<looks around at thread, double-checks that I'm indeed logged into the AI War forum and not the DOTA boards>
<struggles to tape shut the overflowing box labelled "overhaul Ancient Shadows/Nebulae", adds the label "Keith's Problem" to it and loads it onto an Astro Train, hoping there is sufficient postage to get it far, far away from me>
Anyway... Let's see what else has happened here since I chilled out for a day or so.
Hmm... a lot of these minor faction tweaks seem to be getting at some pretty fundamental elements of how the game and the expansions were designed. Might be good to try and get a Q&A with Chris at some point so that folks can get into the design mindset he had when coming up with these ideas. Then we could highlight if there's a disconnect between his intent and the player base expectation and, by extension, better clarify how we can make productive suggestions to improve the game overall.
That said, I've lost track of where we're focusing on this topic. Red.Queen: what scale of "overhaul" are you looking for feedback on at this point? Minor mod stuff, spitballing game balance tweaks, brainstorming on what to ask for the next expansion...? All three are conversations the community should have, but it would help me percolate my own ideas to clarify what the discussion at hand is supposed to be about (even if we end up ignoring it 
)
Good questions, Chris. I didn't lay those out at the beginning as I was curious what the unfiltered response from the community would end up looking like. (Science!) But I think it's now the right time to answer them.
My main interest currently is in the smaller end of the spectrum. (Not that the epic and ambitious stuff isn't fun to read, it's just better suited on a practical level for expansion brainstorming aimed Arcen, at least currently.) Identifying those cases where small to moderate changes would either patch up an irritation/balance problem holding back an otherwise nice plot/MF, or plots/MF that are currently in the category of "no more than 5 people in the history of the game have ever liked this as it is" but that could be salvaged efficiently (fixing some severe balance issue, adding/changing/removing abilities, especially if I can reuse existing code in whole or in part, etc).
Basically, the high "bang for the buck" stuff -- off the cuff, I'd consider things like the tweaks to Resistance/Marauder/Neinzul spawning examples of the first type, and the Beachhead ability reworking one of the latter. There are a couple of reasons for this.
#1, The codebase for this game is massive and it's going to take time for me to learn it well enough that something can come up, and I will be able to immediately recall the entire chain of logic involved, thus making heavier work more feasible without having to spend hours tracking down and researching every point every time before I can even start sketching out possible solutions. Spot-upgrading/fixing stuff in the MF/Plot category sounded like a fun way to start and focus on a manageable subsection of the codebase.
There's a certain amount of overlap, so if I am working on one plot, odds are I am passing by references to another, and absorbing info on how more parts of the game are structured and fit together that will make knowing where to go/what's available later on something different faster. And then the cycle repeats again... and again... and again, but faster each time. Basically, I teach myself a given codebase in an iterative feedback loop. It's also how I pick new programming techniques, expanding the possibilities of what I could potentially do over time.
#2, I'm trying out a bunch of new tools and a new workflow on AI War. On my last project, I worked purely by hand in MSIL (a type of Assembly language). This time, I will be adding reflection techniques to my arsenal, allowing me to do at least some work in C# in cases where Assembly work would be too burdensome, opening up a lot of possibilities. But the tools are pretty experimental and I need to learn their limits, what does/doesn't work, and how far I can push them without having to start finding weird workarounds. As there's zero documentation, it's a lot of testing and studying by doing. The faster and surer way of learning what I need fully is to start smaller and then scale up. That way, I can accurately tell the difference between "I screwed up" and "there's a bug in this tool, better come up with a workaround".
#3, The small to mid stuff has a much better chance of actually being completable with my current knowledge of the codebase and available time to devote to the game. More fun for me, and more fun for everyone else as it means you guys essentially have beta updates for the game again to explore while Chris and Keith are away busy. On my last project, depending on the scope of what I was changing/adding at a given moment, I was kicking out updates every 1-14 days, working by hand in Assembly. I'm not stupid enough to make promises on what I can/will do here as that starts getting into the risk of my mouth writing checks my ass can't cash, but I'm just tossing it out there to let you know that sometimes, when I am on a roll and having fun, I can go nuts. >D That project had a nearly dead community too, unlike here, so I didn't have much in the way of extra bonus fun and motivation from watching people play with what I built or brainstorm with them.
#4, There's a certain amount of racing the patch clock, since this is a game that's still actively supported by the developers. Updating mods when you're reverse engineering is a royal pain, doing it in the middle of developing a new bit is doubly so. Updating in this context is very, very manual, and I would much prefer to fire off a set of tweaks/upgrades/new stuff and everyone have time to chew on it and settle on what version wins the Darwinian struggle for survival without having to repatch the game a bunch of times before a consensus is reached and the uninteresting branches are dropped.
Especially as the winner could potentially end up getting rolled into an upcoming official patch if it's liked enough and the devs agree with it (Arcen-Chris already mentioned this as a possibility over in the "Not So Assault Transports" thread in the Modding forum). So kind of think of this all as a
"Mantis PLUS" system where we can identify potential things to do with the game, and then go beyond simply offering the suggestion by already have a functional possible solution that's had some preliminary testing to confirm "we like it and here's why!". Especially with how the gang is tied up crunching on two other games at the moment, anything that will make their lives easier should be a good thing.
TL;DR -- Nope, can't really shorten this one, gotta read the whole post.
