5 replies to this topic

[hekar]

I've just listened to the podcast at moddb. Funny thing is that I followed a very similar model for LuminousForts, though it's most been me working on LuminousForts and it's not really up to the high quality that Overwatch is and the community isn't nearly as large.

I used to be an avided Sourceforts player during the days of 1.9.1 - 1.9.4. When I first started playing the game, I didn't think much about it, but I eventually began to really like the mod. The Sourceforts mod has a lot of exploits that the majority of people don't realise and you can actually do really fancy trick jumps to get over walls and bases. These trick jumps make it very easy to take the flag and coupled with other things such as grav nading, playing 1v8 and winning isn't unheard of (it can be really unbalanced.)

Overall, I began to get fed up with the mod, because the original development team was kicked out, the releases were very buggy and I didn't like the direction the mod was going (or lack there of direction.) Therefore, I started this mod a few years with the intention of making it completely open source with the exception of artwork. At the time, I was young and really inspired by Richard Stallman and GNU. I was also an avided reader of FreeGamer and I began to believe this could change the mod scene forever.

I knew I could fix more than half the bugs found in Sourceforts 1.9.3 and I thought I could do it in less than a month. Problem was, I didn't want to join the team and I didn't have access. There was also no point, because there was a lot of talk about killing Sourceforts and making a new mod. I decided I'd start a clone of Sourceforts, but I quickly realised it was a lot harder than I thought and gave up. It wasn't until a little less than a year ago, that I started seriously working on a Sourceforts clone by myself.

Everything is redone from scratch and focus has recently shifted from being a Sourceforts clone to being a totally new game, with new mechanics previously unexplored.

The code is licensed under the MPL, because the MPL basically denotes that open source files can coexist with proprietary files (each file has its own license, either the Valve license or the MPL.) The MPL is pretty easy to abide by. Just keep the header at the top of the source file and release that file as open source. Any modifications to Valve files are fair game to take without credit.


We current have a lot happening under the hood, but we're most likely not going to see a release until late November or December (I also have a full-time job and 2 part-time jobs, so I'm very busy.) We're making major changes to the website, adding a wiki, a bug tracker that'll be integrated with the game, etc. We're also working on a ranking system that'll be integrated with the website. I just keep getting surprised with the cool ideas people come up with. Things that would have never happened with Sourceforts, because the developers became so narrow minded and against external ideas.

Anyway, I don't expect anyone to contribute, but using source code from LuminousForts is pretty easy. The source code repository is actually just a differential against the Source SDK and you need to have the Source SDK, along with run a script to merge LuminousForts and the Source SDK together. I already wrote the script, it's written in Python. I did this because of legal concerns. As a side effect, it's really easy to read the LuminousForts source code, because you can just ignore the Source SDK and see only what's been added to the mod. If you want access to the full repository that's fine too. Just private message, add me on steam or email me, etc.

(Sorry for the long post, but I got kinda excited when I saw another open source mod, especially one of this calliper)

LuminousForts Website
LuminousForts Codebase

[The Joker]

hekar, on 10 August 2011 - 09:12 PM, said:

I've just listened to the podcast at moddb. Funny thing is that I followed a very similar model for LuminousForts, though it's most been me working on LuminousForts and it's not really up to the high quality that Overwatch is and the community isn't nearly as large. [...]

I don't believe Overwatch is open source. Where did you get that idea?

[AndY]

The Joker, on 11 August 2011 - 04:57 AM, said:

I don't believe Overwatch is open source. Where did you get that idea?

It's not yet, but as we mentioned in the podcast, that's where we're headed.

Also, Tero just released the source for his mod: http://hlssmod.net/ - It's a step in the right direction, but not exactly what we had in mind.

What we were discussing in the podcast was that mods should develop their titles openly, basically giving the community access to the codebase as they develop. Ideally, this would attract programmers from the community that aren't necessarily looking to join a mod team, but that might be interested in contributing various pieces of code. At the same time, giving others access to the code might spark ideas in them (spin-offs or even completely different mods). You obviously need a good system in place for this to work well.

At the very least, giving the community early access to your game (as you develop it) is a great way to build your fanbase and get tons of valuable feedback in return. Overwatch wouldn't have been the same had we chosen a different route. (Releasing your code once the project is complete is also a great way to give back to the Source dev community).


Hekar: I definitely think you're doing the right thing. People might not show interest now, but once you release and your mod gains some popularity, you're more likely to find willing contributors. Keep us posted on your progress!

[hekar]

The Joker, on 11 August 2011 - 04:57 AM, said:

I don't believe Overwatch is open source. Where did you get that idea?

AndY, on 11 August 2011 - 08:31 AM, said:

It's not yet, but as we mentioned in the podcast, that's where we're headed.

No wonder I couldn't find any source code! I was wondering how many external contributors Overwatch was getting. I'm actually really busy right now, so I was working while I was listening to the podcast and I missed the point where they said they haven't open sourced yet. It was about floss, so I had just expected that they did open source.

AndY, on 11 August 2011 - 08:31 AM, said:

Hekar: I definitely think you're doing the right thing. People might not show interest now, but once you release and your mod gains some popularity, you're more likely to find willing contributors. Keep us posted on your progress!

Yeah, I think mods just need to push open source into the modding culture and that podcast was a first step in the right direction. It's something that's foreign to most mod players and they don't understand the concepts behind open sourcing. They also, usually don't have the skills necessary to help develop mods and they fail to see how they can be a part of it. In spite of this, I think we're going to see a lot more mod players learning programming, modelling, mapping, etc and there's going to be a shift to a different type of development. One where the community is a lot more active.

Developing from a community, rather than a team will give mods a serious edge over commercial games. It allows mods to be more creativity and to have their assets live forever. Rather than die off when a few team members leave and there a failure to find replacements, people will always come and fix mods they love. It also helps the community get the game they want, rather than continually having to ask developers to do things for them. If you're not paying me, it's not going to happen and I'm not going to buiild feature XYZ. Most importantly, we have to finish and release mods and this requires effort from everyone. I don't care how good your idea for a game is, if it doesn't exist, it's useless. I can't play ideas and they're not fun.

Of course there are people that are totally against this philosophy and that's fine, but the rest of us need to push ahead and make open sourcing the status quo for modifications.

[AndY]

Agreed. Unfortunately, for this to work, you need the project(s) to be well-coordinated, or you end up with a huge mess rather than a game.

[hekar]

Well yeah, you still need a core set of maintainers. Most open source projects are like this. A core set of maintainers decide what gets in and what doesn't. These people are usually really dedicated contributors. In the case of the mod, this is just the traditional team and they have control over the SVN, website, etc.

I also argue that you need a dictator that anyone can come to when the mod has problems. This person doesn't need to design the whole game or micro management everything, but when disagreements occur and everything goes wack, people need someone to make the final decision. Usually this person is a programmer or a really good artist.

Of course the dictator has to try to make everyone happy, because the mod developers have to do everything in their power to prevent forking and the scattering of a community. Really open source puts the developers at a greater mercy to the mod players, but it also demands more responsibility and effort from mod players to make sure the mods they love progress.

If you setup a system where you vote in the leader trouble might ensue, because that leader will have to cater to everyone's wants and it'll rip the mod apart. For now LuminousForts is a just an experiment/prototype, so there's nothing really big and the team is really laid back, but we do have a dev team and we do get together and chat.