Towards greater compatibility

[RedRooster81]

In the thread for his Armories ARKO Pack, Arko wrote this in response to requests for integrating his ongoing mod into other mods:

Aasmul resumed pretty well the situation.

-my mod is compatible with almost any mod, even some coa mods.
-the modular system for mods created by PI is a just a pure jewel not used enough.
-It gives me more liberty for updating and developping.
-On updates : this mod keeps compatibility even through patches without updates, or at least shouldn't causes bugs or crashes.
-as said I'm not against integration, I only limit it to situation where there is compatibility issues and complementary ones. (i don't see the point to mix it with a mod expandind the map in China as my mod is about western Coa).
-I try to keep some quality level and coherence for my coa and I don't want it to be mixed with coa made with the sad technic consisting in "take the first coa found on wiki, cut or resize it and voilà" which is a widespread method in mods -that can't be compatible-

Here are my though on it.


Next update soOn with.... more Coats of Arms and a new category of dynastic coa. Stay tuned !

I thought about creating a thread about this issue, for modders working on both large and small projects and non-modding players to discuss what are the best practices moving forward.

A. Integration and its pitfalls
I am seeing a pattern of interesting mods being developed and announced and instantly half a dozen requests are posted on the mod's thread begging for rights to integration. I am as guilty as any, but I am trying to show restraint lately. What I think is needed is to limit such requests until the mod is declared complete or there is some inherent incompatibility because of the files that both mods change.

Let's face it: once you say yes you can integrate my mod you lose control over what is being done with your work. I for one feel the need to have control over what I create, and I feel the need to resist having my work integrated into a huge overhaul where I've lost both creative and quality control. That is my own personal view, and the modder has the right to hold that view and it should be respected.

B. Keeping mods compatible
There is also a need to build the larger mods with more of a view towards compatibility. I have started listing in my mod release notes exactly what files I have changed and added, which I would recommend to others with release candidates. I am ambitious with my own work, but I have taken steps to increase the compatibility of my mods with others. One way for event mods is to use the new namespace feature. Another is to use less-known features like creating new files rather than editing vanilla ones, in the interface folder for example, where I learned that instead of editing the vanilla portraits.gfx, I could add a new portraits_iberian.gfx for example. The devs are also moving towards putting files in the /commons directory into separate folders, as we see now with event_modifiers and traits. So that is also something to think about.

So I am now writing my mods to take advantage of these features. I have separated my map mod from the event mod now in development for the sake of compatibility, and I hope that others will consider doing the same. In the end, what matters is that modders have control over their own creative work and that players can mix and match mods as smoothly as possible to get the experience that they want out of the game. We should consider ways to keep information current on what mods are being updated and compatibility. The mod list that knuckey is maintaining is a great way to do this, but as the modding community develops maybe more discussion will be warranted.

Thank you for reading.

 

[Arko]

I can't agree more.

Any mod that don't modifies any files in the common folder can be easily made as a stand-alone module. Hope PI will continue improving moddability to besten modularity.

Yes, it can it requires a bit more flexibilty and work to elaborate, but it worths it.

 

[RedRooster81]

Precisely, /common is the problem with keeping things compatible. Most everything else can be mixed and matched to suit a player's taste.

 

[ash001]

Some very good points, Red.

Working on Friends & Foes mod, it seems I experienced a bit of everything - added a few mini-mods to mine, had my mod invited to bigger mods (like Project Rodrigo ), and worked on having a compatible version working with other mods (like Ad Limina Apostolorum).


It's difficult to come up with a standard practice.
I can see that sometimes the easiest thing from the player's POV is to pick one mod that has everything or most things and add a few others only if they are really BADLY needed.
Just because of the risk of imbalances and, of course, lack of support from a modder overseeing everything.
That's why, although I'd love to retain some creative control, when the Game of Thrones gang asked it they could use part of my mod (and I quickly understood that only meant 2-3% of the mod), I decided to let them go for it.
I know this means relaying creative control over it but if that's what it takes to fit the whole GoT mod design, I'm down with that (their offer to pay in ca$h for my troubles didn't hurt either I'M JOKING HERE, OF COURSE!!).

Now, on a serious note, if projects are completed and finished (like most mini-mods), I don't see a problem in integration. Just my HO.

Generally, I'd say the best is to try and have modders negotiate some kind of gentlemen's agreement - where a mod version might be released to be compatible with another one, more or less what Rafek and me have been trying with each others' mods.
Oh, yes, I'm totally in favour of keeping a public record of what files have been modified and added.
In my case, I even try to keep a

### CHANGED BY ASH

as a remark alongside any changes made inside original files to let anyone know this was a modder's change.

 

[RedRooster81]

Thanks for the response, ash. I think that one approach could be using dependencies. I'll check out how that will work. But I think that if I click on my mod and your mod is a dependency of mine, then yours will be loaded, too. For that to work, everything would have to line up in terms of your mod having unique files or me making the necessary changes for a version of say Friends & Foes. The difference is that your mod does not get swallowed up into my overhaul, never to be seen again. It's something I'll look into, at least.

This is what Johan had to say about dependencies in DD#21:

- dependencies is a list of mod names. This means that you can set up a dependency chain and support having multiple mods loaded even if they might affect the same files. You can also depend on official DLC here. If dependencies are missing the mod wont be loaded.

 

[ash001]

Wow. I completely missed that (although I did read that dev diary a few times...).

I'd definitely like to know more on how to make it run properly.
Some things look like "magic" to me.

 

[Meneth]

I like this initiative.
Any suggestions as to what I as the author of an overhaul mod can do to help attain these goals?

Also: Shameless plug: The CKII IRC channel (link in my signature) is a great place to discuss this kind of stuff, and pretty much everything modding-related.

 

[RedRooster81]

Wow. I completely missed that (although I did read that dev diary a few times...).

I'd definitely like to know more on how to make it run properly.
Some things look like "magic" to me.

I'll give it a try. Apparently, you can also overwrite stuff, but the load order seems a bit fuzzy to me.

 

[chatnoir17]

Generally, I'd say the best is to try and have modders negotiate some kind of gentlemen's agreement - where a mod version might be released to be compatible with another one, more or less what Rafek and me have been trying with each others' mods.
Oh, yes, I'm totally in favour of keeping a public record of what files have been modified and added.
In my case, I even try to keep a

### CHANGED BY ASH

as a remark alongside any changes made inside original files to let anyone know this was a modder's change.

I think this statement is very meaningful.
A precise, sometimes too precise readme.txt would be also helpful for understanding the changes and the compatibility.

 

[RedRooster81]

I like this initiative.
Any suggestions as to what I as the author of an overhaul mod can do to help attain these goals?

Also: Shameless plug: The CKII IRC channel (link in my signature) is a great place to discuss this kind of stuff, and pretty much everything modding-related.

Come on, Meneth, you don't want to see my manifestos clogging up the IRC channel.

You can encourage mods that would work well with your overhaul. I'm going to check into the dependency issue, which might be a step in the right direction.

If one of the devs wants to comment, that would be great.

 

[dauncosony]

I definitely agree that we should be taking advantage of making mods modular. The problem is however is that some mods are so complex that it requires a great deal number of altered files, I think the most common changed file is opinions_modifiers.txt and landed_titles.txt. This is also why I try to keep my mods as close to vanilla as possible while keeping the core features that makes up the mod. I think the reason why a lot of mods are incompatible with others is that they seem to be "personal" mods, there the modder incorporates everything he fancies instead of only including what he thinks will make the mod different.

 

[duinnin]

I have all the respect for modders and their work. I've modded games ever since Civilization II. I'm all about putting my work out for anyone to use however they choose. No sense making people reinvent the wheel, and honestly, some things can only be coded one way. I understand some modders like to own their work, but it is just code after all.

This week I begain integrating the new titles, cultures, religions and provinces from NAE and SWMH into CK2Plus with a combined map with the Mercator Mod. The compilation should work with other balance mods, but it does edit landed_titles.txt, dynasties.txt, cultures.txt, religion.txt, and all history and map files. I was about done, but the new patch has set me back to ensure those files are up to par. In any case, I began this project because I was playing CK2Plus, with Mercator map, but I saw the new provinces, especially, in NAE and SWMH and those modders had no plans to integrate, so I took it upon myself to do it for myself.

The fun part in this journey is learning how to mod CK2 and seeing what this or that does. It has me looking into doing even more than just compiling these mods, but adding more of my own new content. Learning is a blast and modding is awesome.

 

[duinnin]

My bad, forgot to get to my main point. While working on the compilation, I discovered even having modular files in a dir can be problematic. Sure you can have separate localisation files, but this can result in a lot of overlapping code. I originally dumped all these files together, but each one may have text for k_netherlands, for example. My best option, besides comparing all the files to remove duplicates, was to put them into one file and using a website to automatically alphabetize and delete exact duplicate entries, which was a lot, then manually scroll through those remaing duplicates where in one case k_netherlands is Lotharingia or Netherlands in another. Either way a necessary hassle, but much quicker to lump together. The nice thing with alphabetized files is I can find anything immediately and merging with other mods (localisation files anyway) should be relatively easy.

 

[RedRooster81]

That's all fine and good, duinnin, but I maintain that what I'm doing is more than "just code". My point in this thread is to make it easier for everyone to maintain control over his or her code while allowing for easier compatibility. I'd rather take my mods down than have anyone integrate them into their own publicly released work without my express consent. Arko feels the same way, so I though it worth bringing the issue forward. If Paradox's team can find some good ideas, that's fine, but among the rest of us there is a code of honor that has to be followed.

@dauncosony: I agree with you on the 'personal' mod idea. That's the first tendency that I felt. But moving forward, the thing to do is compartmentalize. For the average user, you can have a portrait mod, any number of event mods, and a map mod. At the core, the most problematic file of all is landed_titles.txt. Event modifiers is being moved to a separate folder, and maybe other things will be as well. We'll see what is to come, but right now what I will do is figure out how to make mods work together. Like other things with this game, the limits need to be established, while the devs keep moving the limits.

 

[CrackdToothGrin]

RedRooset81: Okay... I have to contain my excitement... But, you inadvertently allowed me to discover and test a way to make completely new and separate portrait_properties files for my custom portrait sets... I can literally just append 'underscore'+'culture' to the filename and it works like a dream... This will allow me to keep vanilla files 'mostly' intact, and make my work truly modular.

On the subject of compatibility and ownership: I feel you. I like, nay, encourage integration of my Cuman set into whatever, but I feel the sting of losing control over it. Especially in other portrait sets. It feels like 'stealing' when that particular project took me a combined 400+ hours. That's two-and-a-half straight weeks worth of hours of research combined, and then I see other people getting credit for the hours of painting, ethnic research, translation, digging in obscure archives, contacting Turkic and Russian researchers, etc.! Hell, I even had a few minor arguments with my wife over the amount of time that I put into it... The sheer difficulty in creating entirely new faces (which entailed 558 separately created, color matched, tested pieces) was staggering.

Having true modularity would allow piecemeal, customized mods to take fold. I would love an extremely historic mod that has pieces of a bunch of different mods combined. Maybe I don't want thirty new emperor titles, but like the events of one mod... You get the picture. In either case, I will update my files to allow separate integration at the player's choice. The difficulty for many casual players to try and work everything together is limiting when we heavily modify vanilla files, and a lot of good micro-mods get shoved under the bus, so to speak.

 

[RedRooster81]

Crackd: I'm happy that was useful to you. We are learning some things here and there that make the process easier. Ideally we would have the ability for players to pick and choose what they want to use and click on all those mods in the launcher and go. I am hoping that the devs will come out with an official guide revealing all the little details like what I thought you probably knew about the interface files.

 

[Arko]

well, for the next update of my Coa MOD I'll go a step further with details and info with a new included readme.

Spoiler

MOD NAME : COA ARKOPACK
MOD/DATE VERSION : 2012 may 19
MOD TYPE : mostly graphical
OTHER MODS INCLUDED : mini-MOD Habemus Papam
MINI MOD/DATE VERSION : 2012 may 17
GAME VERSION COMPATIBILITY : v1.05f

FILES AFFECTED


/common
dynasties.txt

/gfx
//coats_of_arms
pattern.tga
+additional files

//flags
see included list
+additional files (flags for other mod -neutral if unused-)

/history
//characters
english.txt
frankish.txt
german.txt
italian.txt
occitan.txt
portuguese.txt
scottish.txt
+additional file (from Habemus papam mini-mod)
//titles
k_orthodox.txt (from Habemus papam mini-mod)
k_papal_state.txt (from Habemus papam mini-mod)
+additional file (from Habemus papam mini-mod)

/interface
coats_of_arms.txt

This coupled with a changelog.
I think it permits a better follow-up for updating packages and better information about compatibilty.

 

[chatnoir17]

A. Integration and its pitfalls
I am seeing a pattern of interesting mods being developed and announced and instantly half a dozen requests are posted on the mod's thread begging for rights to integration. I am as guilty as any, but I am trying to show restraint lately. What I think is needed is to limit such requests until the mod is declared complete or there is some inherent incompatibility because of the files that both mods change.

Let's face it: once you say yes you can integrate my mod you lose control over what is being done with your work. I for one feel the need to have control over what I create, and I feel the need to resist having my work integrated into a huge overhaul where I've lost both creative and quality control. That is my own personal view, and the modder has the right to hold that view and it should be respected.

The worst thing as original modder is, we have a difficulty by getting feedback about our works used in other overhauled mods. The users of these mega mods don't aware of which part of the mods come from where, and will post their feedback or bugreports not in the thread of original mods, but in the compilation mods.

 

[CrackdToothGrin]

Crackd: I'm happy that was useful to you. We are learning some things here and there that make the process easier. Ideally we would have the ability for players to pick and choose what they want to use and click on all those mods in the launcher and go. I am hoping that the devs will come out with an official guide revealing all the little details like what I thought you probably knew about the interface files.

If you have any more tips like that, please let me know. I've mentioned in my own thread that I have a very busy job, am married, and that I paint concept art and portraits on the side; I don't have a lot of time to read and test text files. In terms of actually playing the game, I don't think I've even amassed ten hours excluding testing mods. That whole dynamic naming thing for files is neat, but I would like to know which files that works for. Again, thanks!

The worst thing as original modder is, we have a difficulty by getting feedback about our works used in other overhauled mods. The users of these mega mods don't aware of which part of the mods come from where, and will post their feedback or bugreports not in the thread of original mods, but in the compilation mods.

Exactly. I was only tipped to the errors in some of my files when people commented about them to another mod. Errors which I would have like to know about, and that I have to fix after the fact while my release is bugged slightly during that whole time.

 

[ash001]

Exactly. I was only tipped to the errors in some of my files when people commented about them to another mod. Errors which I would have like to know about, and that I have to fix after the fact while my release is bugged slightly during that whole time.

I wonder what would be the overhaul modders' take on the posts on this thread... Some interesting issues to contemplate, for sure.