 Hello and welcome to my talk. Today I want to talk about the Debian games and free software games in Debian and the status of games in Debian in general. Where am I? I'm a Debian developer from Germany and I've been using Debian since 2002. In 2012 I started to contribute to Debian and one of my first packages was a game of course. At the same time I also started a little project called Vinox UAD where you can play multiplayer online games from the Debian repository including games such as Open Arena or the Battle for the Wiesnacht, Team Worlds and so on. I contributed to, as I said, I started to contribute to Debian in 2012 and at the moment I maintain or contribute to around 500 packages mostly games and Java packages but also to the long term support section in Debian. So what can you expect from the next 45 minutes? First of all I want to talk about the fundamental differences between commercial games and free software games because most people think that commercial games are the norm, which they are, and they don't know very much about free software games. So they ask, can I play a game X on Linux? And 10 years ago I would have said, it will be difficult. Nowadays I tend to say rather yes. It is possible. It is definitely better than it was 10 years ago. I also want to talk about how you can find games in Debian that's not as easy as it sounds. There are multiple tools available for you and there's also a project called Debian Games Blend. I want to explain why we put some games in the main repository in Debian, some in Contra and some in not free. What is the difference and why are they there? Why are there games which we cannot package at all? I will tell something about new games in Debian 10. Well, at the end I also want to explore some new ideas, what we can do to improve the status of Debian games. If you look at the global game marketplace you will see that they deal with thousands of games which are a million dollar business. Many of them attract millions of players around the world. There are big titles, blockbuster games, so-called AAA titles, which consist of, for example, Dota or Counter-Strike, SimCity, Call of Duty, Overwatch and so on. These games attract a lot of people and they make a lot of money with it. I've told this before, last year in Montreal there was a project, it was called the next generation multiplayer online role-playing game by Blizzard Entertainment. They decided after they spent three years developing the game that it was not good enough. Before they made a decision, they employed about 100 developers and creators of artwork, software developers, who worked full-time on the project, and then made this decision, okay, it's not good enough. And they took the assets, all the images, sounds and what they have done before and made a new game. That's called Overwatch now, which is quite a popular game again. Blizzard Entertainment gets it mostly right all the time. They have four or five franchises, but they always attract millions of players and make a lot of money with it. So if you take a look at FOSS games or free software games instead, you will see that the revenue of these games is close to zero. Most developers receive no money at all or receive no donations. And some of them, if they are lucky, they can sell their game on Steam or very, very, very few can attract enough people to support a crowdfunding campaign. For example, ZeroAD tried that and they could raise some money, but that's the exception and not the norm. Successful games attract thousands of players, for example, well, MindTest is a very popular game nowadays. It's a clone of Minecraft, but it has a very active community and obviously it's not as big as Minecraft, but it's still, compared to other games, a very active community. You also have to say, OK, FOSS games are a niche. That's the truth. It's not as big as commercial games, of course, just because nobody funds those projects. These are hobbyists, developers who create a game and they are free time and they are spare time. But yeah, they make no money with it. So you can't expect the same amount of quality that you see in commercial games. When I talk to people who only know commercial games, they are a bit disappointed. But you have to see this in perspective. Well, there are good arguments for free software games. The most important one for me is that we have the source code and also, in most cases, the right to modify images, the sounds, all the assets. So that means even in 10 years' time it is possible to develop a new game or to modify the game. Or as we have seen in the past, it is possible that we can maintain games for 20 years or longer. For example, FreeSIF was developed in 1996, I guess. And still to this day, 20 years later, it is still developed and there's a vibrant community around it. And that wouldn't be possible if sources were closed. It is also a great way to learn about developing in general and it is not only meant for developers, but also people who are not into coding but more into creating artwork can share their work. And it is easily accessible for them in my opinion. So this artwork is reusable for other projects. And that's also not possible with proprietary games, of course. So if you think about it, there are game libraries which are not only used for games. They can be used for everything and a special for projects like the Open Robotics project. There is a story I packaged once a free Orion and I needed bullets to complete the packaging. There is a 3D multiphysics library which makes it possible to simulate deformances. And we have compute different objects to hit each other in space. And someone from Open Robotics contacted me and said, hey, cool, you have packaged bullets and we use it for our robotics project. You can see that not only the libraries are meant for games, but can also be used for different projects. And that's quite cool, so we are not only packaging for a niche, but it can be reused. Another important aspect is that it is excellent integrated into Debian, which you not always have with commercial games. So we apply the same rules as with other programs. And, for example, we are very careful when it comes to licensing. We review not only the code, but also all the assets. And that is a very time consuming task. But it also is very important that we only have software and artwork that is free available. Yeah, major games communities. I mentioned before, MindTest and The Battle for Bison, but there's also SuperTaxCard, Xenotic and COAD and FreeZiv. These are the biggest communities in the free software world at the moment, I think. And they are between 10 and 20 years old. MindTest is a bit younger, it's only 5 or 6 years old. But they are very active and they still produce a lot of new content these days. I have counted 600 different games and emulators last time I checked. That's the number of games, free software games in Debian. And all software and artwork complies with the DSG. Most games are written in C and C++. We also have some Perl games. Most important one is Frozen Bubbles. One Haskell game is Raincat and a few Java games. We are very proud of that. We also have D games, which are quite rare. And the maintainer of GCC suggested that we support them to Java, because he hates to support D, but they are very great. And if you want to look at them, you can find them in the package games, like iPhone, shoot them up. And, well, they are really fun, I think. Popcorn reports more than 2,000 installations for popular games. Is this a good number? I think yes. Because even more popular applications outside of the game universe can have much fewer installations. What I should mention, too, is that we are talking about PC games mostly, because web and HTML5 games are quite rare in Debian. As I know, there's only one game in the archive which could be possibly played in a web browser. That's Cave Express. But it's quite hard to sustain the tool chain. So you need to maintain a compiler that compiles the C++ code into JavaScript. And this hasn't worked all the time. So, yeah, if you're up to the task, please do it. So one of the most difficult problems in Debian is how can you find games? I say this because 10 years ago I tried to get an overview of what games we have in Debian. And this was not very easy, because most people suggested to use the terminal and to use tools like app or aptitude to find them. And, well, not all users are into this kind of stuff. They would prefer something more graphical. So nowadays I would suggest that they use tools like Synaptic, Known Software Center, or if you are into KDE, KDE user, you could also try Plasma Discover or Muon. Yeah, why do I say that? It is... Most people think that users of games are also hard-core Debian users. Most people think Debian is about servers, or it's about this desktop environment and this desktop environment. But people who want to play games are most often very casual users. So they don't want to spend much time to find games and just to relax and enjoy themselves. Yeah, we should make it definitely easier for them. There's another project which I mentioned before, it's called Debian Games Blend, and I come to that in a few minutes. So let's take a look at Known Software Center. You can see a lot of symbols, you can see game names and game descriptions. What you can also see is that some of them are translated and some are not. So the language, which you cannot understand, that's possibly Klingon, and you can see that, for example, the game Killbots has a translated text and, well, then there are others which are in English. So this is a bit confusing. Why is that? Why is it so? Those information are derived from so-called desktop files or AppStream data files. And most games do not ship such AppStream data files, so we derive all the information from desktop files, but they are poorly translated. So most of them are only in English and few of them contain German comments or French comments or Spanish comments, but I know of very little games, for example, who are translated game descriptions, which are translated in Chinese. So in an ideal world, you would see that all these icons and all these games would show Chinese translation text, but we don't have that yet. So there's a lot of work to do in the translation area and at the moment we are not there. What you can also see is the difference between icons. Most of them look quite good, I think, crisp and clear, but take a look at Corruptor or Lure of the Tempris or, for example, Magikor, the game on the left. The resolution is very poor or they are just screenshots, so they can be improved because people tend to play games or are attracted to games with nice icons, so more eye candy is always good. But if you don't have a good eye candy, people mostly think, hey, bad eye candy means it's not a good game, so here we can improve too. The Davy games blend. That's something I started in 2014, I guess, and that's just a collection of meta packages. So if you wonder, okay, you know the tools now, GNOME Software Center, Muon and so on, and you can also use aptitude. But then there's another alternative, you could install just a meta package. If you're into games chess or games cards, then you can just install the corresponding packages. If you are more into Tetris games, you would install games high in Tetris. Or if you don't know what you really want to play, I suggest to install games finest, which contains more than 100 games, that's a lot, you know? But in my opinion, you cannot narrow it down to 10 games, which are good because people tend to say, hmm, I like strategy games and the others say I like road-playing games and there are other ones who like first-person shooters. But those 100 games are of high quality, I believe, and some of them should be enjoyable for everyone. That's our platform page at the moment. As you can see, it consists mainly of the game description and two links to the packaging at tracker.deben.org. In my opinion, this site could also be improved, but as usual, we need more people who want to invest time into it. So we not only need people who try packaging, but it would be nice if there were more people who would like to design a new page, for example. And I think if it is done well, you can definitely promote first games better. But in the meantime, I think it's a solid okay thing and it's informative, definitely. Well, since last year, a lot has happened. We've got some new games and we had to remove some others. For example, G-Tatrainet was removed and a special game, which I liked a lot, OO Lights, that's the clone of Elite, was also removed because the maintainer couldn't provide support for an embedded library, which is called Lipmoss.js. There were security implications and AppStream didn't want to remove the embedded library. So he decided, okay, I didn't want to do all the work and it's enough and well, it was dropped. And there are other games like the aforementioned G-Tatrainet, which depended on GNOME 2 libraries. In this case, we cannot do very much because nobody is interested in porting them to GNOME 3. We even got a response from one of the AppStream developers who said, wow, it was cool that you maintained my game for 15 years. But nowadays, well, I probably should rewrite it and port it to another language or something like that. It's a shame, but that happens all the time and it would be cool if we could improve on that. But, well, not enough people. And another game we had to remove from Debian that was a sprite editor, the last game on the list. And the author decided to change the license to a non-free license so there was not much willingness from our side to maintain it anymore. On the other hand, we... Some other people packaged new games, including DD-Dat, which I've heard is a popular game in Taiwan. I believe one Taiwanese contributor even packaged it. There are other games like Icebreaker, which were reintroduced. So this is a good example for not... You shouldn't worry too much about if games get removed. You can always reintroduce them. And Icebreaker is such an example. There's also Lugaro, which I can recommend. Lugaro was formerly known as Open Lugaro, but the name is a bit misleading because it wasn't open. The assets were still non-free. But this changed last year. And Wolfwire Games decided to change the license. And so we could introduce it into Debian Main. Well, now I've talked about Debian Main all the time, but there are also games in Debian Contrib. So why do we ship games in Debian Contrib? Most of these games are game engines, which are free software, but they don't have any free assets. So in order to ship them in Debian Main, we had to either create new content, or had to cool wins upstream to change the license. Nothing of that happened. Examples are L-General, Q2 OpenMW, and MRB. And for L-General and Q2, I created a new data package. So those engines on those games could be moved to Debian Main, but OpenMW and MRB are difficult because they are commercial games. These commercial games are consisted of, I guess, several gigabytes of data, and they were never made free. So in order to play them, you have to download them somewhere. Well, why are we so picky about that all game content must be free? There's actually a debate. Popular men, which is so unsaid, since the art in the game is not software, it is not ethically imperative to make the art free, though free art is an additional contribution. Some in our team disagreed with this view and said that it is imperative that for a game to function, it also requires data or software images, and they must be free. So without those assets, the game wouldn't function. And that's why we decided to put them in Debian Contra. The engines are free. But the assets are non-free. In my opinion, we should use Debian Contra as a staging area and work on it to either produce free content or take free content from other sources and rearrange it and create our new packages. Sometimes a demo version would be enough. It doesn't have to be always a complete game. Then we came up with another solution, better Simon and Alexander. Simon developed a tool which is called Game Data Packager, and it is written in Python 3, which is able to download content from different sources, mostly good old games, gog.com, or you can also use images from certain CDs or DVDs. And the tool just checks, verifies, is the data valid, and it creates hash sums of files, and they are stored in YAML files. And we have currently 215 supported games, and you can just easily use a command line tool and if you own the original content, then it will create a depth package for you. Why is it useful? Most people want to have a clean, tidy system, and if we can package it in a Debian package format, that's definitely more favorable than just putting those files in your home folder. So last but not least, non-free. Non-free is evil. Packages must be placed in non-free and compliant with the DFSG or are encumbered by patents or other legal issues that make the distribution problematic. The minimum requirement is they should at least be distributable. Most non-free games are assets game data packages, but we also have or had some games like ZengBand, I don't know how it is pronounced correctly, that's a fork of ZengBand, a very popular rogue game. The only difference between ZengBand and ZengBand is, beside the rules and there are more characters and more possibilities to play the game, is that they kept the original license which forbids to make money with the game. That means it is non-free. So if the author of ZengBand tried to convince the original author of the code to re-license it like the ZengBand folks ZengBand folks did, then it would probably be in Maine nowadays. But it didn't. And then we also have Steam. Yes, I should mention Steam, of course. This is the most popular non-free game package in Debian with more than 5,000 installations. Steam was definitely the game changer for Linux. I guess it was released 2012, Steam for Linux. And as of today, according to SteamDB info, more than 3,208 commercial games are supported on Linux. So they worked natively. Should we care? I definitely think so, because Steam is not only the largest digital distribution platform for PC games, but it also attracts new developers who want to create new content that works on Linux or they want to make their existing games available on Linux simply because there's a market. And as I don't know if you know the Humble Bundle project, they sell bundles of computer games and they make it mandatory that they should also run on Linux. And compared to Mac OS users, Linux users are less than Linux users, but they spend more money on those games than Mac OS users do. And that was a very important finding. A few years ago, people thought, oh, if I port my game to Linux, I can make money. That's obviously a good thing and more people joined the train and it goes forward. Well, Steam is not the only digital platform, but it is definitely the most important one. There are other gaming platforms, for example, Lutris, which is not in Debian, but it's free software. And the goal of Lutris is that you have a unified interface where you can just click on a game and then it is installed with all necessary tools like emulators or content and it just works. Or if it is a Windows game, it would install Wine, configure the game and you're good to go. As I said, it is not in Debian yet, but it looks promising to me. But we also have another package which is called RetroArc. It's quite similar. It is also a unified interface and it focuses on emulators and other media players. It's very similar, but it's already in Debian, so you should definitely check it out. I've sponsored PlayIT a few months ago, or two months ago. That's another project which tries to simplify packaging non-free games, which are either DRM-free or not DRM-free. Most of these games are from good old games or Humble Bundle. And the only thing you have to do is install the package, run PlayIT, name of the game, and then it will... I'm not sure if it will download the package automatically, but if you have the executable file, you can just type PlayIT, name of the program, name of the game, and it will create a Debian package for you which is then globally installed on the system and you can play the game, which makes it very easy even for casual users to keep this system tidy. Then I should also mention Wine. Wine is not an emulator. That's the acronym. And Wine made it possible for me in 2008 that I could play World of Warcraft on a Linux machine, which was, well, I was the only one in my online clan who played on Linux in those times, in those days. Nowadays, it is much more common. There's also a company which is called Crossover that provides commercial support for Wine games. So you have your Windows game and you can just install it on your normal Linux system and it will run most of the times. For example, if the game is supported on Mac OS, like World of Warcraft was, it will also run on Linux. That's the experience I have made. Another good way to play Windows games on Linux is to use Play on Linux, which is also front-end for Wine and it is free software, but also in control because it downloads non-free assets and non-free stuff, so we had to put it there. But it's a very good front-end which makes it also very easy to install Windows games on Linux. Who are we? The Debian Games team is mostly responsible for most of the games in Debian, about 400, and we are a bunch of guys who try to maintain the older games and try to package new ones, but we are short on manpower, women power these days, so we can always welcome new members. I myself joined the team in 2012 and it was quite difficult in these days because I couldn't find any sponsors. Nowadays it's a bit easier because you can just go to mentors.debian.net and if you are lucky you will find someone who can sponsor your packages, but nowadays it was definitely harder. How shall I put it? We try to integrate games and try to make it better, but it's very simple. If you want to contribute something, don't think about how or what should I do. You should have a clear motivation. So it's not enough to just say, okay, I'm here and I want to do something, that's a recite for failure in my opinion. You should have a clear opinion of what you want to do. You don't have to be a coder. You don't have to be someone who writes a lot of codes. You don't need to be a programmer. We also need people who create icons, as I said before, because they improve the visibility of games and we need people who write documentation, we need people who translate descriptions. Like I said, if your users can't read the descriptions, that's always a poor experience. So we should try to translate more game descriptions and that would be nice. So you don't need to be a programmer. Don't be afraid to do the seemingly lesser work in Debian. It's important, very important in my opinion, to make games look good. What else can we do? First of all, if you don't want to contribute on a regular basis or if you're afraid to do it, just block about your gaming experiences. Just play the game and tell people about it. You can also package new first games, but you also appreciate if you fix the bugs in our packages. Well, there are a lot of them. Even more importantly, important is report bugs. What I often see is that we release a game and then a few months after Debian was released some people report bugs which we couldn't detect before because those are runtime bugs. It is very hard to find them. We need more people who just install a game, try it out and tell us if it works for them. I've also thought about four new projects. One of these projects is improving our home page, the Blanz home page. We could also create a completely new home page at GamesDebian.org for example which would make it more obvious that we have a game section on Debian. At the moment it is quite hidden under the Blanz framework. Perhaps we could move it to a more prominent place. I also thought about the game contents database which would be similar to codesearch.net that is a Debian project for searching code but you could also create a database for image files or sound files and list them all in these databases and with their license and a preview of the image so that people could find new images which they then could reuse for other projects. Last but not least, a live image about Debian games would also be nice. People tell me we don't need it but I think it would improve the visibility of our game so if you could just hand out some images and say hey, here are 100 games which we support already in Debian let's try them. So far so good. I rushed to all my slides and I'm out of slides. Do we have any questions? My name is Andreas Tiller. I want to agree with you that Blanz pages are not very modern anymore and I would love if somebody would do some better design. This is all templated. You can start today and find better templates. For all Blanz, others would be profit as well. You might know that Ode Streicher has thought Debian Astronomy is some different design but for me it's too less information but you need to decide. Another hint is because you said you had trouble to find a sponsor in 2012. I started in 2013 the effort Sponsoring of Blanz it's just in the Wiki page where you can prove that you have understood the Blanz concept and if you have Blanz package I try hard to sponsor your package. That's a well known project and I knew about it back in 2012 but I tried directly to convince the team members of the gaming games to sponsor my packages and I was a bit disappointed while we have a games team and nobody of the games team sponsored my packages so I had to rely on Debian developers from outside the team to sponsor my packages and when Vincent Cheng became Debian developer he was merely the only one who sponsored my packages then so I had to rely on two people and it was a bit disappointing so I just remember it. Nowadays it is definitely better but I fear not everyone knows that we have a mailing list or that you can reach us on ISC so if people want to contribute they often find it difficult to find the person who they should talk to and I think we could improve that but I don't know how at the moment whether it's definitely something we could do better. Do you have ever encountered some kind of legal issues because just like you mentioned there are many game engines on the Debian package list but most of them may be reverse engineered or reprimanded from prior works so is there any kind of legal issues that they would try to sue the distributor because you distribute some kind of pirated source code? I assume these are all clean room developers I'm not aware of any game engine that is a complete rip-off, a.k.a. remake of an older game so they try to create something new something which is similar to the old game but also new there's nothing that is exactly like the old engine so for example Take El General that's a clone of Panzer General which was a game from the 90s developed by an American studio and I'm aware that a Russian developer well he almost created a clone of the original engine but he was able to sell the new game and still to this day he can sell this game and it looks absolutely similar to the original game but we don't have this problem that General is very different and our source code is visible to all so as long as this is the case and we know the author has created it and licensed it on a free license I don't see a problem for Debian but if we are aware of abuse or wrongdoings we should remove it because there are PlayStation 2 and PlayStation 1 emulators and some legal issues were some lawsuit was filed in the United States and if I remember correctly the EPSX I believe the PlayStation 1 emulator was taken down from several sites because the lawsuit back then was settled out and it was illegal to do this kind of source code I don't know, I can't say much about that at the moment so I can only reiterate if we are aware of wrongdoings we will remove it but at the moment I'm not aware of any game engine that is a problem for us I was told I can only answer one question but we can talk about anything later can you put the microphone close to your mouth please how's that? so in 2012 and 2013 the FSEF worked with Creative Commons and Mozilla and Open GameArt to run a contest called Liberated Pixel Cup and we raised I think it was just about $13,000 when I was put up for people to make free game art and free games to go with them as a contest and I thought it was a really successful project we'd love to do it again but we need more partners to work with I wonder if you think people would be interested in pursuing something like that we could announce such contests on our different media channels we have a new site we could announce it we should talk with our publicity team about that that's the way forward I think you need to communicate it through the community or you should post it to Planet Debian perhaps if you post it on a mailing list it probably gets lost so I think there would be interest in such contests but we must ensure that people get to know it or are informed correctly if you just released some information you probably never knew that there was a contest and are quite disappointed because they wanted to contribute something but they didn't know it I don't know at the moment it would be difficult to think but the publicity team in Debian is the right place to force that we were running out of time I gladly answer all the questions afterwards I would like to thank you for listening and for coming here see you, bye bye one thing Keith Packard is sitting next to me there is another talk about games and Steam and I can highly recommend it to everyone who wants to know more about Steam and the development there go to it