actually is when you are allowed to use that mark. So I think that can be an interesting way to think about this. Likewise, another kind of thought experiment, how do you support the user who only wants to make one decision once that they only ever want to use free software, right? So the Ice Weasel case is not serving that user well because that Ice Weasel is presenting them with non-free software without telling them and giving them the option to install it directly from the program that they got in Debian main. So we worry a lot about making life difficult for the users who need non-free software who want non-free software, but let's also worry about the users who wants to use only free software and only want to make that decision once and not have to be constantly on the lookout for what programs might be prompting them to install and how can that decision be made? So that's the endorsement section and this I'm just going to go through quickly because it's the brainstorming list and I'm hoping this is a lot, what we can talk about. These are things that we can do together regardless that I have thought of and would like to discuss regardless of the status of Debian as a distribution that the FSF actively encourages people to install or not. The FSF has been, has a free software directory which lists our goal is to list all of the free software in the world that runs on a free operating system. Also it can run on Windows or Mac but as long as it runs on a free operating system. Recently, a couple of years ago, we converted the directory to MediaWiki and we're doing it with Symantec MediaWiki so that we can provide machine-readable information about all of the packages including licensing, information, dependencies, that sort of thing. We've been working on importing packages from Debbie and Maine and we've also been working on mirroring the category system so that the Symantec properties and the directory have a relationship to DevTags. So you'll see if you visit the directory, you'll see a lot of familiar forms like WorksWith and Uses and so on. And, I don't know, Amiko, yeah. So we would love to see cooperation there and I think this would be an upstream directory essentially so it wouldn't just be a duplicate of what's already packaged in Maine, it would be Maine Plus. Databases of Compatible Hardware which is something that I have been talking with Lucas and Zach about recently. Hnode.org is the free software foundation's database of hardware that works with a fully free system. It's volunteer run but the FSF hosts it and supports it and this site is different from previous databases of hardware because it's not just about the hardware working on a Linux-based system, it's about the hardware working without requiring any non-free firmware or non-free driver. Previously, the way that information was contributed to the database we required users to run one of our endorsed distributions because that was the easiest official Debian resource that I think Debian could consider pointing to this as a useful resource in order to solve some of the difficulties that users installing the kernel without any non-free firmware in it might be running into you right now, especially people that are in the market for buying a new computer. JavaScript licensing, some of you might have read the material that the FSF had been pointing out about this recently. JavaScript is often distributed somewhere else as free software but when it is delivered to the user and their browser, it comes without any licensing information, making it, therefore, by default, proprietary. And we would like one of the ways to address this could be in the packaging system. We're actually working, trying to work with individual sites right now to add licensed notices and we have a format for that called JavaScript Web Labels, which is both machine readable and human readable. You can read about that on ganodata.org in the JavaScript Trap section. I think that's something that Debian could think about doing for packages that contain JavaScript and whether per file licenses have been sort of optional in the past but that's because those are programs that are compiled. If you are actually serving JavaScript to your user as an individual thing, then does Debian need to ask that those files have license information in them? There's lots of other licensing work we can probably do together. We're both, FSF and Debian, very concerned with free software licensing. The FSF maintains a couple of staff people whose job it is to help the free software community with licensing issues. I know Debian has plenty of experts with a lot of experience in the Debian context also but I want to offer that if we've worked with Fedora in the past to go over issues with package licensing and we would be more than happy to help Debian with that as well if there were any opportunities to do so. Upstream policy changes, like I mentioned, in the case of Mozilla, upstream code, what we at the FSF has a high priority projects list where we try to name things which we see as real holes in the world of free software, things that need to get developed or improved and we're working on revamping that list now. So it would be a very good opportunity for Debian users and developers to take a look at that list and send us some feedback about that because we both have concerns about upstream free software. Debian users want things that aren't packaged because they don't exist. They want to be able to have a Skype-like experience that works from their Debian installation and that's one of the things that's on the priority projects list. So I think there's a lot of potential for cooperation for us to call attention to code that needs to be written, possibly put resources into getting it written. Coordination with GNU. Of course, FSF isn't the same thing as GNU but we do a lot of the organizing work within GNU and if there are issues that any of you have working with GNU packages upstream then we might be able to help improve channels and communication and make that process work more smoothly. We talked about this at LibrePlanet several years ago and the result of that was that several GNU packages moved to Debugs and I think there was some other email addresses and such that were created so that communication about security issues and other critical problems could move faster between upstream and packaging. Privacy and security, the things that were talked about yesterday, I think that we both have a common interest in protecting free software users and there's a lot of opportunity there. And then finally, web application packaging. I know it's very difficult but it's kind of key to our future right now if we are a lot of the decentralized applications that are being developed like Media Goblin, like Pump I O, our web applications and they require, we want users to be able to install them easily and that's what Debian's always been great at is taking free software and making it easy for people to set up. I think that we would like to work, especially in the cases of GNU packages obviously, but for anything that addresses that need because that's something that the FSF has been very concerned about as well. So I have lots of goodies in the back there, saw some people taking them, I brought gifts so hopefully I wouldn't get lynched. There are bulletins, stickers, things back there that you could help yourself to and if you want anything, we have these special GNU Linux inside stickers that are heavy-duty foil and will last forever on your laptop. So you can come see me if you're interested in one of those as well. So that's where I would like to start the discussion. We'd love to hear, I will repeat the question or the discussion comment because I'm not gonna try to answer every, my hope is that we can talk together, right? Yeah, yeah? So you talked about web application packaging and I think that is very important. One of the things that I would like to suggest which is quite a radical suggestion is that we should come up with some way whereby it's easy when installing a web application to supply the source code to that application's users. Even for example, if you're using a modified package or a modified module or whatever. So you can look at that as a way of automatically complying with the AGPL but I would like to look at it as a way of automatically providing with the users with the right freedoms as far as we can with respect to that software. Yeah, I think, so what you're saying is that we want to use AGPL, legal compliance but another issue is just the spirit of the AGPL which is to supply source code to users whenever possible. I think that's a great idea. So you mentioned the areas in Debian where we could possibly improve things in order to improve the working together with the SFN. One of the things you didn't mention is that we do have somewhat different views on freedom. For example, if I want to use make on a perfectly free Debian main system and I want to look up the documentation, I won't be able to find it unless I go to non-free. Is there any thought about moving the FSF's position on freedom of documentation as well? I think that that is something that could be on the table was part of a conversation about if we certainly wouldn't let that be the last thing blocking cooperation between us. I think that's something that issues in various sections, right? It's not the GFTL as a whole. So I think that that is something that we would discuss. Hi, so this is more of a comment, but first of all, I'm going to start out with something I think incredibly positive. You came up with something that I think is brilliant, important, and I really hope it was a project can get behind, which is that the commitment to the idea that people should be able to say once, I want a fully free experience and get it. And I love that as approach because I think that is an important use case. On the other side, I actually don't think it's, I think it's actually a statement that I'm in the right place that, that means not on the FSF's recommended distributions list, but yet there are other distributions downstream of us that do a lot more supportive and non-free. I really think user freedom is something that's very important to me, not choosing for our users and letting our users do what they want even when they disagree with us. And some of our users really want to not care about software freedom. And one of the reasons I'm here is that we support that, that I can care a lot about software freedom. We've said in this room we care about software freedom, but we don't have to shove our opinions down our users' throats. And that's very important to me. In fact, sufficiently important that if we were to change our position on that, I wouldn't be here. And I think it's great that Debian can be a moderate community where some people who are very, you know, the FSF is very interested in software freedom. It's not surprising at all. Some of our downstream distributions are very interested in user experience. And Debian is a community where we can all get together and make useful forward progress. And I really hope we continue that. Thank you. And that's why I proposed that way of thinking about it because I think it is, like you said, it's an important flip side. It's a consequence of maybe going too far toward supporting, just installing whatever comes along at the expense of people that do want to make that decision. So I think if we went down that path, that could lead to some interesting things. And I think we should think about how that could be done technically. And I did, I meant to mention and didn't, you mentioned the non-free distributions downstream, but it's also the free distributions that we have on the FSF's list wouldn't exist without Debian because they're downstream from Debian. And one positive thing that's happened the last few years is a bit more communication between those distributions and Debian. So Triscoll is one that's on the list and that's based on Ubuntu, which of course comes from Debian. And then the other GNU Sense, which was originally based on Ubuntu but recently switched to being based directly on Debian. So that is one thing that's happening and that's why I'm a Debian developer and contributing to Debian because those things benefit the free software world regardless of what happens within Debian proper as consequences of the work spreads out. So I think that's another area for cooperation is that derivatives downstream distribution work that's happened over the last few years. If I can be, sorry, if I can be slightly contentious for a moment, if you go to the FSF's page and look at, so I want to find out, I want to use free software, how can I do this? I go to the FSF page and all I get is a list of kind of niche distributions. By not endorsing Debian, are you actually harming your cause because you can't go to the FSF's page and see anything that looks like a mainstream distribution if you want, if I'm interested in free software, I'm not sure if it's something that's maybe for me or not. I go to the FSF's page and there's not, if I see Debian there and I think, oh yeah, I've heard of Debian, that maybe that, there's a problem there that by not endorsing Debian, you might be turning users off. That's why we're interested in this, right? Like, we would love to be able to promote a distribution that is as well known as Debian and as widely used as Debian. I think that we have made things better in the last few years with that list. I think, you know, Triscal, I'm bummed because I showed up, did a lot of work right before I left to come here to do this presentation from one of the laptops that we recently endorsed, X60 running Libre Boot, Core Boot with all of the blobs removed and I had it all ready to go running Triscal and it has a hardware problem, so I wasn't able to use it, but we run the FSF office on Triscal. Lots of the, most of the servers run Triscal and it's a distribution that really does work and so it's an issue, it's not as recognized like you say, but we have made strides in improving our ability to advocate actually installing a distribution according to what the FSF wants to promote, but yes, we would love to be able to have, to bring these courses together. I wonder if one way to, what, sorry. I wonder if one way to solve that might be to, like even just if we don't endorse Debian, maybe we could put something on the distribution page that said based on Debian, because if I recall correctly, it doesn't say that, so like if we could put just like a logo or anything there, maybe we could get some more recognition and a more balanced presentation. I said I wasn't gonna take the mic between every question that now I am, but yes, and one of the things that's been discussed is, is there another kind of relationship, status, recognition that would be appropriate and something I've been thinking about, that the FSF would describe Debian in a way that's different from the other distributions that we don't endorse, because Debian main does contain only free software. So that's something I would love to hear ideas about. We haven't been able to come up with anything that sounds positive enough on both sides, basically. We don't wanna have it be, you want it to be something positive, but we can't say fully positive and Debian of course doesn't wanna apply a designation that doesn't sound positive for Debian. But I bet there's something there, I do, I think we maybe can add something to the page about that. I mean, it's certainly something that we tell people all the time that those distributions come from, Debian. So just because you asked for possible ways of wording it on the FSF site, I thought of putting more use case related content on that. Content on that, that is okay. So those distributions are completely free, but may fit only the use case of very motivated people. Those distributions are free enough that we would suggest them to people willing to make some trade-offs, and these are the trade-offs, and trade-offs in favor of well dealing with many more use cases. And then if you have use cases that are not even fit with that, then the worst, the least bad options are those. That is, if somebody gave you a very proprietary hardware, a very closed hardware to work with, well, the best thing we can suggest you to do is to use that stuff because, well, it's better than, I don't know, running Windows on it. And so in a way, that would allow you to endorse more without compromising in freedom, because that would be clear. And it may also give you an image of actually caring for the work to be done, because one way that FSF is perceived, at least by many people, who are very strong free software advocates, actually, I know is something that we started to call the Stolman paradox. I have a dear friend who's like one of the biggest free software advocates in Italy that I know, and she has this idea that whenever Stolman points out that the problem is usually right, and whenever Stolman points at a solution, he's usually wrong. And I don't want the FSF to be perceived that way, but I think that perception comes from a non wanting to compromise on use cases, which I do appreciate as far as a general long-term direction is concerned, but it's not that useful when you have people that actually have some piece of hardware in hand and have some problem to solve, and they have no choice along the process. I think the idea of some kind of sort of ranking system is really interesting and could be a good idea. That might be a way to accomplish what I was just describing. You know, it's a really complicated question of getting people software that can work on the computer that they already have. And I'm gonna suggest that we focus on the question of how do we get all people buying new computers to buy something that works with only Debbie and Maine and only free software. And it's partly speaking as the FSF because other people are going to make compromises to make things work as required now, but there's a risk of confusing those compromises with something that's adequate. And I think we will stand where we are, hopefully improving the way that we explain those things and come across. But I think the question we can unify on is what can we do to help users that aren't in a situation where they can't afford, well, that's a question that's important, but what can we do to help users that are going to buy a new computer and help them get the software, because it's possible. You can run, I'm not running any non-free software other than the boot firmware, and I have wireless and a webcam and everything else that people, a touchscreen, everything else that people want on a laptop, it is possible, it's limited, but if we do a better, options are limited, but if we do a better job together of promoting that and encouraging companies to do that, we might get better results and not have to answer some of these trickier questions for much longer. To extend on that, just to quote me, I would love to be able to go to the FSF site and see some documentation on how to use Facebook, which sounds silly, but suppose that I'm a journalist and I have to publish on Facebook because that's the requirement of the media outlet I work with, then if I could go to the FSF and not be told you use Facebook, you're wrong, but be told use Facebook sucks to be you, this is what you can do to make it not as bad. There's a plugin not to be tracked, that's how you make a backup of the content you publish so that you still own your content somehow, Skype, whatever. Because we have lock-in situations, just denying their existence do not help the people that are locked in, whereas providing ways of being aware when using it, actually I think provides a benefit and a pointer towards a way out, but also a benefit because then using something being aware of it means not being bitten too much by it and so on. Is there a concern there for Debian as well to address? I mean, that's like, when we're looking, I think that's legit very, it's feedback for the FSF, but what about us working together on questions like that? I'd like to answer Enrico's point, which is historically the FSF have not been, let us say that the FSF strengths have not lain in that kind of very practical user education and there are other organizations who have done a much better job of that and if I wanted to know that kind of information, I would look at the EFF. The EFF have done a lot of really good work at how to navigate through a world of having to make compromises and to an extent, I think it's slightly unfair to go to the FSF who's, we need the FSF, we need organizations who are very reluctant to compromise. We don't want to live in a world where everybody is sort of just like muddling through, it's good to have somebody who knows where they stand and stick to it. How much of the big picture are we missing? Zack talked yesterday about hordes of developers, young developers not choosing licenses, that's significant. We're seeing so much of our computing move away from the end device, that's significant and to see the FSF kind of squabble over blobs, I mean, I understand the need to be principled, but it feels like we're missing a big picture and it also feels like we're not terribly united as free software advocates when we have such chasms that the FSF can't recommend Debian on its website. It's, there's no unity or fidelity there. Yeah, I'm gonna, I'm gonna, we need to unify, we need to work together. I think that that's, I've got this now, so thank you. So we do need to do that to address the big picture, but I'm gonna say one thing which is I don't think, I think blobs are part of the big picture now. So we have hard drive firmware that's being exploited in order to spy on users, having a proprietary blob running in your ThinkPad that, well not a proprietary blob, a proprietary system on a chip running in your ThinkPad that can communicate over the network without you knowing. These are the types of things that the FSF has pushed for always and it's been hard to persuade people that it's important, but I think these things are part of the big picture now and I think that Debian should have a, well Debian already did the right thing with the kernel, but I think as it relates to other non-free software and the risks that it is introducing users to in those contexts is something that should be thought about in light of the things that have happened over the course of the last year and a half, but I agree with the sentiment fully which is we wanna work together in every way that we can and keep the disagreements, something that we can over the long term try to make progress on without obstructing the broader work that we share in common. You said you have a laptop that you can run entirely free software on except the BIOS as the obvious caveat. This one, the other one has a free BIOS. But at the same time, it's clear that there's probably some firmware embedded in there and you sort of, it's clear that there's some firmware embedded in there and you mentioned this idea of it's adequate and we could maybe even recommend these systems to people. Is that just a sort of a practical distinction because we're so far away from having fully free hardware that you make this practical exception in this case or is there some actual difference in ideology that you honestly believe this is different? Some of it is just new stuff has come to light. Some of it is learning more about the way components operate and what needs to be adjusted in response to that. Some of it is decisions about how we can best push the envelope without not using a computer entirely. But we are, every time we find a piece of non-free software and something we will work to get it out and that's the same thing that we asked for from our endorsed distributions, isn't, stuff gets in there, it happens even with the ones we endorse but there's a strong commitment to address it and get it out. And so that would be the same thing if we were talking about Debian as an endorsed distribution. It's a huge distribution, not expecting you to get everything out all at once just to have a commitment that that's the right thing to do. For hardware the same principle as making an analogy for that. I'd like to make a specific response to that question. So it seems to me the question is why do we make a distinction between the firmware that is shipped in a binary blob that originally came with the kernel and now maybe is in some non-free firmware package? Why do we just differ, distinguish that from the situation where the firmware is an EEPROM on the device? Is that your question? Why do we tolerate non-free software in the device? Well there is a power imbalance if somebody can update the software and somebody else can't. If the manufacturer put the firmware in there and they can't update it then at the very least we're not any worse off than they are now. Although obviously it depends slightly what they programmed it to do to start with. So I think there is an ideological difference there. And why do we tolerate firmware non-free, firmware in devices at all? Well to an extent it's not really practical to do otherwise because the engineering solution for building a complicated system is to put lots of computers in it nowadays. Well I think that's another area where you're tying this all back to user security and privacy that the F7 Debian could work together is on the question of secure boot and what we started calling restricted boot which is the case where you have a security lock on your computer that you don't have the keys for in the UAFI firmware. And Debian as far as I know hasn't adopted a method for installing on computers that come with secure boot enabled by default. So I would love to be able to participate in the conversation about how that gets approached. The F7 doesn't reject secure boot entirely we just want it to be under the user's control that's kind of the definition of free software security. And I think we need that, the reason I thought of that was because of the idea that this power relationship other people have the ability to modify the software on your computer if you have proprietary bits installed things like secure boot actually holds some potential for protecting users from manipulations like that. So we should think about not only how to get proprietary stuff out there but how to work together to improve to give free software security features that proprietary software doesn't have. So I have a question that'll sound either defeatist or like a troll so if you'll feel free to just tell me you'll answer it offline. I'm requesting whoever did the micro update and that's a security update. Well, other software companies have distributed security updates that have enforced anti-features on users along with them. So if you can't see it, you can't know what it actually is. So I know we're falling way down the rabbit hole on this topic but I have one last to say. There's actually a continuum here. You've got software at one end that can be updated, is updated daily on the video. At the far end, you have circuits that can do computations, firmware falls in the middle. When it comes to freedom, walking down the spectrum is, gets harder. And it is up to the end user and to the distribution like how far down the spectrum we go. And if you're out there, the circuit end of computation, all you can say is use this or don't use this because somebody has put an algorithm statement there that you don't know and you can't control and you can't modify. And you have none of the freedoms that we are looking for. So it depends on how far you're going down the spectrum. Just a quick follow up to that. So we go as far as the stuff we distribute ourselves. We distribute here where we distribute software, we do not distribute algorithm. Thank you very much. Thank you. Thank you very much, thank you very much. Daniel was for, can you say what's going on there? Why are you doing this? I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. So just going back to secure boot, we did have a conversation about this last year. So I came up with a plan of things. I think there was a consensus that so long as the, we can support this on architecture where the, where the firmware will, the system firmware that's already on the system will allow the user to replace the trusted keys. But if there was some system that was, some platform that was designed to not support that, that would then, then Devin would not participate. So we can do secure, so basically we can do secure boot if you wouldn't do it with, what's the patient restricted boot? The program got to interrupt here. There was a video team meeting in 327 for anyone interested in that. And we'll probably be leaving as a result. So you guys can continue on, but it will not be recorded. We'll be controlling the last one. Yeah. So good luck. Thanks. Well, thank you. If I can just grab the last word. I wanted to thank you for your work for the FSF. And I wanted to echo some comments that happened earlier saying that we do need someone who stakes out the far end of the spectrum. Sorry for calling it fringe. I'm the FSF supporter. I'm very happy that the FSF is there doing that. And I wanted to say also that your recent focus on security and privacy is critical. And I'm very happy to see the FSF doing that. And that the trade off between security and privacy and software freedom is not a trade off. There is, the challenge that she's posed is two pieces of non-free firmware. And you don't know which one of them has a thing that's going to make you more vulnerable. And so the question of which do you do, you're stuck either way. And the FSF has been pushing for a long time for the security and privacy of the users. And it's now explicit. And I really appreciate that because free software and free firmware and free microcode for your CPU is the only way that we're going to be able to actually protect against the kind of attacks that we're seeing. So thank you for that. Thank you. If anybody is really interested, well, another official Zephyr. Yeah, or official Zephyr.