Transport Tycoon Forums

I think it's easy to call a project 'not according Open Source' if it isn't your project and you aren't experiencing problems that keep you from publishing the source code already. It's been qualified as if TBOT and his friend aren't willing to follow the Open Source definition. In reality there is a fine line between not willing to and not being able to. I don't think you can distinguish between those two if you haven't met that fine line once yourself (not necessairily with programming btw). Amen

I don't care how it's done as long as it's done well.

If the programmers feel that they are doing well on their own, they should keep it closed-source (is that the right term?) to stop other people making suggestions that they don't want or making some code in a different style to the rest and confusing the main programmers.

If they're having problems, open-source may be better as you have more testers and people who can come at the problem from a different angle.

From their website, it sounds like they're doing well on their own, so they should keep it that way unless they have a problem that they get stuck on.

I'm not a programmer, this is just me applying 'logic' and 'common sense'.

Prof. Frink wrote:If the programmers feel that they are doing well on their own, they should keep it closed-source (is that the right term?) to stop other people making suggestions that they don't want or making some code in a different style to the rest and confusing the main programmers.

That's the kind of attitude that causes nothing but problems (SiS anyone?).

Open source isn't supposed to be a lifeline for projects that hit a wall (that would be a pretty lame cop-out). It's supposed to reduce problems, not get around them. All the "reasons" people seem to have given in defence of closed projects in this thread are poor attempts at making excuses.

The usual excuses:

"Someone can just use our code and claim it as their own."
No they can't. No OSI-approved licences permit this, since it would allow people to close an open project.

"Other people may write code that we don't understand, and doesn't fit our style of coding."
There is never an excuse for not having an open standard.

"We want to retain some control over our code."
There are ways of doing this while still keeping to open source ideology. CVS is one example. Maintaining a "control" package and taking patches against that is another.

"Oh, we're getting on fine by ourselves. There's probably nothing here we can't do."
Many hands make light work. It's an old proverb, and proven fact. Look at TTDPatch. How far behind current progress would Josef be if he were still working alone?

"We don't want you having this. Go start your own."
That attitude is both wasteful and selfish. It is an incredible waste of time and effort to have to duplicate someone else's work, and rather selfish of those people to insist that others have to go to that trouble.

Bla bla bla. If I make a project I make a project my way, if you make a project you do that your way and if TBOT is working on a project he can do it the way he likes.
What are you talking about? Where is your authority over their project?

Well, they promised one day everyone can have the source and do as they like with it. Though of course I hope that we will all cooperate and contribute to a better, more fully developed game, rather than going off in ten totally different directions.

ChrisCF wrote: The usual excuses:

"Someone can just use our code and claim it as their own."
No they can't. No OSI-approved licences permit this, since it would allow people to close an open project.

Yes they can, none of the licenses allows us to sue them in any way because we can't prove they used OUR code.

"Other people may write code that we don't understand, and doesn't fit our style of coding."
There is never an excuse for not having an open standard.

Rule 1 in programming, never use code that you don't understand, because you can't fix bugs in it.
Secondly, how would you like if we had to turn down 90% of the submissions because there is no clear standard of coding now. Releasing the source now would be a dead-stab for the project.

"We want to retain some control over our code."
There are ways of doing this while still keeping to open source ideology. CVS is one example. Maintaining a "control" package and taking patches against that is another.

Even when TTS will become open-source there won't be an open-submission CVS. All extensions would have to go through us, as a sort of quality control, to keep the code bug-free.

"Oh, we're getting on fine by ourselves. There's probably nothing here we can't do."
Many hands make light work. It's an old proverb, and proven fact. Look at TTDPatch. How far behind current progress would Josef be if he were still working alone?

At this moment many hands would take up all our time to explain people how to use the code, and correcting mistakes in submitted code. I guess even the core of TTDPatch wasn't open-source in the beginning.

We could have decided to keep the project secret for the time we developed the core engines (which we're doing at this moment). We could have made it public when the source was ready to be used. We could, but we didn't, because we need motivation from other people. If we didn't make this public the project would have been dead already.
(Thanks to everybody for your support up to now, I hope you will be more gratefull about our work)

"We don't want you having this. Go start your own."
That attitude is both wasteful and selfish. It is an incredible waste of time and effort to have to duplicate someone else's work, and rather selfish of those people to insist that others have to go to that trouble.

GO COMPLAIN TO CHRIS SAWYER THEN!


Besides all this, I think you should respect us for putting all this time and effort into this project. Come to think of it, the time I took to write this, wasted at least 10 minutes which I could have spent coding.

krtaylor wrote:..., rather than going off in ten totally different directions.

Releasing the source now would do just that.

TBOT wrote:I guess even the core of TTDPatch wasn't open-source in the beginning.

TTDPatch manual wrote:Josef is the main author of TTDPatch. He started the project in 1996 and made it public in May 1999

For what it's worth, as a keen observer of open-source projects and a project manager of various types and methods, I agree with the position of the TTSD team. I'm sure we'll all agree that Linux is the model of an open source project. Well, Linus Torvalds kept control over what went into Linux for a long time, until it was stable and well regarded. He still has a great degree of control over exactly what goes into the reference version. An open-source project is not a free-for-all, there needs to be some guiding control.

What the TTSD team is proposing makes perfect sense to me, and I am completely satisfied with it. Of course we'd love to get it done sooner, and more people working on it would be great, but until the main engine is done, they are probably right that it would take more time to explain how the code works than would be gained by additional coders. When the main engine is done and working, then that is the time to release it, with documentation on how it works, so people can go on and do the various pieces of an operating game withough having to go into the actual engine code.

Couldn't agree more. Don't forget: open-source doens't need to mean it should be available as soon as the first bytes of a program are ready. The main function of open-source is that you create a reasonable working, or at least beta version of a program, with solid rules for programming (the main engine), and people can review the source code, and fix problems, and add features, which are in line with the original program. Open-source is far from public property. It's still the creators' property. The same goes for designs for example. I can sell a design to someone, but the design is still mine, and if someone want to alter it, he should first ask my permission. Similar situation... That's my vision on this issue...

As I see it, you have our full support except for one person who feels otherwise. Nothing wrong with that. Please continue with our best wishes and eager anticipation.

To change the heading of this conversation, a few minutes ago I had a conversation with Bas.

We have decided to redo the code we have at the moment in C++, opposed to C with which we program now.

We are beginning to run into some limitations and very ambiguous code. We think C++ will sort those problems, and besides that it will make the engine far more extensible, and also easier to extend.

A little downside is that we have to learn some C++ code structures (class definitions etc.), luckily I already learn Java at my study, which will help us quite a lot.

I don't suppose you could do it in Java? I know Java but that's it...

java knows similar structures as seen in C++. Wise decision though, as C++ is much more common these days, and has some advantages over C...

I'm not recommending you change, really, Java runs kinda slow on Windows systems which is what we all use for gaming. It's good for servers though but that's a different story.

I don't know much about Java, but can it run complete games, including all the things we see in TT? With fullscreen DirectDraw and similar things. I always Java can't do much more then this:

http://spaceflight.nasa.gov/realdata/tr ... index.html

You're talking about a Java applet, which is more limited. Java proper is a full fledged programming language that can do everything any other language can. However, I'm not arguing that it is the right language for TTD. Java's advantage is its cross-platform portability. That isn't really important for TTD, and C/C++ is more efficient in hardware use because it compiles native. Plus there's a long history of coding games in C so there may be various code elements that can be borrowed from elsewhere; Java is younger, particularly in the gaming field, so there wouldn't be as much available.

Uhmmm, why are you argueing about java? I said we were going to redo it in C++ and that for the object orientation my knowledge of java might become handy...

I'm not making any argument at all, just discussing things and trying to answer Bart's question. As far as I am concerned, TBOT, your plans are perfectly sensible.

Guys guys!!!

Damn, I made the OP hoping to get some support and interest in the project, but it seems everyone is either bitching about it or arguing amongst themselves about better ways of doing it. Sorry TBOT

Could I suggest we try to bring a more positive attitude to the thread? Maybe even some appreciation for two guys working their arses off to try and create a better version of a game we all love??

Cheers.

Good idea, let's hope they succeed and hold on