Developments of the Last Week

6 more and wow, hopefully in the future ill be able to mod it to show the kill feed with out it crashing on me, any help there?

 

What does this mean ? Classic and Full will be separated into two clients ?

 

I sort of wished for lua but i guess Angel Script is okay too...

 

Modding huh.
Now I'm going to get VERY serious here.

There are some important things I want to point out for you to consider regarding a modding system. If you don't take this seriously, a bad modding system is worse than no modding system, because its a waste of time if you do it wrong.
If you want people to see people make worthwile mods that are of good quality and serious in their design (massive addons, total conversions, overhauls), you must ensure several things.


1. Fully Consistent Vanilla standards for rules and content.
If you intend KAG to become a serious platform for modding, the foundation has to encourage consistency. This means that the KAG universe has solid rules of how physics work, how character behave, how the world is structured and why so. You will need to establish standards for this stuff. Concrete rules of the KAG universe, in other words.

This also extends beyond how content behaves ingame. Consistency is equally important in terms of how they are defined in the files when they are created. As in, HOW do you create the content. Things like formats, conventions, standards and other rules to adhere when making content for KAG. And I mean actual good content, not your usual yet-another-generic mod that you will most likely skip or uninstall after 5 minutes of trying it.


2. No content hardcoded.
If you want to see Total Conversions or Overhauls, its is critical, I mean CRITICAL to not have any piece of vanilla content hardcoded, or it will interfere with any attempts at making said types of mods.


3. Documentation AND well-made, guiding Tutorials for content creation.
Incomplete or no information on base classes, functions and attributes on the coding side can deter many MANY people from ever making a mod that could have been worthwhile. This goes double for tutorials. Brute-forcing your way through trial and error not only takes longer than its supposed to, but it is also frustrating, maddening and blind when making a mod. Or heck, when making pretty much anything creative, not only modding.

Tutorials can not only be used to get people started, but also guided to make the stuff you would want to see people make. Not only you have to teach an aspiring modder on how the tech works, but you must also teach him how to think for himself for what are the possibilities and potential. This stuff is not self-evident and must be taught. Theres a very good reason why many first mods in Cortex Command were bullethoses or bland reskins of existing things.


4. Seperate support for Total Conversions mods and Addon Mods.
Mods are different, but most can be categorized:
1. Edits, Tweaks, Overhauls
2. Addons, new mechanics/features
3. Total conversions (aka new base game)

Thing is, addons often are compatible with other addons in terms of design because they simply ADD something. And many times, there are multiple addon mods that would be great to have functioning at the same time.

Total conversion, however, redefine the entire system, the whole game is different. As such, you can mix two total conversions.

To have a robust modding support, that not only is limited to just one mods or relying on overwriting the base game, there needs to be support for launching a seperate version of the game, as well as support for additional addon mods to further extend a particular version of the game (either vanilla or TC modded).

Point is, from a users perspective, when he launches the game and there are total conversion mods installed, he gets a game launcher program that should allow him to choose in which mode will he launch the game. From there on, KAG will load whatever version of the game, INCLUDING the additional addon mods that were made for that particular version of the game.


Those are the core pillars of a robust modding system and if you fuck up any of that, I will not mod it. You can fucking count me out of it. I'm still bitter from the Cortex Command modding experience and all I just said is originated from the years of experience modding CC and all the things that sucked about it. Things that made me stop. And things, without a doubt have deterred or stalled many aspiring modders as well as experienced modders from getting into the fun of modding or continuing it.

Besides all that, there is also the issue how KAG is developed. You guys are, from what I've seen, build KAG with agile development in mind. You rapidly test and prototype stuff and while you are indeed fast, the code base is a mess and the results are glitchy gameplay, inconsistency in game content and mechanics, bugs, security holes and loads of many more fun things.

My point is, you guys are still at it and I doubt you are considering doing things differently (well designed before implemented, for example), so this will not have a good impact on the modding system you are planning. What I'm saying is that perhaps you should not consider having a proper modding system in KAG and instead, focus on the core game itself.

Building a robust modding framework is a hard challenge and you absolutely dont do it by rushing, so you may want to reconsider your plans for this. One thing I can suggest is evolving KAG like you do now to its final state in terms of vanilla full version gameplay, THEN only consider building the whole thing from ground up with a modding system integrated properly. Building it as you go will cause frustration not only to you but also to modders when there is no stability for the work they do. Think about how fun it was when each new version of CC broke the previous mods.

That is all.

 

I'll hold you to that, but I'm more concerned about MM. Although both of you have worked on this project equally, MM was the one who concieved KAG and the one doing the main code work. Now, I might be wrong since I don't fully know what kind of relationship you two are having with this project (data-access, work-split, profit-split etc.), but I always had the feel MM is the ultimate big boss of KAG development.

I've followed MM's incredible adventures with soldat, with his blog (MM what I do, Defining Awesome), link-dead and now with KAG. I've seen how he thinks and what he values and the way he works. His methods often leaned on simply getting the job done and done fast, regardless of code quality or indepth fore-thought for design.

While you, Geti, do indeed have the experience of modding from CC and have seen what was wrong with it, MM doesn't. If he would, I'd imagine his development style would much be different than what he is doing.
And since he's the main guy doing most of the engine work, he'll have the responsibility to implement the modding system.
For that reason, I am concerned as to whether MM will do it right, is HE aware of what I wrote in my previous post and does he care for it as much as you and I do.

I'm not blaming MM or how he works, because without that crazy workstyle of MM, KAG wouldn't be what it is now, especially in such a short time. While he did sacrifice code quality for development speed and has had to deal with so many bugs that resulted from it, it still miraculously works. We, the players, actually do have something to play with. Thats absolutely fantastic.

But when it comes to a modding system, that development style of his needs to slow down. And I'm not sure MM is up for that, because that requires him to change, maybe not in the way he might find comfortable as it will make him slower than he is today, just for the purpose of writing a more stable framework. He might rather deal with the cumbersome nature of the current engine along with all its bugs, simply because it works right now.

A lot what I said is speculation based on analysis from my time following MM over the internet, so I reserve the right to be utterly wrong.

But if I am right, then I know this about MM: He is a man that lives today, not tommorow.

And to achieve something of a robust, well designed modding framework, it might turn out to be a very tough challenge for him.

 

I already tried to mod but failed (I'm pretty good at sucking at it though) so a modding API would be just fantastic! :)

 

the most i've able to mod is swordfight (give bombs, make builder playable, unlimited items) i was slightly able to mod it so i can see the kill feed and it worked for a brief period. now it doesn't really work anymore. its really pretty easy to mod. any1 know how to make the server stop trying to connect to the master server offline?