Please create an account to participate in the Slashdot moderation system


Forgot your password?
Education Programming Games

Ask Slashdot: Tools For Teaching High School Kids How To Make Games? 237

First time accepted submitter nzyank writes "The other day I bravely (foolishly?) volunteered to conduct a video game development workshop at my boys' HS. This in Smallsville, Vermont with an average graduating class size of about 20. The idea is to meet once a week and actually create a game, start to finish. It will be open to would-be programmers, designers, artists, etc. I worked on a bunch of AAA titles back in the '90s, but I'm pretty much out of touch nowadays and I'm trying to figure out the best approach. The requirements are that it has to be one of either Windows/XBox or Android, since those are the platforms that I am current on. It has to be relatively simple for the kids to get up and running quickly, and it needs to be as close to free as possible. Teaching them to use stuff like Blender, C#, C++, Java, XNA, OpenGL and the Android SDK is probably a bit much. I was thinking of something like the Torque Engine, but they want $1000 for an academic license, which is never going to happen. I simply don't know what's out there nowadays and could really use some suggestions."
This discussion has been archived. No new comments can be posted.

Ask Slashdot: Tools For Teaching High School Kids How To Make Games?

Comments Filter:
  • by InterestingFella ( 2537066 ) on Tuesday December 27, 2011 @12:34PM (#38503750)
    For programmers best suggestion would be XNA and C# as it is really powerful while still being to program with, and you get support to all Windows, Xbox360 and Windows Phone 7. However, you noted that even XNA is probably a bit much.

    However, MS Research also has come up with Kodu [] which is basically XNA and C# in even more suited package for kids. It's really easy to use and you can actually modify your game a lot. It's fully interface based, so there is no need for coding, but it is still fairly powerful and the best of all, you see
    • Re: (Score:3, Informative)

      by aretae ( 1631299 )
      I've been teaching programming for a bit...For kids' learning there's a pretty clear top-of-the-list set Kodu -- XBox -- my 5 & 7 year old enjoy making these games a lot. -- Scratch -- My teen has used it. Logo -- I loved it as a kid, and it has fabulosu learning, but low video-game capabilities Lego Robotics -- Very good for learning programming, less so for video games. Android programming seems pretty easy for kids (My teen)...can use any dev environment you like. Eclipse, Android SDK, Java, you'r
    • by Rhalin ( 791665 )

      I have to second Kodu. Very minimal learning curve, easy to make relatively fun games in a short amount of time. Options like Unity, Torque, and XNA are reasonable if you have the time to invest in teaching them programming on top of teaching them how to make a game (or need the advanced features, such as cross platform dev, which it sounds like you don't).

      With Kodu, you can focus on the game development and/or production, rather than the programming behind it. There are some limitations, such as being s

    • by fermion ( 181285 )
      I would say it depends if you want to teach coding or use of game engines. I would say in either case, unless students have taken a couple years of high school programming, some basic techniques would first have to be taught. Unless they know OO design, structured development, functional programing, and the like, that will have to taught these basics first. The average high school students does not have a strong relationship, with, for example, variables, strict procedure, and following rules. For insta
    • Kodu looks to be a little too simplified for high school students. In my opinion, it would be a disservice to college-bound high school seniors who are interested in software development to teach a course like this without giving them some exposure to actual "code".

      In my first year high school programming class we learned to program in BASIC by creating games. We started off simple with games like black jack and bingo, but by the end of the year some of the more advanced students had progressed to the
      • I think a bigger disservice is in fact focusing on code. Instead focus on what are the elements that make a good game? What makes for a good user interface? Physics (ie, reality) based or fantasy based and what are the difficulties of both? It is always better to think first (a lot) on what it is you want to accomplish than to jump in and start coding.

  • Unity3D (Score:5, Interesting)

    by claytongulick ( 725397 ) on Tuesday December 27, 2011 @12:36PM (#38503766) Homepage

    Unity is pretty much the best option. It is cross platform, easy to develop in, and has everything you need to get started fast. The documentation is excellent, the community is supportive and the entry-level version is free. Unity []

    • Boneheaded malformed link. *sigh*. Here: Unity []

    • Re:Unity3D (Score:5, Interesting)

      by NoSleepDemon ( 1521253 ) on Tuesday December 27, 2011 @12:48PM (#38503938)
      I agree with this, the University I work at runs a game development workshop for 12-17 year olds(ish) that runs for an entire semester, we bring in a lot of big guns from the industry to give them talks as well and next year we're thinking of giving some of the better developers internships at our studio. We've found that Unity3D is a pretty excellent tool for people learning to program, it's also a pretty excellent tool in general, and we use it for our commercial projects as well. I do of course have some gripes with it, for a start the interface is pretty awful (prefabs aren't at all intuitive and nesting them doesn't work right) and source control is a NIGHTMARE - you pretty much need to have the pro version which allows you to turn on the "make my unity project not cause my version control system to tear its eyes out" option or your project's associations will break each time you distribute a new build. - most of the youngsters won't care about that but you're almost guaranteed that one of them will :)
      • I *work* with Unity3D. I love it.
        For our 3D projects in the past we've been the long suffering slave of Shockwave 3D. Now all of a sudden our portfolio doesn't look like it's stuck in 2001. Plus, with Pro, (And about $2000) we can port to Droid/iOS in no-time, using 99% of the same source code.

        • oh, and yeah, Source control is a bitch, even with pro. Most of my time is spend ensuring our last commit didn't screw over our prefab associations.

          3.5 is supposed to give better SVN Control though.

          • Yep their system of building metadata files for every single asset seems more like a bandaid fix than anything else. It wouldn't be such a bitch if Unity would just attempt to reconnect broken associations for you, or list them in a nice way and let you fix chunks of them at a time, but right now, once you have "missing" where a script name should be you're pretty boned. One of the devs suggested we package things up and put the packages into version control, but the problem with that is that then we'd be r
      • Javascript is what you should teach them. Not the high fallutin everything about it, just the code ganking basics. That way they can use both Unity(free) and make simple and neat HTML5 / Webkit CSS stuff (free). Get their feet wet fast & give them real life skills and a subject they can geek out on in their own time.

        Teach them a little bit about "objects" and then using .CSS and an HTML5 / Webkit browser with extensive documentation.. such as Safari *ahem*. It really can do some amazing things w
        • In my experience, code is fine with Unity, I've not had any problems with it mangling that. It's the Textures/Prefabs/Scenes/Shaders and the way they link to the in-game objects that gets mangled by source control. So after a botched commit you might end up with those new objects you added being untextured, or having the components unlinked.

          • Since packages can include all those components and relationships. If you make a "hierarchy" of packages that you import and Unity project folders that contain things that you sellect & export packages from in their entirety, that can work around most of that suffering. (ie click on your scene, select dependencies, uncheck common things you don't want in the package and export)
  • javascript tetris (Score:4, Insightful)

    by larry bagina ( 561269 ) on Tuesday December 27, 2011 @12:36PM (#38503774) Journal
    HTML 5 canvas + javascript runs everywhere that matters. Old basic games (cards, gorillas, donkey, snakes, etc) should be a good target.
    • Right on. HTML+Javascript is a great place to start. It's simple enough that you can see results right away. You can put it online and then play it anyplace to show friends and family what you're making. You can easily tinker with it after the class is over. 2D games will give a far quicker reward for their efforts. The problem space is far smaller so experimentation is more likely to have understandable results. Scripting/running is more straightforward than compile/debug. These all result in maximizing fu

    • Agreed. This is what I first taught my 10-year-old nephew, and he picked up on it right away. I just started real simple and gradually got more and more advanced. The best thing is that there's no compiling and no required IDE. I actually started off just showing him in Notepad so he could see how easy it really was to create a webpage. I think that would allow the students to be able to easily continue using it on their home computers after class. The students might not be able to setup a complicated IDE w
      • I agree with the JavaScript/HTML approach. I would also point out the some what language similarities between JS and the likes of C++/Java.

        I was also going to mention there's a HTML developer-oriented editor already included with office that few actually know is even there, called Microsoft Script Editor. Default located at:
        "C:\Program Files\Microsoft Office\OFFICE11\MSE7.EXE"

        It's actually pretty good even if it does seem to default to VBScript for some reason. It's kind of like Visual basic (drag but

  • by andi75 ( 84413 ) on Tuesday December 27, 2011 @12:37PM (#38503782) Homepage

    Check out []. It sure looks like kiddy stuff at first glance, but its awesomeness cannot be described, you have to try it yourself.

    Since scratch takes care about all the nitty-gritty details, you can focus on actually *designing* good games, which is awfully hard.

  • But you can't use something that must be licensed for them. These kids need to learn a skill they can go home and practice - if any of them are going to use the skills they learn from you, the tools required had better damn well be Free and Open to them.

    • So why not start with an open source game? Start with ioquake, or now the new iodoom3? [] YOu start by analyzing a existing game, and then build on it. The whole concept of open source...
      • by plover ( 150551 ) *

        That's a good suggestion if the goal is to teach them what it takes to combine artwork, levels, and puzzles into compelling gameplay.

        Of course it has a first-person shooter bias, and won't be suited to a sudoku or Tetris type of game. And a zero-tolerance school board may frown upon creating a shoot-em-up in class.

      • So why not start with an open source game? Start with ioquake, or now the new iodoom3? [] YOu start by analyzing a existing game, and then build on it. The whole concept of open source...

        So you really want to bore them to begin with? Analyzing someones existing code is a really tedious work for even existing programmers. On top of that they wouldn't get to imagine and make what they actually do want to make. Your whole suggestion is terrible.

        If you want to "build upon it", there are far better solutions, like modding Valve's games. Garry's mod is really fun too. Yes, they aren't open source, but do you really want to draconically push such views on newcomers? Because if you do, then congr

        • So you really want to bore them to begin with? Analyzing someones existing code is a really tedious work for even existing programmers. On top of that they wouldn't get to imagine and make what they actually do want to make. Your whole suggestion is terrible.

          Really? So you jumped from nothing to high level coding without ever looking at anyone else's code? Wow! That is amazing! I don't know anyone like that. We all learned from looking at other code...

          If you want to "build upon it", there are far better solutions, like modding Valve's games. Garry's mod is really fun too. Yes, they aren't open source, but do you really want to draconically push such views on newcomers? Because if you do, then congratulations, you just ruined all the fun from starting programming.

          You do know that ioquake and iodoom3 run all the quake and doom3 mods, right? And mod editors, and so on? So you can do all the stuff you want without having to buy a half life for each student, or resorting to pirating it.

    • by vlm ( 69642 )

      if any of them are going to use the skills they learn from you

      Now that, that right there, is where you have to decide if you're going to merely provide low level vocational training or provide (higher?) education. Its difficult/impossible to do both, and both are going to have radically different plans, and results. Decide that first. Then pick your toolset.

      • decide if you're going to merely provide low level vocational training or provide (higher?) education. Its difficult/impossible to do both

        Tommy rot. I've known a number of people with both - who did apprenticeships and/or vocational training at 16-18 and a few years later went on to university. Late bloomers, if you like.

        Did you get that from the same book that says intelligent people have no common sense?

    • But you can't use something that must be licensed for them. These kids need to learn a skill they can go home and practice - if any of them are going to use the skills they learn from you, the tools required had better damn well be Free and Open to them.

      I really don't see how licensing would play any role in this. Those kids aren't making games (or apps) to sell them, not yet. Licensing mostly applies to that only. Hell, now a days even Microsoft offers Visual Studio for free if you don't sell the programs created with it.

      • The Visual Express versions have no "can't sell" limitation.

      • Actually, there is nothing in Visual Studio Express licenses that forbid programs being developed in it being released using any license, open source/free/commercial.

    • by plover ( 150551 ) *

      Most environments offer academic licenses that range from "steeply discounted" to "free as in beer" to "free as in speech". Money shouldn't be the only factor to take into account.

  • I'm really liking Construct ( []) at the moment. It's an HTML5 game engine that's easy enough kids should be able to pick it up and it has the added bonus of being free for non-commercial use. If you want to see it's capabilities, I threw together a little game in around 5 hours of work to learn it's functionality and it turned out ok. You can view it here ( [])
  • by mrbill1234 ( 715607 ) on Tuesday December 27, 2011 @12:42PM (#38503874) []

    There is a free version - and paid for. You can code via their visual tool - or on the command line. My 11 year old son makes fine games using this!

    • I second this. Used it for years (ironically enough right up UNTIL yoyogames bought it out). It is rediculously simple to make basic games using drag and drop stuff, and the basic engine takes care of all the window creation, image loading/drawing, ect, and there's also a C-like language (I think it's loosely based on Python actually, but doesn't really look like it) that is quite powerful. It's an amazing beginners tool, and was actually created by a professor to teach game design.
  • by vlm ( 69642 ) on Tuesday December 27, 2011 @12:45PM (#38503910)

    The dominant /. mindshare definition of gaming is that it is exclusively 1:1 mapped to 3-d FPS.

    If you're willing to break out of that ultra-narrow mindset, there is a possibility of RPGs, text adventures, maybe hex based wargaming, (semi)numerical simulations... A whole world of human computer interaction exists, but only for the open minded.

    Reimplement Oregon Trail as a flash game? (try not to get sued)

    Supposedly HS kids like vampires and zombie books, so write a text adventure fanfic in the anne rice or twilight universe (try not to get sued). Make all your game lines less than 160 char and play over twitter?

    Stock trading game using real stock market data? Or YetAnotherRealWorldFuturesMarketImplementation? Maybe give it a modern twist by implementing it over text messages or whatever?

    Hex based wargamer vampire vs zombies? or plants vs zombies? (again try not to get sued)

    Actually, "try to write Fing anything without getting sued for copyright and patent violations" might make an interesting and informative meta-game?

    • My first thought when I saw this was the Adventure Construction Set [].

      And wasn't Myst [] originally written in HyperCard []?

      If you're looking at writing text-based games, there's MudOS [] and other MUD/MOO/MUSH engines out there, most of which are free.

      • by vlm ( 69642 )

        Maybe you could implement a "choose your own adventure book" style adventure entirely in a HTML editor. "click here to go north" links to rm6342.html etc.

        Not exactly meeting the degree requirements for AI implementation, but ...

  • Python + Pygame (Score:3, Informative)

    by sharp3 ( 1195261 ) on Tuesday December 27, 2011 @12:47PM (#38503930)
    Pygame is a pretty nice little package for quickly building 2D games. Fairly decent documentation and best of all, free! []
  • Have you looked at Unity? ahref= []>

    Its free and a lot of the complex underpinnings are taken care of and hidden away. A simple to use scripting language is used to create the game mechanics. I have never used it but I have seen it used for the Global Game Jam.

    I would keep away from teaching programming unless the students already have programming skills which I doubt is the case here. Keep it simple.

  • but Blender ( is free and open source. It has a great community behind it.
    • the biggest problem with Blender is its VERTICAL learning "curve"

      If any of y'all think you are a Blender Guru then email me with the subject

      and i will send you back a few questions i need answering from somebody that really knows blender.

      • by Animats ( 122034 )

        the biggest problem with Blender is its VERTICAL learning "curve"

        True. If you haven't encountered Blender, imagine a user interface for 3D designed by someone who likes EMACS. The condensed hotkey reference is 19 pages.

        Blender's game engine is interesting. Programming is done by wiring together blocks in a graph. You can write new blocks in Python if you want. It's a nice demonstration of the fact that graphical programming does not scale well. I once wrote a program to simulate LIDAR processing for a mobile robot using the Blender game engine. This was Not Fun.

      • Good point. I also personally think that learning blender doesn't do a whole lot for learning game development or programming in general. However, the tools it comes with for free are pretty great. Not just for game programming, but also for modeling. If one wants to do a game development course for High School I think it would be good to show and share some demos and let students see what they can do with it. Drop it if it doesn't stick anywhere. A survey of many development frameworks and technologi
  • Although I wholeheartedly agree with all the people who are going to recommend Unity [] (which is also the platform I prefer), you might be better served with UDK [] when demonstrating to students. I'd say that Unity is a 3d game engine/platform made for programmers whereas UDK is a 3d game engine/platform made for level designers with support for programmers. You can get a lot of mileage from both platforms without much programming, but UDK is specifically designed so you can create an entire game without one


    this always seemed to get reviewed highly... I haven't used it as my kids aren't old enough yet.
  • The project you are looking for is Bootstrap [].

    Bootstrap is a standards-based curriculum for middle and high-school students, which teaches them to program their own videogames using purely algebraic and geometric concepts.

    Bootstrap uses Scheme/Racket and focuses on the algebraic/functional aspects of programming. The teaching materials are freely available online. They even sell "I program my own videogames" T-shirts.

  • I suggest the Corona SDK. It uses LUA, supports Box2d Physics and it's easy to work with a tilemap editor like tiled to put everything together. Plus, you only need to buy it if you want to publish or sell your app.

  • by roman_mir ( 125474 ) on Tuesday December 27, 2011 @01:00PM (#38504106) Homepage Journal

    In fact $49 [].

    It's called 'impact' and games like this [] are made with it. []

  • Processing can be a lot of fun for small projects: [] Pros: Cross platform, java based, simple IDE, supports keyboard/mouse input, 2D/3D. Cons: Only downside is limited audio support.
  • Don't know what to suggest as tools least the question is being asked. High schools today teach kids to be...users, not creators. Their idea of 'tech ed' is to teach kids to use a recent version of MS Word or Excel. Even the old English class standby has become more about reading stuff that someone else wrote and answering questions about it and much less about writing something new. The high school yearbook class has become all about snazzy software to present graphically-attractive pages rat

  • I am teaching my nephews how to program during their week off from school. I went with a platform that all households had readily available: Excel. Excel VBA is robust enough to create fun games from the Atari generation, forgiving enough to keep new programmers from being frustrated quickly, and the skills learned will carry my wards into many business environments for years to come. Even if Excel goes away, learning to manipulate data, graphics, and data in a spreadsheet program will be

  • Since you have previous pro game dev experience, you should know the entire crew is divided up into teams by task. Assuming some sort of 3d platform, not all the kids will have equal interest in art, modeling, testing, coding, rigging, etc. Break them up into teams.

    Also, don't dismiss the allure of 3d... it pretty much is the main reason we don't all spend countless hours playing sidescrollers anymore, but aside from phones its the only scenario these kids likely know. Even if you just recreated an old 8

  • by Hentes ( 2461350 ) on Tuesday December 27, 2011 @01:22PM (#38504328)

    Game programming would be the last thing I would teach to novice kids, as it has several different parts, from low-level hardware-oriented code to networking, high-level scripting, databases and map design, not to mention a non-programming parts like the graphics.

    If you really want to create a standalone game I would suggest something simple using Flash. But if you want to get them into game programming (and teach them actual techniques that it needs) get them into modding. There are many games designed to be easily moddable, the instant feedback and success will be a great motivation, and the kids will learn plenty of stuff they can later use.

    • +1 to this, start small with simple mechanics in a 2d space. Flashdevelop [] is a great, free tool that is easy to setup and use.

      If you want to try using a Flash game framework, try something like Flixel [].
  • As someone who's also been roped into the "teach game development" trap, my advice is to stay the heck away from programming and 3D and any other component that requires specialized skills. If you want the kids to actually finish something, then do what you can to make that happen. Not all of them will be programmers or artists, and you don't want to spend all your time teaching those skills. Plus, it's also pretty common for those that can't program (or model) very well to have really huge ideas that fa

  • I did the same thing for a group of middle school students back in 2005 and after evaluating a bunch of graphics and sound libraries, we settled on Basic4GL [].

    Basic4GL is everything BASIC was, except without line numbers and with all the GLUT functionality built in (minus the initialization cruft). It also supports sound, loading a bunch of texture formats, and has the NEHE tutorials ported to it, and runs on VERY low end hardware. Download and run the demos -- you'll be impressed.

    The kids did exceptionally

  • by marcosdumay ( 620877 ) <marcosdumay@[ ] ['gma' in gap]> on Tuesday December 27, 2011 @01:55PM (#38504724) Homepage Journal

    I think it would be advisable to start with something 2D, that is simpler to understand and to code. On their first game the kids will have too much to learn, so not making them learn analitic geommetry, lightining, all the tools you'd need for 3D, and lots of other stuff (like "why is my game that slow?") is a good thing. First focus on general programming and basic I/O.

    Now, if you take that advice, you'd need a good library for general I/O that is available in a good language for novices. Well, here I can recomend Pygame, on Python.

  • if you want a physics dynamics engine, you want ODE. if you want a small amount of code (a high bang-per-buck ratio for the students i.e. they get results fast), use python. ODE has python bindings, so you win both ways. there are plenty of example recipes for python-pyode with OpenGL and pygame, which you don't entirely have to "understand", just copy cut/paste just like any other programmer would, and it gets the job done.

    now, if you're looking for web-based, i cannot recommend "pure javascript". it's

  • Game programming is a very complex kind of software development, prone to errors (all professional game programmers write hideous, insecure, unportable code with utterly broken networking), and requires massive amount of effort placed into non-programming-related parts of the project (art, music, story).

    Teaching something that complex in high school will inevitably degenerate into mucking around with pre-made templates, with no educational value whatsoever. If someone really wants to teach kids programming,

    • I agree that it will likely turn into hacking some premade code. However, for a 1 meeting course a week for a couple of weeks, this seems more like a survey course where the only educational value that can possibly be expected is just awareness of the tools and some guidance using them. You're not going to make a programmer, but you may inspire a few kids to be one.
  • by Tronster ( 25566 ) on Tuesday December 27, 2011 @02:19PM (#38505016) Homepage

    Speaking as a current indie & AAA gamedev who has built game-related curriculum for 3 schools (middle school through college)...

    Because of the diversity of student types (artist vs game designer vs programmer, etc...) I recommend teaching EPIC's Unreal Development Kit (UDK).
    The tools are mature and will immediately offer something to every role on a game project.
    As for yourself, the UDK uses Unrealscript which is based on a C++/C# syntax.
    There is a wealth of knowledge via books and internet tutorials. (e.g., Just typed in "UDK tutorial" in YouTube and received over 4500+ results!)
    A nice bonus is that learning UDK is something the kids can actually put on a resume and/or help them get an internship.

    Milage will vary with other pre-built gamedev environments.
    Below are a few all-in-one-solutions that have editing features, based in a Windows environment:

    App Game Kit (AGK) - []
    Construct2 - []
    Game-Editor - []
    Game Maker - []
    Game Salad - []
    Scratch - []

    Good luck!

  • Surprised nobody here has mentioned Adventure Game Studio [] yet. It's free, runs on Windows. Ports are available for linux and mac. Some pretty great games have been made with it, including the excellent Sierra classic remakes by AGD Interactive []. If you want something more old school, there is also ScummVM []. It's mainly used to run old games by enthusiasts, but it can probably be used to develop new games as well.

  • 1. To make it interesting, it must be something that they will be able to use at home, not just in a school setting.
    2. You also need something that will enable them to learn how to communicate while doing the work - and if it's interesting enough, they'll work on it more than just in your allocated time, and possibly even do their own things too (see #1).

    As a result, it is probably best to use something that is open source. From that perspective, you have a few options:

    1. Python - easy to program, an
  • Most people speculate about motivational and educational benefits of certain tools and programming activities. We actually measure them. Scalable Game Design, using AgentSheets, teaches kids how to make games starting with simple 1980 arcade games such as Frogger and them gradually move on all the way to modern SIMs like games including sophisticated AI. With middle and high schools all around the US we have a close to 50% participation of girls. And don't think this is just for K-12. The curriculum + tool

  • by Tom ( 822 ) on Tuesday December 27, 2011 @05:20PM (#38507206) Homepage Journal

    Do you want to teach games or programming?

    If you want to teach games, the first one should not be a computer game. Make it a board game, a card game or something else that you can create with paper and pens.

    Anything beyond that adds complexities that distract from the game design itself. There is very little design-wise in a computer game that you can't have in a board game.

    And yes, I am a (hobby/indy) game designer. I've made some board games, a card game, a play-by-mail game, two pen&paper roleplaying games and a bunch of computer games. Largely in that order.

  • I see a lot of suggestions on how to avoid coding which seem silly to me. High schoolers that are interested in making games are probably smart enough for a little coding, and it'll do them a lot of good. It certainly doesn't even rule out other people (visual and sound design, etc) as often the design takes as much or more time than the coding.

    I really like pygame, it's:
    a) python
    b) fairly straightforward
    c) engine-less
    d) cross-platform
    e) free and requires only a text editor (I like komodo edit for python,

  • Look at programming boot camps or summer camps for kids in the same age range, and then see what software they are using: []

    The above is a two week course aimed at kids: 7-10 | BEGINNER – ADVANCED

    Software: Arcade or Platform game using Clickteam® Multimedia Fusion 2 Developer® and Adobe® Photoshop®. Build custom characters in Spore Creature Creator and import them into your game. Take breaks with supervised outside play, s

  • Contact lego. They might even sponsor the school. The game is Soccer

Given its constituency, the only thing I expect to be "open" about [the Open Software Foundation] is its mouth. -- John Gilmore