Why is CM closed source?

[Challenger]

I don't know if you're being dense on purpose. But I'll respond

We're talking about digital goods. Also open source game code. This means my point has nothing to do with piracy. It feels like you're putting piracy out there just to try and discredit what I'm saying. You can't pirate open source code.

You seem to be stuck the house analogy and not the part where if a dev, coder, spriter, etc. leave the game they have no say over what happens with the game. They can't suddenly be important when up to that point their absence has made them inconsequential to the games current direction.

No pirate has ever said anything about making a game better because they're pirating the game. You're thinking hackers and modders. Which has actually helped games. Take Nier:Automota and Gearbox's Colonial Marines as examples. The entire point of open source is to make a good game out of a good game and you seem to ignore the fact that CM is a result of an open source code.

We're confusing each other with the piracy thing and it's not even leading anywhere so whatever. Let's just talk about CM and the devs.

As far as I understand you, you have a "practical" perspective for how CM and CM's devs interact with each other: CM should only give a shit about what the devs want so long as the devs have leverage over CM, namely being able to withhold future contributions from CM. Hence retired devs' wishes don't matter because they're not gonna contribute any more. This is basically how most open-source codebases work.

What I'm understanding is the case, according to the staff in this thread, is that they have an implicit agreement/contract: the devs provide work, and in exchange CM gives them "a say in the vision" that lasts beyond their departure.

CM open-sourcing would break their end of the "contract" because the devs' work would no longer be in control of a party that respects the devs' wishes, which is why they don't want to do it.

[KingKire]

Well, it seems it really was an actual compile-able code. That leaves the question of the future of cm, a little bit in the air. Will it burn everything down, or will it be the fire that lights innovation across the cyber seas... I dub thee, the CM2018 Prometheus code branch. lets hope this life goes better than it did in 2016... huh, ironicly the year when we diverged from the standard timeline and entered trumpolics.

[Casany]

Well, it seems it really was an actual compile-able code. That leaves the question of the future of cm, a little bit in the air. Will it burn everything down, or will it be the fire that lights innovation across the cyber seas... I dub thee, the CM2018 Prometheus code branch. lets hope this life goes better than it did in 2016... huh, ironicly the year when we diverged from the standard timeline and entered trumpolics.

Well, the new code doesn’t actually have a file that it needs to work. So a dedicated cider may be able to run a CM server, but anyone else who tries will he missing a vital part.

[Nickvr628]

Then how come I can go play the leaked server right now, courtesy of BR?

[Casany]

Then how come I can go play the leaked server right now, courtesy of BR?

Oh it can LAUNCH but half of the things won’t work. I spoofed through the git and basically while you can hurt rig it to work horribly without the three files that are missing, or at least the main one it isn’t gonna work nearly as well

[Khlorghaal]

https://spectrum.ieee.org/semiconductor ... esign.html
A good read about why open source is generally an ethical imperative.
Especially with engineering IP, sharing things only makes development for everyone better. Yeah competitors will use the tech to increase their share of sales, but their contributions to the net result outweigh the lost sales.


I notice lots of people approaching it ethically as 'If someone volunteers work why should they be forced to share it', but with free / public domain IP that work is not diminished with the consumption of its product. Rather it should be approached as 'If you volunteer work into something, why would you not want to share it? If not volunteering for the purpose of sharing, then for what purpose?' If a project has altruistic intent, then its utility is increased with its proliferation. Which is the opposite of commodities, which is why I think most people don't realize this, as people instinctively (literally instinct, this behavior is observed in other animals) only think in terms of the labor->exchange->profit, instead of labor->utility.

Yes it is someone's right to hoard their IP to themself, but if the IP is for a public domain product they have absolutely no reasonable motive to do so.


But thats open source in general, while CM is an exception because its source proliferation would have potential to make hard forks which dilute the playerbase, which would decrease its value, along with other things mentioned.
BUT theoretically this could happen with ANY open source game, and as far as I know there are absolutely no cases of this happening; because the consumers have no motive to use the fork unless it makes very large improvements.
And in the case of a fork making improvements that the original was incapable of achieving, then it would be a net positive. But that is unlikely, as in pretty much all cases, the original authors of a product are the most qualified to design and maintain it.

if the source of CM was leaked again there could be no legal trouble

Note I'm not a lawyer and none of this is legal advice.
LEGALLY there isn't even an entity which owns the CM's IP, thus the IP belongs all rights reserved to the original authors, so they could choose to pull it arbitrarily, meaning spooky's decision to require unanimous approval is warranted.
BUT bay12 is GNU AGPL, meaning all derivatives of it must be open source, meaning the CM's legal status is additionally contingent on the implied license from the owner / all the authors of bay12.
Also CM's leak contains a file stating that it is also GNU AGPL, which is very comical.

But legal matters are irrelevant, if any of this were enforceable then getting contributors would be as simple as having them sign an NDA.

[Challenger]

I remember Apophis775 posted something on reddit about some SS13 server that was open source. It got forked due to a split in the community or someshit, and the forkers managed to get a large share of the playerbase to play on their server, and the forked codebase tracked the original codebase. Both servers eventually entered a death spiral and died, because the devs' contributions didn't do anything to increase the original server's playerbase as the parasite server would run the same code and retain its players, and they didn't want to contribute to the parasite server for whatever reason, so they simply stopped contributing as a result. Unfortunately he's deleted his account since.

A basic reason to hide your IP is, like Nicboone's said, that you get to decide the context in which it's used. I might be willing to develop content involving Nazis if it's going to be used in a mature, historical context, but not if it's used by Lebensraum tier trolls. I might be willing to develop mechanics that create an interesting story if it's going to be placed in a server where it plays out in a rare and patient manner, but not if it becomes the daily rotation of a TG-tier ADHD fest. Being able to determine the context and community that your content is used in is a reward for developers' work, and they might not be willing to develop without it.

Goon, CM, and the russian guy's server are some of the most unique and polished servers in SS13. Is it a coincidence they're closed source?

[NethIafins]

I remember my mentor times on Paradise server, long before we got "directive" not to give out any information from the code, and tell that "if you want to ruin it for yourself, go ahead"

So we had:
1) Us telling them EXACT damage of every single thing in game if they asked
2) Telesci roundstart getting removed for being TOO FUCKING GOOD because of the prevalence of reverse-engineered scripts that were giving you what to write to teleport shit. I mean if you are antag, you can steal ANYTHING. If you are loyal, you can teleport people WHILE THEY ARE BEING KILLED (I done that more than once)
3) Chemistry and plasma being absolutely almost useless, because you can just check code and find out what is the best

The only thing that saves open servers is a great amount of RP and non-combat based gameplay style. You can have open source combat based game but it has to be super competitive (like OpenRA or DooM II) and balanced.

But this is just my opinion

[NoahKirchner]

I remember my mentor times on Paradise server, long before we got "directive" not to give out any information from the code, and tell that "if you want to ruin it for yourself, go ahead"

So we had:
1) Us telling them EXACT damage of every single thing in game if they asked
2) Telesci roundstart getting removed for being TOO FUCKING GOOD because of the prevalence of reverse-engineered scripts that were giving you what to write to teleport shit. I mean if you are antag, you can steal ANYTHING. If you are loyal, you can teleport people WHILE THEY ARE BEING KILLED (I done that more than once)
3) Chemistry and plasma being absolutely almost useless, because you can just check code and find out what is the best

The only thing that saves open servers is a great amount of RP and non-combat based gameplay style. You can have open source combat based game but it has to be super competitive (like OpenRA or DooM II) and balanced.

But this is just my opinion

The only things here that are applicable to CM are 1 and, to some degree, 3.

Damage values can be hidden serverside with a config. If it's not needed to compile the server it's not necessary to be open to the public under AGPL so you can keep damage values, chemistry recipes etc hidden from the general populace.

Chemistry on CM is already heavily babyproofed, all of the best recipes are public knowledge on the wiki and the issue with making them isn't a lack of knowledge generally it's a lack of knowledge on part of the chemist and, occasionally, a lack of materials. All chemicals in CM are useful in some way or another, either as a supplement to other meds or stuff to give grunts because it's easy to produce. Even still, you can hide all of these recipes behind a serverside config as written above because as long as the server compiles without it, it's not necessary to include under AGPL iirc.

As for your second thing, I understand that it wasn't aimed directly at telesci but there are really no roles left that could have such an impact on the round. Researcher chemistry/weapon stuff was gutted, atmos was gutted, engineering was gutted, combat stims etc were gutted, there's no way for shipside roles to really contribute much to a round directly because all of their means to do so were removed. Some were certainly justified and I'm not making a case for or against their removal or reinstatement, but things like this almost certainly aren't an issue in CM because the good tactics rely more upon timing and common knowledge than they do knowing specific variables.

--

The thing that saves open source servers, specifically referencing the larger codebases like bay, tg, the vore fork of polaris, aurora, para etc is that they are being constantly updated and developed, TG today is vastly different from TG a year ago in a staggering amount of ways and exists in a lowRP enviroment with a lot of combat based gameplay and has two servers that sit at 50 and a european server that when I've seen it gets 20. You can also argue that keeping the code closed is better for development direction, but you can just keep anyone but devs from being able to contribute, have the code open and can then port features from other codebases to enhance the CM experience. Even more minor stuff like backend code or finally getting SQL nonsense finished so that we stop losing our char prefs every time the server goes bad, this stuff takes a long time to code on its own but porting it speeds that up a lot.

[Khlorghaal]

Apophis775... posted something on reddit about some SS13 server that was open source. It got forked due to a split in the community or someshit, and the forkers managed to get a large share of the playerbase to play on their server, and the forked codebase tracked the original codebase.

In theory I was thinking, why would people prefer the fork over the original, and why would the original not merge the fork, but I guess things aren't that straightforward in practice.
So that instance is excellent evidence.

[NoahKirchner]

I remember Apophis775 posted something on reddit about some SS13 server that was open source. It got forked due to a split in the community or someshit, and the forkers managed to get a large share of the playerbase to play on their server, and the forked codebase tracked the original codebase. Both servers eventually entered a death spiral and died, because the devs' contributions didn't do anything to increase the original server's playerbase as the parasite server would run the same code and retain its players, and they didn't want to contribute to the parasite server for whatever reason, so they simply stopped contributing as a result. Unfortunately he's deleted his account since.

A basic reason to hide your IP is, like Nicboone's said, that you get to decide the context in which it's used. I might be willing to develop content involving Nazis if it's going to be used in a mature, historical context, but not if it's used by Lebensraum tier trolls. I might be willing to develop mechanics that create an interesting story if it's going to be placed in a server where it plays out in a rare and patient manner, but not if it becomes the daily rotation of a TG-tier ADHD fest. Being able to determine the context and community that your content is used in is a reward for developers' work, and they might not be willing to develop without it.

Goon, CM, and the russian guy's server are some of the most unique and polished servers in SS13. Is it a coincidence they're closed source?

I do vaguely remembering hearing that story as well, but the kind of community split required to do that is pretty earth shattering. Paradise has their fair share of drama as well and are open source, with a lot of complaints about how they handle staff situations, but despite this nobody has made a fork and tracked it because doing that requires a solid staff team, forums, servers and a direction for the server still. If everyone jumps ship, perhaps it was meant to be.

--

That last bit there is a valid point, but as far as I can tell it's the only refutation for going open source. In every other aspect, being open source is superior to being closed source because a closed source project stifles progress. Sure, in the past year CM has gotten some beautiful sprites and awfully neat mechanics, but compare those few large additions supplemented with a number of smaller additions to some of the leaps and bounds made by other open source codebases. Optimization is always happening in open source because you have a large number of people working towards a goal, bugs are fixed at a much faster rate and you get more additions generally. Even if you go with the route of disallowing other people to contribute, which would still allow the devs to steer server direction like you mentioned being an issue in your post (I recognize that it was more aimed at forks, though), you can still port code from other servers to better optimize your codebase and bring it into the modern.

[Khlorghaal]

Optimization is always happening

CM is pretty performant due to no atmos though. The only time things get laggy are expected stuff like round start and DS crash. And the leaked code I looked at from years ago was surprisingly nice and procedural.
Only big improvement would be latency, but Byond itself is probably the limiting factor there.

The big thing which needs gauged here is, tangibly, how much contribution is the codebase actually missing due to not being open, and I'm estimating from what I've read here, probably not a ton, or at least little enough that it outweighing the benefits of being closed makes it contentious.

[NoahKirchner]

CM is pretty performant due to no atmos though. The only time things get laggy are expected stuff like round start and DS crash. And the leaked code I looked at from years ago was surprisingly nice and procedural.
Only big improvement would be latency, but Byond itself is probably the limiting factor there.

The big thing which needs gauged here is, tangibly, how much contribution is the codebase actually missing due to not being open, and I'm estimating from what I've read here, probably not a ton, or at least little enough that it outweighing the benefits of being closed makes it contentious.

You can always improve optimization. Explosion code taken from other servers could help speed up the dropship crash and I'm sure other servers have found ways to optimize lighting so that effects like light flickering won't be so laggy.

CM is, as far as actual features go, frankly pretty barebones. They have no R&D, no Atmos, a stupidly easy engine system, there is no mining and even a lot of construction/deconstruction has been entirely gutted, note that all of these are things that CM once had. What replaced it was some great sprites, a tank and a reworked orbital system. A tank is just a slightly modified and resprited mech, still cool as hell and I'm certain that it took a fair bit of work to get up to scratch but in terms of actual feature density CM is incredibly lacklustre. All development goes towards combat with basically no additions to shipside roles aside from tweaks, removals or very minor additions at the cost of the aforementioned.

[Casany]

You can always improve optimization. Explosion code taken from other servers could help speed up the dropship crash and I'm sure other servers have found ways to optimize lighting so that effects like light flickering won't be so laggy.

CM is, as far as actual features go, frankly pretty barebones. They have no R&D, no Atmos, a stupidly easy engine system, there is no mining and even a lot of construction/deconstruction has been entirely gutted, note that all of these are things that CM once had. What replaced it was some great sprites, a tank and a reworked orbital system. A tank is just a slightly modified and resprited mech, still cool as hell and I'm certain that it took a fair bit of work to get up to scratch but in terms of actual feature density CM is incredibly lacklustre. All development goes towards combat with basically no additions to shipside roles aside from tweaks, removals or very minor additions at the cost of the aforementioned.

What does CM need tho. They don’t need mining (where would they even do that) they don’t need atoms (lag for the sake of lag, Space isn’t used for anything else other than actual atmosphere, like the way the game feels.) R&D is needed but not like it works on other servers, I’ll give you that. Construction got a huge update, for building barricades and FOBs. That’s not barebones at all. Engine system yeah I’d like a rework there. So you have two things that need a rework, research and engineering. But look at what they DO have compared to other servers? A quite good combat system, even if a bit bullet spongey, entirely reworked and new alien castes and alien system, OBs in general, the tank in general (and when looking at the code it doesn’t seem to be a reworked mech, seems to be its own entity) all sorts of weapons, armor and other combat items.

Like yeah all the things you mentioned are needed for a normal SS13 server, but they don’t fit in CM. At least the ones mentioned. I get it you like the the TG style of adding tons and tons of new systems even if they aren’t optimized or even if they fit or not. But that’s why TG exists, for those wacky and zany antics. CM is combat, it’s literally called Colonial Marines. If you come expecting more I guess it’s just not gonna work.

[Jonesome]

But look at what they DO have compared to other servers? A quite good combat system, even if a bit bullet spongey, entirely reworked and new alien castes and alien system, OBs in general, the tank in general (and when looking at the code it doesn’t seem to be a reworked mech, seems to be its own entity) all sorts of weapons, armor and other combat items.

Yeah but all those are for combat classes. If you're playing a shipside role things can get pretty boring. The Almayer feels kind of empty. Honestly if I had the choice of choosing between Sulaco and Almayer - even with the new fantastic additions of Air support, tanks, and the overwatch system - I would still choose the Sulaco for the sheer unpredictability and creativity of individual rounds.

[NoahKirchner]

What does CM need tho. They don’t need mining (where would they even do that) they don’t need atoms (lag for the sake of lag, Space isn’t used for anything else other than actual atmosphere, like the way the game feels.) R&D is needed but not like it works on other servers, I’ll give you that. Construction got a huge update, for building barricades and FOBs. That’s not barebones at all. Engine system yeah I’d like a rework there. So you have two things that need a rework, research and engineering. But look at what they DO have compared to other servers? A quite good combat system, even if a bit bullet spongey, entirely reworked and new alien castes and alien system, OBs in general, the tank in general (and when looking at the code it doesn’t seem to be a reworked mech, seems to be its own entity) all sorts of weapons, armor and other combat items.

Like yeah all the things you mentioned are needed for a normal SS13 server, but they don’t fit in CM. At least the ones mentioned. I get it you like the the TG style of adding tons and tons of new systems even if they aren’t optimized or even if they fit or not. But that’s why TG exists, for those wacky and zany antics. CM is combat, it’s literally called Colonial Marines. If you come expecting more I guess it’s just not gonna work.

I'm not saying that they need any of these things, what I'm saying is that features which gave depth to the game and were removed never had equally deep features added, other servers have had the equivelant of guns and combat items added alongside major ports like integrated electronics, goonchat, bay servers adopting and refining very well functioning multi-z and certainly a lot more that I don't even know about. This plus minor additions to all of the departments that base SS13 already has does not even begin to compare with the feature deficit in CM, the only code at all that they have worth taking is projectile code and niche things like OB, SO consoles and tanks. Beyond that everything in CM aside from the sprites, another server already has. Their xeno castes just utilize abilities that were either already allotted to the xenos which came in base ss13, like building, spitting or fighting and adds a pounce, a charge and boiler gas.

[spookydonut]

I'm looking to re-add complexity to some parts of the gameplay but understably our time is limited by a dozen other things that need doing.

[NoahKirchner]

I'm looking to re-add complexity to some parts of the gameplay but understably our time is limited by a dozen other things that need doing.

Of course and I'm not trying to knock your work here by any stretch, but the CM development is slower because of closed source is the point I'm trying to make.

[spookydonut]

Of course and I'm not trying to knock your work here by any stretch, but the CM development is slower because of closed source is the point I'm trying to make.

It would be even slower if I had to go through a bunch of PRs

[NoahKirchner]

It would be even slower if I had to go through a bunch of PRs

You could see people adding good code, add them as contributors and have the dev team manage PR review. Or you could only allow contributors to make PRs, make all devs contributors and then, since you're open source just not community developed, be able to port things from other servers that would otherwise require massive dev time or have people able to give you more comprehensive bug reports.

[Casany]

You could see people adding good code, add them as contributors and have the dev team manage PR review. Or you could only allow contributors to make PRs, make all devs contributors and then, since you're open source just not community developed, be able to port things from other servers that would otherwise require massive dev time or have people able to give you more comprehensive bug reports.

But just as well porting things over wouldn’t really work. You can’t just copy paste TG code in and have it work, sadly. I’ve tried. So the whole point of adding in new things for optimization won’t work because they themselves need to be optimized. And I know from experience that adding something in in a totally different area can sometimes just break code, happened whilst I was coding the other night.

[NoahKirchner]

But just as well porting things over wouldn’t really work. You can’t just copy paste TG code in and have it work, sadly. I’ve tried. So the whole point of adding in new things for optimization won’t work because they themselves need to be optimized. And I know from experience that adding something in in a totally different area can sometimes just break code, happened whilst I was coding the other night.

Porting is not as easy as copy pasting, it requires you to "stitch it in" so to say. It's the same idea of implanting an organ, if you just out someone's organ and drop another one in, it's not going to work, you need to hook it up to the rest of the body properly to ensure that it works, likewise with coding, but it takes a lot less long in most cases to integrate a feature than to design an entirely new one.

[spookydonut]

Porting is realistically only something one of our current dev team could do, anyone else trying to do it would either be outright rejected for code quality or take so much of our time to improve it/give feedback that we might as well do it ourselves.

I won't lie, if you find a code related bug in the leaked code and I can find the same bug in our actual code, I won't be upset about you making a gitlab issue/bug report about it.

[NoahKirchner]

Porting is realistically only something one of our current dev team could do, anyone else trying to do it would either be outright rejected for code quality or take so much of our time to improve it/give feedback that we might as well do it ourselves.

I won't lie, if you find a code related bug in the leaked code and I can find the same bug in our actual code, I won't be upset about you making a gitlab issue/bug report about it.

Aye aye.

[Khlorghaal]

It would be even slower if I had to go through a bunch of PRs

Yeah productivity is definitely not proportionate to number of devs. Especially merging anything that changes interface, good luck to anyone who is on a downstream branch.
But at the same time, you're impling that having to check and approve someone's work is slower than doing all of it yourself? I understand where you're coming from, but you're likewise complaining about having too many people willing to help.
If you're worried about low quality PRs, just strictly enforce your already existing design/style guides, and delegate it to an approval process that sends it to lower ranking devs first muh social hierarchies.

be able to port things from other servers that would otherwise require massive dev time or have people able to give you more comprehensive bug reports.

I don't see feature porting as being a huge advantage, unless the person from the other server was the one porting their own code. I feel like with the bit of DM ive glanced at so far, the individual features are really simple if not trivial classes, but theres significant challenge in keeping it contained and consistent with the rest of the codebase. Since DM is dynamic type and essentially a modding environment, there is huge potential for people to jury rig implementations, use inconsistent design patterns, etc that if you tried to shovel into a repo could quickly lead to pastafuck.
I see Datums as essentially a global prototype, and hooooh that could lead to maximum badness if you have people who don't know what they're doing touching that.