An open letter from the Kingdoms of Arda team, and the total-conversion mod community

[Jance]

Hi!

A subset of total-conversion modding community has penned an open letter regarding some serious issues we have with Bannerlord's code, and it's development in general. We are extremely frustrated with how the vast majority of our feedback to TaleWorlds has been handled and frankly, discarded. It's been nearly a year since early access was released, and we've seen little to no work on issues we have expressed as critical to our mod's development. In addition to our feedback not being taken seriously, TaleWorlds leadership has taken it upon themselves to make decisions with adverse effects on total-conversion mod development without consulting the community (as detailed in the letter).

We hope this letter provides some insight on the widespread frustration our community's developers are currently feeling, and that TaleWorlds will start considering their stakeholder's feedback with more weight going forward.

Here is the link to the complete letter. We encourage any criticism, support, or whatever commentary you may have to be expressed in this thread.


Sincerely and respectfully,

The Total-Conversion Community

 

[sakirma]

+1

 

[Xylord]

I am a programmer for the total conversion mod Shokuho. This letter may seem harsh at a glance, but it truly comes from a place of love and passion. We want Bannerlord to reach its full, tremendous potential, and the problems detailed here affect this process significantly. It also offers possible solutions to those issues that are, while non-trivial, also practical and realistic.

 

[Jance]

I am a programmer for the total conversion mod Shokuho. This letter may seem harsh at a glance, but it truly comes from a place of love and passion. We want Bannerlord to reach its full, tremendous potential, and the problems detailed here affect this process significantly. It also offers possible solutions to those issues that are, while non-trivial, also practical and realistic.

Indeed. The intention here is entirely constructive, and in the end we all just want Bannerlord (and its mods) to thrive

 

[Arktos]

Thanks Jance! Greatly appreciated effort.

 

[Peanut_Brother]

+1

"Help us help you make the sequel that leaves Warband in the dust."

Say it louder for the people in the back! ?

 

[Bloc]

The letter makes valid points, but all of you are missing main goal. The game is still in EA and their primary focus is "releasing" the game. So they don't care about this at the moment. This doesn't mean that they don't care about mods - this means, as a primary objective, they don't care about the mods and/or side effects of their updates to those mods.

Once they are "done" with adding whatever they think is good, afterwards they can increase the accessibility of the modules/classes. As far as I can tell, Bannerlord is not following any known architecture or design pattern. Hence, many of the classes/modules are simply "read-and-learn" type of code. So even if they make it accessible, you have to re-read what they changed ( codewise ) in each update. Then the next thing you will complain about is the sudden changes, no deprecation annotations etc. As you can see, this can go on like this. I'm a fan of Total Conversion mods but I wouldn't recommend anyone to waste their time for this unstable mushy codebase at the moment. Let them release and stabilize the game first.

If there are any statements regarding to this - saying that they are planning to make the game as moddable as possible even though it's in the EA, then please, provide me some quotes - I'm currently unaware if there are any.

 

[Lonewarrior]

As someone working on a large mod. It has been sad to see over time some essential modding tools have not been fixed when reports were made a long time ago.
A number of issues are holding back my mod and despite getting together with some of the best modders out there none of us can find workarounds to some issues that I feel should have been addressed months ago.

I hope this post will get the attention it needs in order for us to continue to develop great mods, and is no way a dig at the team behind the game.
It’s just very disappointing for me that so much potential is being lost at this time due to things holding modders back.

 

[Jance]

The letter makes valid points, but all of you are missing main goal. The game is still in EA and their primary focus is "releasing" the game. So they don't care about this at the moment. This doesn't mean that they don't care about mods - this means, as a primary objective, they don't care about the mods and/or side effects of their updates to those mods.

Once they are "done" with adding whatever they think is good, afterwards they can increase the accessibility of the modules/classes. As far as I can tell, Bannerlord is not following any known architecture or design pattern. Hence, many of the classes/modules are simply "read-and-learn" type of code. So even if they make it accessible, you have to re-read what they changed ( codewise ) in each update. Then the next thing you will complain about is the sudden changes, no deprecation annotations etc. As you can see, this can go on like this. I'm a fan of Total Conversion mods but I wouldn't recommend anyone to waste their time for this unstable mushy codebase at the moment. Let them release and stabilize the game first.

If there are any statements regarding to this - saying that they are planning to make the game as moddable as possible even though it's in the EA, then please, provide me some quotes - I'm currently unaware if there are any.

their primary focus is "releasing" the game.

That is exactly the point. They are tunnel visioned on expanding the game rather than refining it.

Documentation is a necessity at this stage of development. Systems are well established (at least at a high level).

We also have no problem adapting our code to changes made with updates. Currently since we need to rip out and replace entire systems to implement them in our own assemblies, we have to make exponentially more changes to make our code compatible than if we were able to extend classes with atomic/modular code.

 

[eddiemccandless]

That is exactly the point. They are tunnel visioned on expanding the game rather than refining it.

Documentation is a necessity at this stage of development. Systems are well established (at least at a high level).

We also have no problem adapting our code to changes made with updates. Currently since we need to rip out and replace entire systems to implement them in our own assemblies, we have to make exponentially more changes to make our code compatible than if we were able to extend classes with atomic/modular code.

The thing is though that the game doesn't need to be refined. The game needs to be finished. It's not done yet. I sympathize, and I definitely want the finished game to offer as much support for modding as possible, but I think that what you are asking for simply can't be done at this time (at least, not by Taleworlds).

Keep in mind, most of the complaints that you bring up in the letter (which I didn't think was harsh, everything you say is just common sense) are the product of bad coding practices on their side. I simply don't see them being able to build a modular system like the one that you ask for. It would be great, it would probably be better for the overall architecture of the game itself, but I just don't think that it's going to happen. If they tried to do it they would drown even more than they are drowning now.

Now with that said, if you want for someone to actually answer this you probably want to tag one of the community managers, e.g. @Dejan (well, I suppose I just did it ).

 

[Bloc]

That is exactly the point. They are tunnel visioned on expanding the game rather than refining it.

I agree with you. But we also have to be realistic. TW is tunnel visioned since from closed beta and they will always have an EA shield about such topics. Once EA is out of the way, then they have no excuses.

As a person who worked on mods for BL since from Closed Beta, I don't think it would be logical for a company to waste their resources on this. This is not good for company or good for modders.
I don't think EA will last another year, they will release a somewhat stable game in 2021. With lame features? Yes. But that's how TW plans the finish the game. In this scenario, it's not logical to make this change while they can simply put all effort to make it stable as soon as possible.
If EA lasts longer, in that case it's also not logical to put effort on this since, as a modder, you will always have to update your code and read TW's codebase over and over to understand what has changed ( and why ) - you are saying that it's okay for you to adopt all these changes but it's not okay for the company to always think about this aspect while doing something or doing something experimental. They are not making an engine - it's a game, and it's in EA.

But as I said, would it be better if TW suddenly decides to go that path by accepting all the effort this might cause to them? Yes - for the modders and for the community. Since TW won't add anything meaningful into the game anywhere soon, mods are the only hope. But will they do this? Probably not - and as I explained, they don't have to do that anyway.

 

[Jance]

The thing is though that the game doesn't need to be refined. The game needs to be finished. It's not done yet. I sympathize, and I definitely want the finished game to offer as much support for modding as possible, but I think that what you are asking for simply can't be done at this time (at least, not by Taleworlds).

Keep in mind, most of the complaints that you bring up in the letter (which I didn't think was harsh, everything you say is just common sense) are the product of bad coding practices on their side. I simply don't see them being able to build a modular system like the one that you ask for. It would be great, it would probably be better for the overall architecture of the game itself, but I just don't think that it's going to happen. If they tried to do it they would drown even more than they are drowning now.

Now with that said, if you want for someone to actually answer this you probably want to tag one of the community managers, e.g. @Dejan (well, I suppose I just did it ).

Arguably, building off of an unstable foundation in itself is a bad coding practice :p

Refactoring code with a multitude of dependencies is pretty much guaranteed to be much more work than refactoring systems from the bottom up. It's like building a house on top of a crumbling foundation, there isn't really a good reason for it.

 

[John_M]

The letter makes valid points, but all of you are missing main goal. The game is still in EA and their primary focus is "releasing" the game. So they don't care about this at the moment. This doesn't mean that they don't care about mods - this means, as a primary objective, they don't care about the mods and/or side effects of their updates to those mods.

Once they are "done" with adding whatever they think is good, afterwards they can increase the accessibility of the modules/classes. As far as I can tell, Bannerlord is not following any known architecture or design pattern. Hence, many of the classes/modules are simply "read-and-learn" type of code. So even if they make it accessible, you have to re-read what they changed ( codewise ) in each update. Then the next thing you will complain about is the sudden changes, no deprecation annotations etc. As you can see, this can go on like this. I'm a fan of Total Conversion mods but I wouldn't recommend anyone to waste their time for this unstable mushy codebase at the moment. Let them release and stabilize the game first.

If there are any statements regarding to this - saying that they are planning to make the game as moddable as possible even though it's in the EA, then please, provide me some quotes - I'm currently unaware if there are any.

It's not as simple as them only needing to focus on EA now and putting effort into modding later, points in the letter point out choices made by TW that are adverse to modding. While there is a want for action in the short term, I think if there is a possibility to change TW stance on certain aspects of modding, that would be a good thing in it self, with the implication that change will come at some point.

 

[eddiemccandless]

Arguably, building off of an unstable foundation in itself is a bad coding practice :p

Refactoring code with a multitude of dependencies is pretty much guaranteed to be much more work than refactoring systems from the bottom up. It's like building a house on top of a crumbling foundation, there isn't really a good reason for it.

Oh I don't argue that it would be a good thing. I just don't think they have the know how to do it properly.

 

[Namakan]

I personally prefer Multiplayer over Singleplayer so I haven't really gotten into modding yet since it's entirely impossible for Multiplayer at the moment due to lack of features that TaleWorlds called the "lifeblood of M&B Multiplayer" just over two years ago.

The points made in this letter worry me (even more) greatly for the future of this game, both Single and Multiplayer.
Taleworlds, please.

There are a lot of very talented people very happy to help out in the community, there is no shame in working with them.

 

[adrakken]

+1 Been talking about this for some 6 months and its finally getting to the point its going to hurt the future of this game.

Bannerlord only exists BECAUSE of the modding community keeping the original alive long enough for TW to keep "modding" it themselves and re-releasing it like it was an expansion only for the modders to keep those alive also so they could keep repeating the process.

 

[Bloc]

by TW that are adverse to modding

Was it all public at the first stage of EA and they suddenly changed classes to internal or something? As far as I know, that never happened. Was it not hardcoded at the beginning and then converted to hardcoded all the sudden? No.
So basically, what they are doing is not "adverse" to modding but simply following what they were doing in last 3 years. And that's not modding friendly - that's for sure, but it doesn't have to be. They can't think of modders all the time while coding a new class/feature. They don't even have a proper structure internally to handle such things.
I.e. Currently entire rebellion system is handled in one CampaignBehavior - there is no helper class, there is static method to trigger it. When I communicate this to devs, their response was something similar to this "Oh yes, we are planning to add this in the future, let's see"

TW's initial plan ( for about literally everything) is "Do it first, make it work, then make it accessible and moddable".
Another example, the entire particle system was heavily hardcoded till 2016-2017 and then they slowly moved it to more accessible area ( modding tools/in game calls etc )

And:

But as I said, would it be better if TW suddenly decides to go that path by accepting all the effort this might cause to them? Yes - for the modders and for the community. Since TW won't add anything meaningful into the game anywhere soon, mods are the only hope. But will they do this? Probably not - and as I explained, they don't have to do that anyway.

 

[Namakan]

Was it all public at the first stage of EA and they suddenly changed classes to internal or something? As far as I know, that never happened. Was it not hardcoded at the beginning and then converted to hardcoded all the sudden? No.
So basically, what they are doing is not "adverse" to modding but simply following what they were doing in last 3 years. And that's not modding friendly - that's for sure, but it doesn't have to be. They can't think of modders all the time while coding a new class/feature. They don't even have a proper structure internally to handle such things.
I.e. Currently entire rebellion system is handled in one CampaignBehavior - there is no helper class, there is static method to trigger it. When I communicate this to devs, their response was something similar to this "Oh yes, we are planning to add this in the future, let's see"

TW's initial plan ( for about literally everything) is "Do it first, make it work, then make it accessible and moddable".
Another example, the entire particle system was heavily hardcoded till 2016-2017 and then they slowly moved it to more accessible area ( modding tools/in game calls etc )

And:

ah yes the game that has been advertised as modding friendly for years doesn't have to be modding friendly. what the **** am I reading.

 

[John_M]

Was it all public at the first stage of EA and they suddenly changed classes to internal or something? As far as I know, that never happened. Was it not hardcoded at the beginning and then converted to hardcoded all the sudden? No.
So basically, what they are doing is not "adverse" to modding but simply following what they were doing in last 3 years. And that's not modding friendly - that's for sure, but it doesn't have to be. They can't think of modders all the time while coding a new class/feature. They don't even have a proper structure internally to handle such things.

If the assumption is that choices are made randomly or without thought about modding, but that is not the case. As stated in the letter, choices and design philosophies were made with specific intent in how they wanted BL to be modded(not stated in public channels afaik), which is part of the problem.

 

[Lonewarrior]

Was it all public at the first stage of EA and they suddenly changed classes to internal or something? As far as I know, that never happened. Was it not hardcoded at the beginning and then converted to hardcoded all the sudden? No.

Actually yes.
The game main map campaign height used to be linked the the map itself.
Suddenly in 1.5.5 when you load a custom map you spawn below the ground and it uses the vanilla game height And cannot be changed.
A group of very talented coders who I worked with from my mod and some others mods all tried together and found it had now been hardcoded into the game almost completely removing the possibily of custom main maps

so there’s at least 1 for you of something suddenly being hardcoded if that’s what your getting at unless I’m misunderstanding what you say.