Handling Mods with the mod manager

[youtoo]

I have several ideas on how to handle mods now that the mod manager and more jsons are coming with this release. The ideas are based on

Goals:

1. make it easier for people to find the mods they want.
2. Make it easier for mod developers to promote their mods and get people to use them.
3. Little to no additional work for main branch developers. They have enough to do .

Mod Forum Change

Mod forum is how a hodgepodge of mods, discussions about mods, development discussions… there is clutter. Break it up into 2 forums atleast

1. Forum for development discussions, coding tips, main branch discussions
2. Mods in Development: One thread per mod. Hype your mod. Ask for help with your mod. Ask for tips with your mod.
3. Released Mods: 1 thread per mod. Same as #2.

#2 and #3 can possibly be 1 forum. However, I think we will get alot more ‘working on mods’ then completed and released ones. I think people who release a mod would appreciate a separate place to advertise’

Package all mods into the main game, but have them turned OFF in the mod manager

1. Cataclysm is a small download so there is little fear of it getting too big.
2. Makes it easier for people to find mods so they don’t have to download

Enhancing Mod manager to make it easier to use with lots of mods in it

1. have basic categories for mods. So when people people release a mod they give it a simple category. So we don’t get a big dump. Just list a few to start with.
   Make them in alphabetical order
2. Have an ability to see descriptions.
3. Top of description has an entry for an http link to the forum post with more details
4. Mods can be added to this that are in development, but go in an ‘in development’ tab. This way its easier for mod developers to get testers
5. Mod Developers can state which mods work well together. Possibly add a way that you click on this and you get the other mods that go with this one automatically turned on
6. Mod Pack page: Mod Pack developers can test out mods together. You can enable these that are tested by clicking this and it enables all mods in the pack
7. Need to include a latest version tested in the menu so you can avoid older mods that may not work. This would require mod developers to atleast update the version to see that its still active. You can check forums, but coloring it makes it eaiser.

Github including all mods

1. WHat do the developers think the best way is to do this? Add a flag for ‘mods’ when people upload code?
2. Goal is that main branch developers don’t manage this. Mod developers upload, manage has them set to off, and people can choose to use them, if they don’t work, they can be set to off. if there are alot of complaints, someone can change the color of a mod in red, if its known to have bugs.

[FunsizeNinja123]

Such idea, much wow.

[Binky]

There are going to be two types of mods it seems:

1. Default mods of things included in the game (modularized content from the default game which is separated on theme)
2. Content not deemed suitable for the main game (either due to being too different or not high enough quality)

I don’t see the point in putting the second type in with the normal download (unless they were very high quality) as otherwise they would be included in the main content. Also, and this isn’t meant to be an attack/moan, but most content created currently does get pulled in to the main game so there isn’t a huge need for this.

I like the ideas of having a ‘completed mods’ forum and enhancing the mod manager though.

[kevin.granade]

Seperate subforum: If someone wants to work on a mod long-term a thread is probably sufficient, if you are working on a mod and you want more than a thread or group of threads we can talk. If you have an actively developed mod and want your thread stickied just ask. Otherwise I’d prefer to wait on there being enough mod-specific threads to populate a new subforum before taking action.

Package mods with main game: Very yes, this is what’s happening by default.

More categories for mod manager: I’d like to see a bit more than 5 mod packs (what we have now) before worrying about beefing up the UI.

Merging mod content: The “github way” to do something like this is that someone developing a mod (possibly accepting contributions of their own) maintains a repository of their own and makes pull requests to the main repository. That might sound familiar, that would be because it’s how all contributions happen, there’s no reason for it to be any different for mods. What IS different is that if you’re maintaining a mod that isn’t part of the core game (dinocataclysm, some scifi mod, triffids everywhere mode, no guns, whatever) YOU are in charge of game balance, which is one of the main sticking points with reviewing new content. Really most json entries are some prose attached to a bundle of stats, if we don’t care what the stats are it becomes trivial to review, therefore not imposing additional workload. One note, if you have sexist, racist, or otherwise offensive* content, you’re on your own, we’re not touching it. However if it’s “silly” or “unnecessary” isn’t a reason to exclude it, if you want a DF total conversion mod, have at and we’ll gladly release it with the main game.

Excluding “low quality” content from inclusion with the game: There’s no reason for this, compressed json is TINY, so it won’t impact the download. And if it’s not on by default it’s up to the player to turn it on, so I fail to see how that’s a problem either. The problem is that if you force people to develop a mod outside the normal distribution channel it raises the barrier of entry unnecessarily and makes bugs more likely. Overall it just discourages contributions for no reason.

*Offensive means we the core developers find it somehow offensive, reasonable people disagree on what is offensive, so there will be disagreement on this point, that’s how it works.

[Binky]

stuff

I get that it doesn’t take up much extra space but I’d really advise against including all mods with the default download package. Granted they can be turned on and off at will, but having lots of incredibly unbalanced/possibly broken mods (which some will no doubt be) seems like a bad idea in terms of quality control.

For instance, if I made a mod that after a week of in game play caused all monsters to drop hundreds of armed mininukes at the players location as soon as they saw you, that’d probably not be a great addition. Anyone turning it on would just have a broken game after a week. Similarly, mods with really awfully designed content that doesn’t work would be similarly off putting. I get that the mods are all ‘at the players own risk’ but in my opinion a lot of people would believe that anything that comes packaged with the game will more or less work.

I do realise it’s extra work to go over everything and that it’d cause yet more aggro from the childish parts of the community, but having some sort of filter really is necessary.

[kevin.granade]

"I get that it’s at the players own risk"
You evidently don’t get the concept. A message making that clear on the mod selection screen might be warranted, but that’s as far as I’m planning on taking it.

[i2amroy]

Yeah, I’m not so sure about including mods by default in the experimental either, notably for a few big reasons:

[ol][li]If a mod stops being developed the code will just hang around until some other dev deletes it. This means that every once in a while the git managers would need to check every single included mod and make sure it’s still being developed if it isn’t compatible and to drop it.[/li]
[li]In the event any sort of thing changes that requires updating the json files we are going to end up with a whole bunch of broken mods, or a huge amount of developer time that needs to be sunk to update them. This might not happen often, and is definitely something we would try to avoid, but it is almost certainly going to happen sooner or later (if only when we update the mod manager format itself to be more detailed).[/li]
[li]Lots of PR’s. Now right now it might not be much of a problem, but as we get more and more modders with community expansion its gonna add up pretty quickly. Even if they are all insta-merge things (assuming json cooperates, it’s been known to have problems merging before) that’s still a bunch of PR’s that need to be looked at manually and that clutter up the PR and issue lists on git.[/li]
[li]Lots of modders aren’t going to want to go to the trouble of learning how to use git. With a simple thing like JSON editing, there are going to be a lot of mods out there that are just done by people who don’t know how to code and don’t want to go to the trouble of git. Making the main hub for mods being on git is going to impede mod formation since a lot of people are going to feel that they “have” to use git to get their mod distributed.[/li][/ol]

Personally I’d advocate containing maybe a few of the most actively developed and popular mods, if even that. The mod manager should (if it doesn’t already) support simple drop and load functionality, which should allow people to find their mod on the forum, download it from any of the numerous file sharing sites, and then simply drop it into their cataclysm folder and boot up the game. I don’t see any real benefits to including potentially dozens of mods into the main git when the player can easily drag-n-drop, especially when compared to some of the costs it would create (developer time, mod entry level, etc.).

[Dominae]

[quote=“i2amroy, post:7, topic:5185”]Yeah, I’m not so sure about including mods by default in the experimental either, notably for a few big reasons:

[ol][li]If a mod stops being developed the code will just hang around until some other dev deletes it. This means that every once in a while the git managers would need to check every single included mod and make sure it’s still being developed if it isn’t compatible and to drop it.[/li][/ol][/quote]

… plus, there is a firm backlash against “removing content”. Broken or not, people will flip tables if some random mod they made and abandoned gets pulled.

[ul][li]In the event any sort of thing changes that requires updating the json files we are going to end up with a whole bunch of broken mods, or a huge amount of developer time that needs to be sunk to update them. This might not happen often, and is definitely something we would try to avoid, but it is almost certainly going to happen sooner or later (if only when we update the mod manager format itself to be more detailed).[/li][/ul]

I’d be afraid that yeah, each time there was a game update it’d break half the mods and they’d have to be reworked, then re-PR’d, then remerged, etc. Much easier for players to just swarm the mod forum and bug the mod maker to upload a patch/new version.

[ul][li]Lots of PR's. Now right now it might not be much of a problem, but as we get more and more modders with community expansion its gonna add up pretty quickly. Even if they are all insta-merge things (assuming json cooperates, it's been known to have problems merging before) that's still a bunch of PR's that need to be looked at manually and that clutter up the PR and issue lists on git.[/li] [li]Lots of modders aren't going to want to go to the trouble of learning how to use git. With a simple thing like JSON editing, there are going to be a lot of mods out there that are just done by people who don't know how to code and don't want to go to the trouble of git. Making the main hub for mods being on git is going to impede mod formation since a lot of people are going to feel that they "have" to use git to get their mod distributed.[/li][/ul]

Agreed. Just take a look at, say, the Nexus website for modding big budget games like the TES series. Each mod has its own page and following and is nicely separated and all that happy stuff and STILL they fall behind in development, get left behind by better mods of the same variety, get abandoned, get found out they’re crap, get downloaded-then-you-decide-you-didn’t-actually-want-that-mod-afterall, etc etc.

Plus, I feel like a mod-splosion will happen as even just a few people get started on mods, that the PR’s will back up fast.

It is a weird spiral … you want Cata to be popular and attract people and be fun. Adding the ability to make mods will help that, but will also make more work for people manning Git. Exponentially, it could get out of hand pretty fast if the modding scene takes off.

Personally I'd advocate containing [i]maybe[/i] a few of the most actively developed and popular mods, if even that. The mod manager should (if it doesn't already) support simple drop and load functionality, which should allow people to find their mod on the forum, download it from any of the numerous file sharing sites, and then simply drop it into their cataclysm folder and boot up the game. I don't see any real benefits to including potentially dozens of mods into the main git when the player can easily drag-n-drop, especially when compared to some of the costs it would create (developer time, mod entry level, etc.).

If mods-in-progress kept the forum they’re in now, and a sub-forum of “released mods” got tacked on it’d sort them decently then from there, if a mod got a lot of traction and people decided it’d be GOOD to slip it into the main game, then someone can step in and hand them a “your mod make it to the major leagues” trophy. That is a boost to the mod maker’s ego and the game as a whole. “Hey person-X, your winter weather mod is so good we think it should get put on the list of mods that ship with the game client” is a lot more rewarding then it just automatically being merged on Git into a massive list with mods that turn all zombies into the kids from One Republic.

The community can work as its own filter, in a way. People DL and tack on the mods they like/want, and those that were bad ideas should just fade into oblivion on their own.

[GlyphGryph]

I plan on building a section of the site (rather, a subsite) in the next several months specifically to host and rate mods.

Personally, I’m opposed to packaging any mod that we don’t see as worthy of being maintained by the “core” team. Right now, that’s probably 2 or 3 graphics packs, Classic, and Dino, that I’ve seen.

I’d much rather it be clear and simple on how to acquire additional mods through a system that dynamically and effectively filters content.

Perhaps each stable release after that can include mods that get over a certain “popularity level” before the release.

If it falls below that later on, it doesn’t get included in the next one.

But these would only be packaged with the release - ‘mainline’ mods should be the only tracked in the git repo, if-even-that.

[Zireael]

I think Glyph’s idea is the best.

[John_Candlebury]

Yep I stand with Glyph on this one.

[Binky]

Yeah, I agree, and really awesome mods could certainly get merged.

To be honest though, I really hate the idea that mods would take away from main dev time/become a huge issue. I mean, every other game does fine for mods/people finding them without an official system/integration. If the game and mods are good enough they’ll get enough attention and people will find them - there’s no need to babysit (although a link to the mod forum in game/documentation seems fine). A separate download of ‘approved working mods’ could easily give a bit more attention to the full ones, and could be forum maintained (like most other games that have ‘master mod lists’ and stuff)

[GlyphGryph]

On the other hand, other games are significantly better because of their support for mods. Wesnoth comes to mind here, with mod finding tools (however poor) built right into the software.

If no one has any better suggestions though, I’ll probably throw up our own version of PHCDownload. That’s the same thing used by, for example, dffd.wimbli.com

Better suggestions appreciated, I just don’t know of any better.

[Binky]

[quote=“GlyphGryph, post:13, topic:5185”]On the other hand, other games are significantly better because of their support for mods. Wesnoth comes to mind here, with mod finding tools (however poor) built right into the software.

If no one has any better suggestions though, I’ll probably throw up our own version of PHCDownload. That’s the same thing used by, for example, dffd.wimbli.com

Better suggestions appreciated, I just don’t know of any better.[/quote]

Yeah, I definitely get what you mean, and I’m sure it does help, but I just don’t think there’s much need for loads of time to be spent on it once people can easily create mods which add/remove stuff (until the game gets a huge community/modding community)

PHCDownload sounds like a good idea though, especially as it has ratings built in.

[whiran]

I don’t really have a view one way or the other as to what mods are included with a release version.

The only thing I would suggest is that all mods are turned off by default in world generation.

One should have to actively select what mods they want to be a part of their game. No mods, at all, should be part of the default world generation configuration.

If there are mods that are considered good enough or essential to the default settings then they should be rolled into the default game.

Looking at the release candidate I noticed that there are a couple of mods that are active by default (I’m assuming that any mod under “Mod Load Order” is active which means the default world conditions include “Medieval content pack” and “Old Guns Pack” - there should be a naming convention for mods as well).

I was thinking that maybe this was a behavior of the experimental builds but it’s still there and it shouldn’t be.

[Binky]

[quote=“whiran, post:15, topic:5185”]One should have to actively select what mods they want to be a part of their game. No mods, at all, should be part of the default world generation configuration.

If there are mods that are considered good enough or essential to the default settings then they should be rolled into the default game.[/quote]

Although I get where you’re coming from (and I know some people on the forum will jump with joy that you’ve said that) - I strongly disagree.

A good portion of the content which has been added, whilst great, doesn’t really fit into a coherent theme (or is seen as too unrealistic - like lots of very old guns kicking around everywhere). Instead of arguing over what fits and what doesn’t, the mod manager allows people to pick and choose which I really think is the best of both worlds.

However, this content is deemed to be good enough that it’s a set as default, and some aspects of the game may not live up to their full potential with them turned off - such as when monsters get mod-packed the challenge may vary without a full compliment of monsters (much like classic does).

I do understand what you’re saying, and if the game had a very coherent vision and closed dev team then I’d completely agree, but the nature of the DDA community really necessitates it.

[CIB]

One should have to actively select what mods they want to be a part of their game. No mods, at all, should be part of the default world generation configuration.

That’s not a very good idea at all. With this, either “default” is going to end up the least common denominator of everyone’s preferences, or minorities who want to play with less content than the rest aren’t going to have many options at all, other than replacing the .json’s of the main mod and hoping it doesn’t blow up in their face.

Modularity is a core principle in any kind of game design, so putting stuff into mods where we can is generally a good idea. The way to display this is another matter. Maybe you’d argue that very “common” stuff like e.g. medieval weapons shouldn’t appear in the mod selection. Maybe I’d agree. But separating out the JSON’s and reducing hard dependencies still makes sense, even just from a developer’s point of view, and at that point there’s really no reason to prevent a player from deciding “yeah I don’t really want all those medieval weapons in the game, thank you very much”.

[whiran]

[quote=“CIB, post:17, topic:5185”]

One should have to actively select what mods they want to be a part of their game. No mods, at all, should be part of the default world generation configuration.

That’s not a very good idea at all. With this, either “default” is going to end up the least common denominator of everyone’s preferences, or minorities who want to play with less content than the rest aren’t going to have many options at all, other than replacing the .json’s of the main mod and hoping it doesn’t blow up in their face.

Modularity is a core principle in any kind of game design, so putting stuff into mods where we can is generally a good idea. The way to display this is another matter. Maybe you’d argue that very “common” stuff like e.g. medieval weapons shouldn’t appear in the mod selection. Maybe I’d agree. But separating out the JSON’s and reducing hard dependencies still makes sense, even just from a developer’s point of view, and at that point there’s really no reason to prevent a player from deciding “yeah I don’t really want all those medieval weapons in the game, thank you very much”.[/quote]
I’ve read over what you wrote a few times now and I think I may have written what I meant badly.

What I meant is this:

Have mods. Have a mod manager. Have lots of mods.

But, make it so that the default world does not have any mods. One would then get to pick which mods to choose from the selection presented.

I understand where Binky is coming from on this subject and that is a great point. Part of the development of this game is open ended so that things that might otherwise be considered ‘default’ are being handled by other people. That makes a lot of sense to me and upon consideration I am okay with it.

I suppose I am just of the belief that a mod is a modification to the core game. And a default generation of a game world would only be the core game not the core game with modifications. With Binky’s comments, I can see how the default generation of a world in this environment could also contain a couple of mod packs.

I just don’t see how certain mod packs get to be part of default and others do not. It seems to me that this is a path forward open to claims of favoritism and the rest of those kinds of headaches. But, if the default was just the core game then one could list as many mods as one wanted to in the mod list without worry.

By making a mod ‘default’ it is, implicitly, stating that the game devs believe that the mod should be part of the core game. If that is the case then that’s cool and I now see why they might not just take those mods and make them part of the game.

[Binky]

[quote=“whiran, post:18, topic:5185”]I just don’t see how certain mod packs get to be part of default and others do not. It seems to me that this is a path forward open to claims of favoritism and the rest of those kinds of headaches. But, if the default was just the core game then one could list as many mods as one wanted to in the mod list without worry.

By making a mod ‘default’ it is, implicitly, stating that the game devs believe that the mod should be part of the core game. If that is the case then that’s cool and I now see why they might not just take those mods and make them part of the game.[/quote]

Well made points. I think the key to it is that these aren’t ‘modification packs’ but ‘modularization packs’, and so it’s not a core modification but just modules of a core. They’re more options than mods I guess. You’re right with your last sentence that the devs believe they should be part of the main game; but the thematic differences have caused such strong divides that it makes sense to modularize it (and it helps a lot from a developing perspective)

[Soyweiser]

It seems to me that this is a path forward open to claims of favoritism and the rest of those kinds of headaches.

Why is that a problem? Some mods are just more liked by some of the devs. And not all mods are equal. I don’t see the problem.