As many of you know, I'm working on building an online RPG from scratch. I've been working on making a GUI for the character creation code the last few months and trying different things.

I won't bore you w/ the details of it but in the last week I realized I may be going about this whole thing wrong. My original plan was to create a program that people could install to their computer and then play the game through the application, similar to the OpenRPG project. (Yes, I could just use their project but I want to learn more about programming by doing it myself).

But now I see that, since my program is designed to let people play via the Internet, why not just make it use the web browser? If I created a web app then I wouldn't have to worry about making a custom GUI and making sure the program is cross-platform installable. Plus, I've never made a web app before so it would be another learning experience (which I consider a good thing).

I guess my question to you more experienced programmers out there, in your opinion do you think it's better to continue working on a stand-alone application or would it be better to make a web app? If I did create a web app, what considerations would I have to account for vs. making a stand-alone app? Obviously the GM would have to act as a server but that's what I intended to do w/ the stand-alone version.

The only code I've really created so far is a console-based character creation module so making a web version of it wouldn't be any more work than what I've done trying to make a GUI. If the backend data would be the same either way, is there a decided advantage to going web or not?

__________________
Start Programming with Python-A beginner's guide to programming and the Python language.

Interested in pen & paper role playing games from the "golden age" of gaming? Visit Old School Role-Playing Games

I think that I'd separate the ideas of being a networked game from the idea of being a "web" app or not. The browser is a matter of personal choice and supporting the same look-n-feel means testing on several common browsers on a variety of architectures for similar game aesthetics. You're also directly limited by what the browser is able to do with regard to networking protocols and the related statefulness of those protocols. You also tend to require a server URL in order to initiate the game play, since the browser isn't going to respond to requests, rather it issues them.

Perhaps using an Applet or JApplet in a browser would be a suitable alternative? But then, the question would be why not just create a Java application rather than an applet?

I'd probably do away completely with the notion of using a browser and common "web" protocols, namely HTTP and to a lesser extent, FTP. While there are a lot of benefits in using web-centric technologies, there are also severe limitations. Perhaps one of the biggest benefits of using a browser is that you naturally overcome the firewall and related proxy-access hurdles when it comes to making connections to game server(s). Of course, then you have to write a game server in addition to the game "client," too.

I'd probably use Qt and OpenGL for the game client and the Qt networking facilities (including writing a custom protocol) for networked play. I would dispense with the "server" component of it and allow a client to "host" a game by simply opening up a server socket.

In the ever expanding land of broadband, there are a lot of people who can't host games because they can't receive incoming connections since such are not forwarded by the service provider. This suggests that a traditional game server is a useful piece, though I'd try to figure out if it is really necessary or not, since it could be a lot more code to write depending on the design and features of it. At the minimum, it is a medium complexity multi-threaded service-based (daemon) architecture that requires administrative "console" capabilities for configuration, startup, shutdown, restart, etc.

My choice for using Qt is due to its rather advanced capabilities and its reasonable platform independence. Also, with the OpenGL module, it will be much, much faster than Java in terms of 2D and 3D graphics. Qt is also free for open source projects. Qt supports the notion of plugins so that expanding the game over time is feasible using the plugin facilities such that your app simply searches the "plugins" directory (or whatever you want to call it) and loads them. That architecture is a good choice for a lot of the game components such that as you define a component, by following the plugin API you naturally separate roles and responsibilities of the project into suitable plugins. Updating a component of the game is simply a matter of swapping the plugin by name (file) so that the new fixes/capabilities/whatever are then available when the program is restarted. Of course, Qt is very C++ centric, so it is similarly demanding to some degree. One of the biggest benefits of Qt is the choices of looks-n-feels that you can use...and, of course, you can create your own custom look-n-feel specific to your game for yet another aspect of making your game unique.


:davis:

Well, thanks for your advice. I've thought about Qt since it's one of the recommend UI designers for Python (my selected language). PyQt can be used to convert the code. The only problem is getting PyQt to work correctly on my Mac. Because of this, I'm looking at using wxPython, since it's cross-platform as well.

I was thinking about using TurboGears or Django to make a web app, if I went this route. Can't I make a fairly universal look using CSS? What about using AJAX to make it more interactive?

I do like the idea of not having a "central server", whether from a browser or stand-alone. Can this be done w/o Qt? If so, how?

For a final question, is it better to make a console-based version first then make the GUI later? I'm thinking that by doing that, I can verify the game design and logic w/o being distracted by GUI elements. The problem I'm finding is that I can readily port the text-based code over to a GUI since much of the GUI is tied into the logic differently. I can still use the same though processes but I essentially have to rewrite the logic flow to account for the non-linear functioning of the UI.

__________________
Start Programming with Python-A beginner's guide to programming and the Python language.

Interested in pen & paper role playing games from the "golden age" of gaming? Visit Old School Role-Playing Games

The game "engine" should probably be a "console mode" type app and then the front end will be interchangeable. You may want to make it a "daemon" in that it is basically a server socket waiting for a front end to connect to it. That makes it much more networkable and distribute-able.

In answering some of your other questions, I tend to try to simplify when it comes to "technology layers." In other words, I don't like to add too many disparate technologies to the cake, as it then becomes a challenge in managing all of the technologies and the constant revisions of software packages involved in the app. With Qt, not that it will work for your needs, but with it, you are focused on one package that basically does everything for you with regard to complete GUI, networking, database, OpenGL, etc.etc.etc.. The "alternative" IMO is Java. It has a very robust library, and is well supported on multiple platforms. However, you seem very much dedicated to Python, so this "argument" is completely moot. Perhaps you should write the game engine and let others write the front ends in whatever languages and GUI toolkits they choose?

Still, I'd think that a C++ (Qt or not) game engine would be far better in the long term given a lot of players, if such is feasible for the game.

I don't know if I understand the rules of the CMRPG. What is the objective/purpose? It seems as if it is designed as a first-person shooter where you roll-up (generate) your character in a D&D (et al) fashion and then go about contributing to some greater or mission objective. However, it seems to me that the preferred method would be similar to a C&C kind of game rather than a FPS where the player is the "Commanding General" of the various troops and positions them and utilizes them accordingly to the battlefield scenario. But then the whole individual "stats" for the "character" are certainly less meaningful and the RPG aspect is gone.

The basic fallacy with "Colonial Marines" is that a military is composed of units and units are composed of troops and troops follow orders and orders come from a Chain of Command. That means, in order to play your game, people have to "volunteer" to be privates, and corporals, sergeants, lieutenants, captains and so forth. I mean, you can "promote" them as play moves in time, but what is the definition of the chain of command at the start of a new game?

A good game needs to be single player and multiplayer where multiplayer means two or more.

I can see a RPG-like FPS whereby there are not ranks, only skills and upon initial roll-up (character generation) some basic skills and traits (you seem to be missing these) are part of the character such that as battlefield time increases, skills increase. This puts each player in a peer-to-peer basis and sets up opportunities for chatting to evolve mission objectives.

I'd also want to research the copyright aspects of using the concept of Colonial Marines, which is obviously a copyrighted motion picture "entity" in the Aliens movie, script, book, etc.

I think that you need to come up with some kind of overview document that describes what CMRPG is and how it should be played (statement of vision) so that technology choices are measured in terms of their ability to effect the vision. If you have these and I didn't see them, sorry.


:davis:

No, it's not an FPS. My idea is to recreate the old table-top experience on a computer with people being able to play via the Internet. So it would have things like a dice roller, a map, a chat program, etc. that can all be shared and everyone can see the results and people can interact as needed.

I'm actually not expecting this game to really go anywhere. I really started it as a way to learn Python. Sure I could make it in Java or C++ or whatever, but I wanted to learn Python so I came up w/ this project. That's why I'm so "tied to Python" for this game; I want to see how capable the language is. I'm also thinking about making another game based on the old Twilight 2000 game using Java or C/C++. Right now I just want to learn more about programming; my game idea will make me learn networking, graphics, UI design, etc. and I can't think of a more comprehensive project right now. Though I could join another Sourceforge project, I'd rather start from the beginning rather than jumping into the middle of something.

That being said, since the GM would control the whole experience like a traditional pen & paper game, it would be up to him to decide things like promotions. I (or whoever codes for the game) would just provide the tools and let the GM decide how to make the world. It wouldn't be like a RTS or typical computer RPG where the computer makes all the decisions.

I'm aware of the copyright issues but I believe I'm using it under "fair use". It's my own project and I'm not making money on it. If I ever do get a "cease and desist" letter, then I can keep the same game structure but make up my own terms and create a whole new universe.

Since I'm the only one working on the game right now (some offers but no follow through) I'm thinking of pulling the web site since it's not doing anything for me. I don't update it as often as I should because I'm working on a Master's degree right now and it's just not a priority anymore.

__________________
Start Programming with Python-A beginner's guide to programming and the Python language.

Interested in pen & paper role playing games from the "golden age" of gaming? Visit Old School Role-Playing Games

...then the GUI is at least "minimalist" compared to a more graphically intense gaming experience, yes?

To me, the idea of a dice roller is a bit absurd (not trying to be rude!). If the call is for 2D10, then the "outcome" should simply appear on the screen and the hit-points taken, IMO. Obviously, this is a "speed-up" element, but in the process loses the "old fashioned" RPG element of selecting the proper die/dice and begging the gods of random number generators for minimal losses.


I believe that your expectations are correct, but I tend to think that inwardly at least, that you hoped that it would be popular...in a good way. What I think may have happened is that you've basically found out how many people want to choose the same path as you. If you've accomplished your goals of learning Python, then kudos. If you want the game to progress into something that others are more likely to contribute toward, then maybe a language change or at least the notion of "generate your own front end in whatever language you want" might be a path to greater successes? I really do not want to be considered the person who is or is trying to shoot holes into your project because I think that it is a cool idea. In some ways, I want to learn Python too, but I know myself too well and know that I don't really appreciate languages that are not more strongly typed. I always "think" in type-specific ways and that's a problem with languages that let types float all over the place! At least, it is a mental barrier for me, though I can do it (read: pulling teeth), I prefer not to.


Have you accomplished this objective?


How about a summary of what you've discovered? I know that I'd be interested to hear it.


...are you talking about the "old" BBS "door game?" ...my memory is fairly foggy on it, as I ran a very large BBS and had many, many doors, but sometimes the names escape me since the fact is that BBSes died at the advent of the WWW.


The fundamentals of programming tend to be quite similar for many languages with only a few exceptions relative to the whole. The key question is WHY do you want to learn more about programming. For what reason(s) is that important to you? For me, it has always been career-based. If I had my choice, I'd never write another program again (though I do find it fun, too)...because I'd be retired on a beach somewhere sucking down cold drinks with stupid pink or purple umbrellas sticking out of them brought to me by topless Polynesian women or something similar. I do sometimes try to learn "something new" in order to just better understand some aspect of the particular language, but most of my directed education effort is spent trying to enhance my profitability in the industries where my skills are aligned.

You see, I'm the kind of person who would want to sell the game to a game manufacturer and walk away with a nice and nifty lump sum at the end of the day in exchange for my labor...even if some of it was learning something new. I am also very supportive of the notion that it should be an open source project where there is no money, but that it is platform independent to the point of running on Linux first and, if someone else cares, Windoze, too. I spend as much time as possible trying NOT to start up a M$Windy box.

All that this really says is that between the two of us, there are some obviously different sets of motivations. However, I think that it can be inferred, that among all programmers, a super set of all of their motivations will be relatively non-intersecting in terms of any "major percentage group" if so analyzed. This can be wrong based on the language choice, is my "argument." I have no intention of "bad mouthing" Python or any other language, as all have some useful and relevant purposes, but the "80/20" rule kind of plays against it in terms of mass appeal, I think. Granted, the same can be said for Windows vs Linux...maybe to an even greater extreme? I certainly commend you for your efforts in being platform independent. To me, that, by itself, is a much more important "sign" than language choice.


...and, possibly, so would a lot of others?


I have a "major problem" with this since it requires a kind of unity among the players that is reminscient of the days of D&D, Blitzkrieg, etc. A group of people would meet regularly at some scheduled time/place and bring all of their "equipment" (or, if you were really severe, you had a permanent place set up for it--usually in someone's garage no matter what time of year!) and play would commence. I don't want to be negative, but the "Internet Age" suggests a more dynamic and less "interpersonal" kind of system, in my opinion. I remember the days of D&D and so many others where players would wait weeks or even months and years to get into a game with a really good DM/GM. Try to find some kind of overlapping notion of that in today's "online gaming" metaphor? I'm not saying that it isn't possible, but that the technology lends itself to a different model compared to the yesteryear version(s). Let me state, however, that I'm perfectly open to "enabling technologies" that help promote the more "nostalgic" gaming experiences. I am a bit confused as to how one would "know" that a DM/GM is "good" in the potential for thousands if not millions of simulataneous ongoing games?


Okay. No worries. I just wanted to point out a possible area of conflict. I think that as long as you do not have a component of the game that assumes particular character roles equating to movie characters, that your fair use may be valid. Otherwise, the rights of the original book (if exists) author, screenplay author, etc. are likely to be considered plagerism. I'd like to add that I don't think that anyone here thinks that that is your motive, rather, perhaps just a tidbit of common sense to be considerate of...


I think that you bring up a really interesting point in the "who's got the time to be involved?" question. Generally, I'd say that if it isn't costing you any support time (or very, very little) then leave it alone, your "marketing arm" may actually reach out to someone over the span of time. However, back to the point of time and "spare time" as in being non-existent, I'd see if there wasn't some appropriate way to further delegate responsibility rather than try to be in control. Sometimes by inviting others to rewrite the entire thing in a different language, you might gain the benefit of momentum that helps move the vision to completion? Of course, you can always invite them to write a portion of it in a different language, too.

I think that the "problem domain" becomes one of what is good for you and what is good for the game? Since you have no expectations of it going anywhere and it really is the foundation of an educational activity for you personally, then why not "let it go free" to see what roads and directions it may possibly take? Of course, nothing is to say that by doing that, that anything at all will be different tomorrow or even in 6 months. I'm just suggesting that perhaps it is something for you to consider?

I remember watching Aliens in a theater in Tacoma, WA in 1986 and thinking that the effects and "toys" were way over the top compared to anything else done up to that time. I am also prior service, Army and a Special Forces combat vet. So, I had a lot to relate to in the CM portion of it, particularly since a lot of my good military friends were Marines. I really connected with SSG (if I'm not mistaken, I can't remember if he was an E6 or E7) Apone...even though the credits show him as SGT.

I think that my point is that how many others here are connected with the story in the same way that you and perhaps I are? It was 20 years ago now and because of the R rating, it would mean that someone would have to be a minimum of 37 to have seen it or at least been enthusiastic enough about it to see it on cable or rental video since then. IOW, you may be chasing a group with diminishing potential for return.


:davis:

Exactly. That's why I'm looking at Tkinter, wxPython, Qt, or other "widows" GUI rather than making a custom game interface. I just want it simple and "low tech".

The dice roller isn't really anything special. It's really just a few buttons labeled 1d6, 2d10, etc. that output the result of a random number generator. It would be tied into the game engine such that the HP or whatever are automatically deducted. In essence, the game would show you the result but do much of the "paperwork" for you.

What I've found is that people generally aren't interested in learning Python for various reasons; some don't like the "white space counts", some don't like not having brackets, some don't like not having types, some prefer more control over their code, and so on. A lot of people just don't want to learn a new language. I can understand that.

I wanted to learn Python because I've read many people who prefer to use it for "hobby programming". C/C++ are powerful but I don't like how easy it is for errors to occur and I like not having to compile to test it. Basically I feel lazy and prefer to use tools that make my life easy. Since I got a computer engineering degree and not a computer science degree, I didn't get hot & heavy into data structures and the "hard" programming stuff. Python is the first time I actually understood how the OOP paradigm works; I could never figure it out when I took C++ or Java.

Obviously I wanted the game to be such that people would want to contribute to it. One of the reasons I chose a game such as this is that it's not a 3d graphics-intensive FPS or anything that requires lots of programming skills. Ultimately it's not too different from making a standard Internet-capable office application. Because of its "simplicity", I also thought it would be a good way for people curious about Python could use it to learn the language without reading a lot of books. That's why I wrote the Python tutorials for GIDForums.

I still want to learn how to do socket programming or other networking things, learn the benefits of serialization vs. file I/O, and figure out how to save things from a window then pull the data up in another window. So I'm still using Python to do this, since it's just easier for me.

Besides my wants, I have learned quite a bit about the language. I think it's the perfect language for beginners. It lets them think about the algorithms and logic required in programming rather than having to worry about proper syntax. You can start out using procedural programming techniques then move to OOP as your skills develop. You can combine OOP and functions together as needed; you don't have to think of everything as an object like other languages, but you can use classes if they'll improve your program.

I feel that my programming skills have increased at least three-fold. I may not remember how to program in C or C++ without looking at book anymore, but I think I would have a much easier time now than I did in college. I absolutely love the idea of lists, dictionaries, and tuples. Not having to make linked lists, arrays, and data structures from scratch every time is fantastic; I can just tell the program what I want to do and move on. The language will take care of the dirty details.

Because Python has bindings to C naturally (Python is actually written in C), if there is something that needs the speed of a compiled language, just write it in C/C++ and bind it to Python. Contrarily, you can write a program in C/C++ and bind a Python script into it when the convenience is good. Many video games do this; writing the graphics code in C/C++ and making much of the game engine and supporting scripts in Python.

If you are familiar with or need to use other languages, Python can probably do it. IronPython is Python for the .NET framework; the creator now works for MS so the IronPython framework is supposed to be in the next Visual Studio release. Jython is Python for Java and works the same way with Java as standard Python works w/ C.

Python is slowly becoming a popular language for companies. Google and Yahoo have Python APIs, NASA uses Python for a lot of their code, Doom and Civ4 use it for their scripts, and many corporations are starting to use it internally. Python programmers are moving up in the recruiters sites and supposedly make more than a C++ or Java programmer (I only know that second hand so I don't know how true that is). Many places use Python to prototype their projects then rewrite it in another language.

Basically, I've learned that learning Python makes me more marketable and valuable regardless of whether I actually get a job as a Python programmer. I still don't know what I want to do when I grow up but whether I program for a living or just to make software for my needs, Python is good to know.

No, this is a table-top pen & paper RPG made by the Game Designers Workshop. You can find a few companies that sell PDFs of the rules and modules.

I first took programming after I got my AAS degree in electronics. I wanted to get another Associate's and couldn't decide between laser technology and programming (I should have took lasers ). I dropped out of that program because they were teaching COBOL for a year and C++ for only a semester. I knew that COBOL was a "dead" language and that all the new stuff would be C++. Plus I didn't like the COBOL syntax and didn't want to be a maintainer of old code; I wanted to be making "shiny and new" programs.

After that I never touched programming again until my computer engineering degree required that I learn Java, C, and C++. The first class was Java but the book was horrible. Actually I think it was J++ which could explain why it was so horrible. The course was bad enough that the school cancelled the course while I was still enrolled and just gave me credit for the class (I only had 2 tests left).

I didn't actually start to enjoy programming until I took C, then I could actually see my programs run correctly and get that warm fuzzy feeling. I had fun learning C and C++ until I got to OOP, then I got lost.

I was still interested in programming because I figure it's something good to know (like changing a tire or CPR) in case I ever needed a program that I isn't already written. (I know, pretty naive). Plus I wanted to just have a better knowledge of programming in case I ever get a job where it would make me stand out. Hence, Python.

Right now it's good that I learned Python since I was able to change shops at my command because someone saw me reading a Python book. I'd much rather mess around on a computer rather than pulling Ethernet cables and moving towers from one desk to another (my old shop).

I still have no idea what I want to do when I get out of the Navy, but I wouldn't be adverse to programming. I like computers but I don't want to do tech support anymore (did it for 12 years) and networking doesn't interest me enough to pursue it.

Possibly. Everyone has to start somewhere and it's usually easier to start at the ground floor than be thrown into a full-blown project. I didn't want to try to understand the logic of the program while still learning the language. I had to do that at work and it stressed me out too much (the language is ACL by the way).
If you don't even know the syntax of the language, it's very difficult to learn how the program is supposed to work.

My idea is not that the game would be like an IRC channel or chat room but that a small group of friends could get together over the Internet and play a game. Perhaps it's a group of friends who played together in school but moved away, or maybe they don't want to worry about hauling boxes of books, papers, and dice from house to house. I don't know. But it was really going to be a small group (half a dozen or so) that feels playing over the Internet just makes sense for them. That way everyone knows everyone else and there's no worry about playing with strangers and putting up w/ online jerks as some people encounter in MMORPGs.

Noted and considered. Thanks for the concern.

I've already told people who've contacted me that I have no problem with them using a different language. Since I've licensed the game under GPL, I honestly don't care if someone wants to take the idea and redo it in Java or whatever. When I said I'd kill the site, I was also thinking of setting it in the public domain if people wanted it. But Sourceforge doesn't cost me anything so I'll probably just end up keeping it. It does make a good place to put my code if I make some when I'm away from my Mac.

I realize that many people just don't play traditional RPGs anymore; if it's not a Final Fantasy remake, people aren't interested. I do know there is a niche market for old text based games like MUDs but the "big money" is in graphical games. I'd like to learn how to do this also, so maybe there will be another game in my life sometime.

Since I've pretty much accomplished what I set out to do, I'm not to worried about other people any more. It's still about my personal journey, which is why I think I may make two versions of my game: one stand-alone installable and one web app version. This way I'll learn how both styles of programs work, increase my knowledge, and make me (hopefully) more valuable when I get out of the Navy.

__________________
Start Programming with Python-A beginner's guide to programming and the Python language.

Interested in pen & paper role playing games from the "golden age" of gaming? Visit Old School Role-Playing Games