make Cantr open source

[fishfin]

I don't know if this has been suggested before (I searched for 'open source' and I didn't see anything similar in the first couple of pages of results).

I got this idea because the programming department is understaffed and so new things are being implemented at a pretty slow rate. I know that I (and probably others) don't want to join the programming department because I know I don't have any time for it, yet, knowing myself, I would probably find time if I was working on something I really wanted to see implemented (like wooden locks).

I'm not sure if the costs of this would outweigh the benefits or not. I made a short list of the pros and cons that I thought up.

Pros:
1. This could really help get things done with the programming department as understaffed as it is
2. Cantr could be advertised to the open source community and potentially get more players

Cons:
1. Other games will spring up everywhere that are very similar to Cantr, possibly diluting the player base
2. All those security issues that come with everyone know the code
3. People could find ways to implement things (like invincible armored cars) that the staff don't want in the game and then have hard feelings about it not being added to the game
4. Everyone will know the map

Some of those things could (possibly) be avoided by only making open parts of the code and then only to players.

Edit: I was also wondering, what programming language is Cantr written in? I know that the front end is php but is there something else as well?

[SekoETC]

This was brought up before and Jos said no because Cantr is a unique game and we can't afford to have loads of copies springing up. It's php/mysql.

[formerly known as hf]

I think your list of cons pretty much cover why this is a non starter?

Open source projects require lots and lots of admin time, and I doubt it would attract enough interest to increase the programming base and be worth that time.

[ceselb]

Open source projects require lots and lots of admin time

Huh? That's not true. How would making it open source suddenly increase the time needed? I think I know where you're coming from, but that paper was a pile of lies by a think tank funded by microsoft and has been debunked pretty thoroughly iirc.

Cons:
1. Other games will spring up everywhere that are very similar to Cantr, possibly diluting the player base

For a game with only a few hundred players as is? I don't think so, but I suppose someone will try.

2. All those security issues that come with everyone know the code

Huh? Not really an issue, unless there's some risk of sql injection. But that comes from sloppy coding and has nothing specific to do with open source, plus it would be noticed more easily.

3. People could find ways to implement things (like invincible armored cars) that the staff don't want in the game and then have hard feelings about it not being added to the game

No, ProgD has nothing to do with implementing items and such, that is RD.

4. Everyone will know the map

No, hardly anyone has access to the map. I suppose someone with access to the server could get the map, but that also has nothing to do with open source, it's who we let into the server.

Anyway, the problem is that the current ProgD doesn't have time to train someone new, so nothing is happening. I've talked to Jos about this, that is part of why he's back now I think. The problem is that currently we don't let anyone into the server without training. My suggestion was to show aspirants the code, so that they can improve things and in doing so train themselves. Eventually we would get a new batch of people we trust that could get server access. Trouble as I understand it is that such a policy change requires input from Jos, since gab/gac also is pretty defunct and unwilling do do much of anything.

[Tiamo]

Anyway, the problem is that the current ProgD doesn't have time to train someone new, so nothing is happening. I've talked to Jos about this, that is part of why he's back now I think. The problem is that currently we don't let anyone into the server without training. My suggestion was to show aspirants the code, so that they can improve things and in doing so train themselves. Eventually we would get a new batch of people we trust that could get server access. Trouble as I understand it is that such a policy change requires input from Jos, since gab/gac also is pretty defunct and unwilling do do much of anything.

This can easily be solved by setting up separate development and test environments, containing exact copies of the actual game software (using near-empty databases, of course). That would be a perfect (self-)training ground for ProgD applicants.
I guess some kind of test environment already exists (hopefully on another server than the game server).

Making the game open source would, in my opinion, give people (players) too much detailed insight into the game system, like exact formulas and possible shortcuts. Cantr is mainly a role playing game. Let's keep it that way.
Opening the source code for anyone to offer 'improvements', whether or not asked for, will probably induce a flood, if not a tsunami of ready-to-use changes. Many of those changes will be inspired by personal wishes regarding the game, not by an overall view on the game structure. It will be hard for the game's governing body to stay with the overall game design, and refuse all those nice, but not really fitting features.
A third problem i see with open source development is the testing process. Contributors will not know who is working on what change. Changes can therefore easily interfere with each other, making testing a complicated and even more necessary process. It means ProgD will have to oversee the testing process, and actually spend time on this, probably more time than is available...

[ceselb]

I guess some kind of test environment already exists

There is one, yes. But guess what? ...it doesn't work. Plus it may not be possible to isolate privs enough for that to be a viable solution, I don't know.

hopefully on another server than the game server.

There is only one single server for everything.

[Jos Elkink]

There is one, yes. But guess what? ...it doesn't work.

Is that true? Can you contact me privately, through email, about this?

There is only one single server for everything.

Nope, the test server is in a different one.

[ceselb]

Is that true? Can you contact me privately, through email, about this?

This is all I know, there was some post stating that it wasn't working. It may have been fixed since, but I haven't read any such message.
Anyway, the access to the test server isn't given out, afaik. Can we synch the two servers codebase and get some aspirants access perhaps?

[formerly known as hf]

Open source projects require lots and lots of admin time

Huh? That's not true. How would making it open source suddenly increase the time needed? I think I know where you're coming from, but that paper was a pile of lies by a think tank funded by microsoft and has been debunked pretty thoroughly iirc.

It depends how you do it. But if you plan on releasing regular, coherent updates and a single product, someone or people are gonna have to take the time to facilitate that. Projects only reach that state of emergent development where things work almost non-hieracrchically if there are enough people working on it. It's rarely the case for small teams without at least one person having strong, guiding input as director/facilitator/admin or whatever.

[fallenval]

I don't know alot about code so maybe I'm just ignorant but I was thinking instead of ProgD having to make every implement why not put it up on an "Open Board" someone makes their peice of code and sends it to a reveiwer they just reveiw it instead of making it by scratch.

Again I dont know alot and even to me it doesnt seem realistic but if it was possible it would save time.