 Okay, good morning. It's a pleasure to see so many people after nine days of death camp and several days of death camp, especially if they are all still awake. I hope you don't change that. I don't think I've got any risk in for these people on this subject matter. Okay, no more jokes. One of the most important documents are our free software guidelines and he's about to tell us something about them. So please welcome Matthew Carrick. So this is strictly down as a burst of feather session. So I'm going to try to go through what I'm going to say reasonably quickly and then we can all argue viciously over what any of this stuff actually means. Again, so what are the Debian free software guidelines? Strictly though this may have been changed to some extent by the general resolution on foundation documents, the Debian free software guidelines are part of the Debian social contracts with its users. They describe what freedoms must be provided by a software license for us to be able to consider it free software. In terms of the history, in 1997 Bruce Perence, who was the Debian project leader at the time, emailed in answer to somebody posted a software license to a mailing list and Bruce responded with what he considered to be his set of criteria for determining whether or not a license could be considered free software or not. This was later posted to Debian privates as part of his proposed social contract and they then followed some argumentation on Debian privates and a moderate amount of privates emails that aren't archived anywhere. And so as a result the process by which the Debian free software guidelines, perhaps one of the more important documents in the entire free software community, were written and the thought process that led to certain wording being adopted is completely hidden from anyone who cares in the slightest except for Debian developers and even they can't see most of it. Which is a shame. So here we have a set of criteria for free software. Software must be free to use, software must be free for people to study, software must be free to distribute and people must be free to modify and distribute their modified software. This isn't the Debian free software guidelines. This set of points is the free software foundations for essential freedoms for free software. The Debian free software guidelines are quite similar but a little more clearly defined and essentially what they are are the free software foundations for freedoms plus a small list of explicitly acceptable restrictions such as you may require that the name of software is changed, you may require that modifications only be distributed via patches against the original software rather than as an integral part of modified software. We are more explicit about the fact that freedom must be for everyone. You can't have a free software license that's free for everyone except employees of nuclear electricity companies or the military. It must be free even for people who don't like you. And we explicitly state that free software must be able to coexist with the world of proprietary software. It may not need to support it but it must coexist. We do not accept as free a software license that states that all other software on the same distribution medium must be free software because we believe that people should be able to use free software even if they are working in a situation where they're not entirely using free software. And essentially that's without some publicity the difference between the Debian free software guidelines and a free software foundation for freedoms. So what I'm going to do now is to look at some of the debate over certain aspects of the free software guidelines. So for instance we have Debian free software guideline eight which states something along the lines of license must not be specific to Debian. The rights associated with the license must apply to everyone who receives the software not just those who are not just Debian. So this is to prevent licenses such as this where you have something saying Debian is allowed to treat this as free software but anyone else who receives it has additional restrictions. And pretty much everyone would agree that this is not free. And this is the sort of thing that's occasionally appeared around the time that the Debian free software guidelines of the recent people would say oh well I've written this piece of proprietary software but I point my Debian to be allowed to modify it. And so they come on and they say well here is a license that allows Debian to modify the software and now surely that's free software because you're allowed to modify some distributed modified version. Can you recall a specific example because I can't? People suggest this at various points. It was a common response from because we would mail, I don't know if people still do, we would mail the maintainers of people who would see the license it saying would you care to fix this city license and they would often write back and say well I'll never fix for everybody but especially for you. There's certainly a couple of them so I have successfully found and I think there's a reference for at least one of them in my paper which is available from somewhere. I should read. That's obviously something weaknesses are not free license. Now the problem with certain interpretations of DFSG8 is that you might end up with a license like this that states that the software is released under the new general public license which is the free software license and the Debian project is grant permission to distribute the drive words under a license they're choosing which means that the license grants Debian permissions that it doesn't grant to its other users which you could then interpret as favorite DFSG8. Or five or six. Or five or six whichever. But we have this issue that certain literal interpretations can have slightly odd side effects such as saying that a piece of software that grants more freedoms than the new general public license is not free which I think again people with 10 degrees probably not be desired side effects of this sort of clause. So one thing that can be important while reasoning DFSG is to think about why they're risen the way they were. And unfortunately at the moment the largest end that that's involved trying to get inside the mind of Bruce Parrons. Eww. And yes this is possibly slightly less than desirable. We do not within the project have any desperately good rationale for why various points in the Debian free software guidelines say the things they do. And a lot of the time we can come up with arguments for why they do say that but it's difficult to justify some of those interpretations. And something that the question is should we attempt to interpret should we take the literal text of the Debian free software guidelines and attempt to interpret that literal text within today's free software social context. Or should we assume that what we mean by free software hasn't changed in which case we need to interpret them in the situation that we found in 1997. Which was a slightly more innocent and less hostile environment in some respects. Less hostile in the sense that we were largely being ignored. Exactly we didn't have people trying to screw us over and sue us repeatedly except the University of Washington. So one thing that appeared we have this the GPL is an example of a copy left license. It's designed to allow it to be the software to be used by anyone and modified versions distributed and used by anyone. But it's also intended to ensure that anyone using the software can gain access to the source codes of the software so they can change how the software they use works. Now this is excellent assuming that you live in the world where software is executed on the computer on your desk. It works less well if you never actually distribute the software that somebody is using to them. So say if somebody writes a application that sits and presents a web interface to something. If the user never actually gets hold of the server side of the software then under the GPL they have no right to the source code despite the fact that they're using the software on a regular basis. So this is sometimes called the um uh what's the euphemism? Web loophole or is it? Remote procedure loophole something like that. It's to refer to things where you communicate over a well-defined network interface. A well-defined network interface and as a result you can't complain that there's any distribution of the software. I think the interestingly this loophole is surprisingly similar to one that's that doesn't have anything to do with remote communication whatsoever. It's the uh it's the old argument about the GPL and using wrappers for um well for instance if you use the shell interface to go out and do something then you're not if you write a shell script that evokes GPL licensed command executables then your shell program doesn't have to be under the GPL. And that's that's a common interpretation but it's I guess the reasoning is that there's a sufficiently large barrier between the interfaces of the shell and the presumably C program that you're executing that the virality of the GPL doesn't hop over yeah it's hard to consider that's a derived work under copyright law though we'll never really be sure until somebody actually starts a lawsuit over it. But it's a lot of parsing, it's a lot of handling. Uh so this sort of brings us to the AFRO general public license which is a pretty much a direct copy of the GNU general public license except with one additional clause and that additional clause effectively says that if the software release under that license is intended to be faced over the network with the user and if that software contains functionality that allows the user to download the source code for programs. So say imagine a PHP script uh PHP content management system that's all your web pages are in that has a little button down in the bottom right hand corner that says download my source code. The AFRO general public license says you are not permitted to remove that functionality. Uh huh, you're me. Not misremembered, you've missed my own films. Oh sorry, right, yes. It's a good thing we don't let you write licenses. Well I hope that's where I'm writing the licenses then there was of course be checked thoroughly by Devin Legal for attempts to release any software. So uh and there will be unless we get distracted by one of our own internal claim orders. Well absolutely. So this is intended to ensure that software, the source code software is always available to the users of the software even if they're separated from that software by the void of IP of raving carrier. And I think it's probably reasonable to think of this as a fine upstanding attempt to ensure that uh the source code of software is available to more people and people who use it. On the other hand it's demonstrably a restriction on modification and it's a restriction on modification that goes beyond what the GPL allows. I can't take code out of the web server application and stick it into some random piece of crap because it doesn't have a web server in it. Well at that point it's not software that's intended to communicate to a network anymore so it's not an issue. It uses NNTP instead of HTTP. But then you in that case functionality could possibly be the Fusser will append a URL that provides the software to it. And if I'm correct you're allowed to change the functionality doesn't say that you have to maintain the code or the method just a way to get the. Yeah I think the idea is there's every person who interacts with the software should be able to obtain the source code. Right so and that's something that doesn't generally present a technological barrier. So in that sense it's a it's similar to some of the sub clauses of section two which say if the program is interactive you must clause and be printed a banner or you know you may not remove an existing yeah set of code that prints a banner identifying the coding that's licensed or whatever. Yeah. So if you can rewrite it you can port it to a different programming language but you have to preserve the existing functionality. It's the functionality that's important rather than the specific implementation of the functionality. Another point you might mention about the affair of public licenses I understand that it was created with the blessing of the Free Software Foundation. Absolutely. Which has caused a lot of speculation about GPL v3. Well the Free Software Foundation science explicitly states that it's a non-GPL compatible free software copy that license but is expected to be compatible with GPL v3. And since traditionally GPL does not allow any restrictions that's in the presence of GPL itself it does seem strongly implied. So yeah it's worth pointing out there that it also if it provides the functionality means that if the software does not include the functionality to to to someone to download the source code of the network it doesn't you don't you don't have you don't have to you don't have to include it. It's just like the interactive banner clause in that sense if it doesn't start out with one you're not forced to add it and modify it. So it's probably the case that even if we accept clauses like this as free then it's possible that somebody could provide a mechanism that was sufficiently obnoxious that it resulted in you being unable to usefully make the right words of that software. And in that case that specific software might be non-free but then we also have people who attempt to interpret certain clauses of the GPL in such a way that it's blatantly not really free. I think there are pathological interpretations of most licenses that's allowed us to twist it into being non-free software even if fundamentally we agree that this software should be a basic concept behind this clause should be free. And that's not a problem when the person doing a pathological interpretation is some random devian developer. It is a problem when the person undertaking an interpretation is the copyright holder in a piece of software they're trying to distribute. In which case we can always either send suits to us or not ship this. In general I think going for the not-shipping is probably a better option. Anyway Paytons, the Paytons weren't too much of a problem in 1997 when it came to free software though they are explicitly listed in they are explicitly mentioned in the GPL which states that it's not permissible for the fact that you can provide somebody with a copyright license doesn't mean that you can distribute GPL software if you have a restricted patent license that you're enforcing which would mean that somebody could not exercise all of their GPL rights but that's about as far as it went up until fairly recently and then suddenly people started threatening to sue free software projects over their use of license of patents and strangely enough as often happens the free software community has adapted by including clauses that attempt to prevent this sort of behavior since the behavior is considered harmful to the free software community and therefore we don't particularly want people to have the freedom to sue us for patents infringement for those parts at least we should make it awkward for them to do so there should be a cost to suing us for patents infringement and in existing software licenses there are two main clauses that deal with this sort of situation one of them is if you sue us for patents infringement then you're right to uh where we are the people who give you a piece of software yes so if i as someone who gave you a piece of software get sued by you for patents infringement you no longer have the right to use any patented uh any patents that i hold patents for regarding that piece of software and so i get to counter sue you for patents infringement if you want to carry on using the software assuming you have a patent with the force yeah uh which yeah yes so how strong this is depends upon how because a set of patents the free software community holds and how they are infamous by specific piece of software but as an example that way it's just like her work for proprietary software that companies share software and they if you sue us we see you because monthly they have patents it's a problem for us that we don't have we don't want them in the uh proprietary world uh companies mostly deal with the patent situation by in hinting at mutually short destruction yeah uh if apple and microsoft got into a patent battle then it's likely that neither company would be producing greater the software for the next 25 years or so so lotus and borland fought their way to the supreme court three times over software patents in the united states doesn't always result in the desired outcome but effectively if we don't have any patents it becomes difficult for us to really survive in this world but that's also appearing on top of it the other source of flaws which is a bit stronger is that if you sue us for patents infringement uh you lose the right to use the software full stop which is unfortunate say if uh you're if you're distributing it um it's unfortunate if you integrate this heavily into your own proprietary software and it's slightly irritating if you've just shipped 200 000 copies of it to your users desktops uh now basilar includes both of these clauses effectively one of them says uh the first one here the one that's strictly related to the patent license is only invoked sorry is invoked if you sue anyone for if you sue basilar foundation for any painted infringement whether it's a specific piece of software or not uh then you lose any patent license the basilar foundation has granted you so if you sue over thunderbird then you lose your patents granted by firefox as well what if you sue over a business benefit like like you guys are cleaning your office in the wrong way i mean like no in my proprietary way which is good and i mean seriously seriously i can't actually remember i believe that the npl tries to limit it to software type situations so i can't remember when we have a game and try to get that license parsed before the end of the talk uh whereas the termination of the copyright license is only triggered by um you suing over that piece of software so if i sue you over paint infringement by firefox i lose the copyright to firefox but not to any other works that i may have obtained there's a kind of logic to that because the the penumbra of the copyright protection only goes as far as the work itself whereas patents go everywhere yeah so this sort of clause also appears in the apple public source license and the apple public source license says if you sue us for patents infringement you uh we terminate your copyright license to all software that we've given you under this license so then they jump out of the penumbra and say right um the downside again is um the apple public source license uh doesn't restrict stats to software paintings uh with the result that if i say uh claim that apple has violated my patent on their case design i suddenly lose my rights to use uh various chunks of aksl software release my apple uh i think that's probably going a bit far especially since while we will in general agree that software patents do have a somewhat chilling effect upon us the broader idea of patents uh the painting of hardware designs and the like is not something that there's anywhere near as much consensus on the evilness of but for the most part patent holders often really do want us dead and so as a result of that it's not surprising that people within the free software community are seeking to find ways of protecting ourselves from them the problem is there's no good consensus on which of these things are acceptable in free software and which aren't deviance certainly not taking lead here the open source institute has explicitly said that uh they don't really care and the free software foundation have maintained a wonderful silence so arguably this is something where we should be making a stand but that would actually require us to agree on something sorry a few just fans according to um section 1.10.1 the definition of patent claims in the npl it's a patent a patent claim is blah blah blah including without limitation method process and apparatus claims and the patent licensable by the grantor and I scanned the part about termination it did not further restrict the application of patents so the mozilla public license is completely neutral as to regards what kind of patents are okay okay so uh painting licenses um complete lack of consensus and uh general depth of universe protected unless we actually do something let's say it's deviant I try to do lots of things and um maybe we'll actually do ones that usefully let's have multilateral talks with the website and the fsf you can get those two to sit now so the fsf so so it's worth knowing that the fsf um is now kind of their legal department has basically been spun out into the software freedom law center right and they're employing the person from the public who started the public patent foundation um and there was lots of experience doing stuff for patents so maybe it's just something we can get an answer from uh jan robertser over what's he pleased to be acceptable yeah yeah I mean most of what public public pat does is doing things like storing you know basically storing software software patents for three softwares you can get into that when the shit hits the fan right you can get it we build our own they can say listen we've got 10 patents and they're fucking useful and you need them and you and we're gonna have to make you know we've got a patent done x or so pay up right or we'll actually leave us alone is the uh trademark licenses this has come to the floor with firefox lately the mozilla foundation called a trademark on firefox as related to the field of software products uh in the united states at least possibly some other jurisdictions that matter they call the trademark on the word mozilla as well but it's not clear to me what the scope of that one is so the idea of this is to say if you modify this software uh you are not granted a trademark license and so if you continue to call this firefox we can uh send you cease and desist letters and the idea of this is to prevent situations where people impair a brand by taking advantage of the free things granted by free software license the specific source concern that the firefox people worry about is that anyone can take the firefox source code and integrate spyware into it and then put it up on a download site as firefox and they would prefer five versions of firefox that do bad things to not be called firefox because otherwise you'll have people setting the homepage to guptex for instance uh that's not yeah i couldn't use so the idea is not to prevent somebody doing this because obviously there are cases where they might well be say um if i'm doing a study into uh user behavior when using web browsers that would make sense of my modified web browsers to track what people are doing while they're using it uh but that's crazy science of its setting uh it's not really possible to prevent it in a free software license people need to have the right to modify software whatever it is that would be a field of endeavor restriction yeah people need to have the right to use software uh to modify software to do things that you didn't expect them to do and may even not particularly like them doing so the idea of this sort of trademark license is not to prevent people taking advantage of the reasons granted by software it's to ensure that the user can always tell the difference between a product that's upstream produce and a product that's downstream produce that is drive-worth of here it's not and there's been some argument over whether or not this should be acceptable in free software doesn't i tend towards thinking that it's this irritating um but in a situation where we can provide software given a specific name given the same name as upstream then that reduces user confusion and is probably a worthwhile goal because it doesn't actually prevent the user from modifying that version of sales providing it is not too difficult to rename piece of software yeah so the mozilla foundation in this case have basically said that they will provide a set of instructions and if you do everything requested by those instructions then that will be considered uh sufficient to satisfy their concerns they won't sue you over any remaining instances of the one at firefox um just to make sure i understand correctly uh it's an old even from the old days it was like university of washington has these clauses that you you need to to make it to change the names but this is extending is you cannot even use the name is that correct you cannot so um represent your work as the uh fine license for instance requires locally modified versions to carry a version number that was different to make it clear that it was a locally modified version that was perhaps in that case more to ease support burden so if somebody reports to bug to the university of washington with the version number make immediately say this version is being locally modified so this goes a bit further if it says you can't use the name but it's only bounded by trademark law you can use the name in any way that is not confusing that does not seem free to the trademark license so the fact that i am not allowed to call it firefox does not mean this i am prohibited from saying we built this software on top of firefox in an interview with the press that's true in the same way that uh the fact that force holds trademark on the word force doesn't mean that i can say that i used to own the force so it's just to clarify actually what what trademark is about sorry so so what they are saying is that's the standard of trademark business uh default state is that uh you are not permitted to use a trademark term yeah most uh a trademark term in areas that would be confusing this way that's not generally an issue with free software because very few pieces of free software use trademark terms we're pointing out that debbie is the registered trademark yeah debbie is the registered trademark you are not permitted to call something debbie in in various cases uh in some other cases i believe you're allowed to use the debbie in trademark where it relates to debbie in so there's some i'm going to take where that do i think but descriptive you can't you can't necessarily an as non-profit there's a number of other things a number of people do use the word debbie in places where they probably shouldn't and you know at a soccer notebook and just as a lawyer and goes around and there was the case of the uh what was this trusted debbie yes that was didn't necessarily present a very flattering image of debbie by implying that what was untrustworthy and so i don't think we actually in cases any sort of formal proceedings but it was suggested that maybe they might want to reconsider their name and i think it's the dpl at the time who sent them a nasty frame and that they changed it yeah and i think there's more than they like uh and they changed without any argument it's so much better but okay they changed without too much open arguments and unwillingness and have sort of be grudged it ever since is jeff look we in here if not okay um we we kind of fought our way to a a kind of principle on debbie legal with regards to the latex project public license because what they wanted to do was enforce a restriction wherein if you modify the file you had to change the final name and we fought that pretty strongly and they'd actually get their agreement uh not to to try to push that into the license because the reasoning is is that this is uh is that you can use trademarks or things like trademarks to prevent people from passing off a work as something that it's not but you can't use it to uh to impose restrictions on modifications of computer interfaces and a file name is a computer interface so uh i think we would have a problem if the mozilla foundation were very strong without saying you can't call you like user bend firefox is not loud you could call it user bend firefox and then when it comes up it says you know debbie and ice legal yeah and yeah in the band because because then you're communicating directly with the user more or less and not trying to communicate with another software interface and that's an important distinction to keep in mind and i think it's one that has rough consensus on the debbie and the mailing list yeah at least hasn't a year or so again the mozilla foundation suggested that they don't object to say binaries having right that's name uh as long as in the obvious user interface it's not referred to that way so in the graphical environments they would prefer us to appear with the ice legal term rather than the firefox term in the menu right but that's when launched if being launched by other software that's other software would be allowed to launch it by attempting to call firefox right so it presents the same software interface but not necessarily the same direct interface to the user there there has been at least one court case in the united states where somebody was used was embedding their trademark termed inside of bios or something and so when somebody was uh creating a clone uh the presence of the trademark termed in the bios was was used as an instance of infringement i wish i could find the case for this i've looked for it at some point in the past month but the court ruled that because the presence of that string was used by other software to configure itself and the purpose of copying the trademark term was to provide interface compatibility that the trademark licensor could not initiate trademark infringement proceedings so it's it can be dangerous to a trademark holder to wed the use of their mark too tightly with technological dependencies upon that i believe that case may actually have been of the nintendo entertainment system and people producing unlicensed cartridges for that possible uh sony did something similar with the cop protection for playstation there's a section of codes on playstation cds that is copyrighted and represents their trademark in some way so they can attempt to use as many cease and desist less as possible if they're not all enforceable yes even more so in fact um but part of my day job i would ask you to get involved with such a case apparently the playstation will not start up unless it finds certain playstation graphics on cd and of course they're copyrighted trademark for everything called graphics yeah so uh people have tried to use trademarks to inhibit uh people's ability to make interoperability uh i think people attempting to do that are probably not on the side of free software i think we may have some reasonable degree of consensus there so since 1997 that the world has changed somewhat uh we have many more things that have been used to attack free software we have people who will deliberately attempt to use every loophole they can to take advantage of what we have risen without giving anything back and that's fine uh in some cases obviously the bsd license specifically is written in such a way to encourage this sort of behavior but there are other licenses that are used by people who would prefer this sort of thing not to be allowed and inevitably when we try to protect ourselves from more of this sort of behavior we impair the ability of people to make certain types of derived words or to use the software in certain cases and we need to accept that there is a balance here it's always possible for people to come up with new ways of attacking free software and if we want to protect against those then we may need to prevent people from being allowed to do certain things that would in an ideal world be allowed so the gpl is an obvious example of something that very much restricts the rights of derived of derivers uh in order to protect the community as a whole it says you may not embed this software within software if you're not going to reach the source codes of all of this under the gpl and if say we imagine a hypothetical world where the gpl had not been written in the late 80s if in 1997 uh Debian was based almost primarily on top of bsd license software if the learners kernel in fact were under a bsd start license and then we've got to this point and somebody came along with this copy left license uh maybe it wouldn't have been registered or perhaps i don't know who things would have worked out differently and so if somebody today presented the gpl to Debian as a new software license of license which had never been seen before do you think there's any remote chance whatsoever that we would look at that and say yes this is a free software license i think we would make them get rid of some of the flaws in section 2 about the banners that we have to keep to be honest i think we'd be quite dubious about this idea of imposing uh source code availability who's we well we various people with it wouldn't be here in the first place it's unlikely that we would be successful enough to be here in the first place because we'll be running microsoft winix yeah that sort of thing so it's been something of a problem but the instinctive response to any new license is to look at ways as it restricts reasons beyond what we consider acceptable without looking at what benefits those restrictions bring and possibly times come to actually start thinking more strongly about where we think the line should be drawn in terms of the balance between restrictions and freedoms so that's basically everything i had to say as there anything anyone else to want to bring on just stretching though okay no i know i'm sure um one thing one thing that is really uh kind of hidden for me is in a lot of conversations with with i wouldn't know if he'll always say things i'll say well you know this part is really annoying it causes this to happen and all this thing says is this a freedom issue and you have to think about it right because i think that that often often we confuse the less than often situations or the plain ugly or confusing situations or the you know any number of different problems that a license can have and there are lots of problems that a license can have that that are not necessarily freedom issues right there can be there can be some license that are suboptimals in a range of ways that that ultimately do not take away anybody's freedom or put it i don't know against the spirit of free software and i think that's something we need to kind of ask if it's the free software is not the freedom to do absolutely anything free software is the freedom to do some things maybe a large number of things but not necessarily everything well let me rebut make this point a little bit the concern i have with licenses that are written like shit uh even if only in some respects is that is that we have to remember what's the goal the goal here is that we want to have freedoms and we want to have those essential for free software freedoms and if you have a license that's that's full of ambiguous or unclear language what does a careful person who's trying to do the right thing when they're modifying free software do they read this and go hmm can i do this or not well i get sued it's not really clear i might have to pay a lawyer and they just won't exercise their freedoms even even the ones in the subset i mean not the freedom to do absolutely everything but even the freedoms that they're supposed to be guaranteed under the fdl so you know i want to i want to do better than having a world of licenses that are technically dfsg free i want to have a world of licenses where people can go forth and exercise their freedoms with some confidence that they're not going to be squished by you know some lawsuit because that that threat is real and they're going to have to cope with that on a rational basis anyway um if license proliferation and the proliferation of poorly worded licenses increases to the point where they you know are smothered by irrational fears uh then we're not really servicing the community as we talk to real corporate lawyers about the gpl i have not they're incredibly terrified of it well many of the crap ones are and the problem is that you can tell people that their fears are irrational but they're irrational and the reason that they have these fears is because they're not thinking clearly and very often you coming along and trying to you know explain it to them isn't going to help but who is more who is more likely to make a persuasive case joe q garage developer or the deviant project i mean 10 years ago we were joe q garage developer now we have a little more force we have a stronger voice and we know other people in the community who can help to make a persuasive case let me give you a good example the creative commons people have been very indulgent of our you know this picky proofreading of their licenses and it looks like that they're going to be like licenses are very you know favorable to us by the people that you all not consistently much more you know most of them you know we're interested in talking to um compare that with the people who are scared and don't understand what's going on most people are all those people who aren't connected to our community and they've never heard of us but they have heard of us they see this as some kind of weird hippie crowd and some of that people are the country's centre-lines so you turn up to a UK court and you go to the UK and they'll go well this phrase doesn't mean anything because the right work has though been a warning to the court i think it's clearly the case that proliferation of software licenses um does nothing whatsoever in most cases benefit free software and the open source institute must take actually OSI initiative initiative yeah have actually started taking a stance against this which is a surprisingly useful thing that we've done because like i can't remember doing anything else welcome never know yes and that's i think is undoubtedly a good thing because okay we don't particularly care about what the OSI does but uh lots of companies really want to be able to take that little OSI approved license tick box when they're saying so are we going to be able to say we're engaged in open source so one of the things that i think is really interesting is when we look at the body of licenses that existed back in the BFSG was written it was pretty easy to determine which within that set were ones that we thought yeah fell on our side of one which ones we thought fell on the other side of our body and in fact the set that's explicitly referenced in our cause channel yes in clause 10 was not particularly arbitrary it was in fact at the time you know almost a complete list of licenses that we thought were reasonable at that time and what's been intriguing to me is that there's this combination of the facts and pressures that over time have caused real licenses to appear there's really only been one sort of economic model change in licensing since then that was at all interesting for free software community and that's the difference in sub rights on the Zillow license objective other than that in almost every case what we have seen is a proliferation of licensing subtleties that have made it harder and harder for us to figure out which side of the license policy so just blaming Debbie and legal for getting more argumentative doesn't doesn't paint a complete picture the the market has evolved to the point where we're being asked to make more and more difficult decisions yeah with the issues at hand are more subtle and harder for the average to bring it software professional that's why that's one of the reasons I wish people would cut legal some slack because it's being asked to do harder and harder things all the time the reason for this is obvious though it's because nowadays getting your license approved by Debbie and or OSI as free software as well has a lot of value even if your license is not really free software or you don't really care about free software or anything like that and so lots of people are very motivated to screw everybody over and this is one reason why new licenses are you know treated as a great deal of suspicion but on the other hand you look at these licenses and the problem in most cases with you know if we're not talking about giant weird shit licenses from microsoft um or sun um often yeah or apple but sometimes we you know we've got some random projects out there and they wrote some pile of nonsense or they you know with the best of intentions and what those people need is you know help i think at this point we're out of time so um i'm sure we've had ample opportunities to continue this discussion at length in a variety of different ways as uh thanks well i should point out that when i was ppl i tried to get people to agree to a reparative