It's 1 person doing the development of the game.
And that's exactly the problem. There's a potentially sizable developer community on Warlight- just look at the activity of some people when it comes to CLOTs, the participants in the Warlight AI challenges, and the people doing analysis.
This wouldn't be easy (would probably take a week or so, maybe 2-3 or a whole month if done extensibly and thoroughly) but what I think Fizzer could do to completely solve the limitations that inhibit Warlight's development would be to open up its APIs (and, of course, create or publish usage caps; right now, based on the lack of limits on calling, it's very evident that Fizzer doesn't expect those APIs to actually be used and didn't design them with that expectation in mind). Here's a simple roadmap:
Create Game API: fix the parser (looks like there's a formatting error that throws an exception for perfectly-formatted JSON resulting from widely-used and nearly bug-free Python libraries; I figured it out a month ago or so- before I had API access, so I forgot about it but I'll edit it in when I find the exact bug again), remove the absolute restriction that keeps you from inviting players (replace with a cap of some sort) as it renders large-scale automation near-impossible especially when things like
Order Priority/Delay require L43 and aren't accessible to most players.
Query Game API: remove restriction on querying games that you weren't a part of (outside tournament/ladder games) as it inhibits analysis; at the very least, make it possible for you to analyze games you hosted but didn't participate in... or let the host decline their own game (the easiest janky workaround here)
Game ID feed API: allow for reverse sorting (really cosmetic but still pretty solid), restrictions like "didn't end in a boot" or "not expired", etc.
Clan interface: make the APIs public; Fizzer says that fixing the clan management system isn't in his "immediate roadmap." That's understandable- it's just like reddit, for example, not fixing their moderation tools for
years. However, the difference is that reddit opened up its moderation tools API and along came Toolbox- something the mod community has been fine with using for years; it's kept us from complaining too much about the shitty vanilla modtools, at least until the Paocalypse happened. Right now, I'm actually working on fixing this after reading Fizzer's response but frankly I suck at reverse engineering private APIs when anything more complex than simple HTTP requests is involved; if he made this public, give it a month and you have a much more robust clan management interface available as a browser extension or script.
Tournament interface: allow for tournament creation and inviting to be automated; this would allow, for example, the widely-operating leagues (Clan League, Promotion/Relegation League/etc.) to be automated without using the CLOT framework. Also allow for tournament querying beyond the Game ID feed API- for example, getting win/loss directly, getting final ranks, getting invites/declines, getting clans/etc.
These are not entirely without their pitfalls and, of course, would have to be done carefully. But just these four would do a whole lot. I understand that it's an investment to overhaul the APIs, but it's got dividends that easily eclipse that of virtually any other Warlight change that's just as simple or simpler to implement.
The things you call trivial are probably not as easy as you think.
They would still be trivial if the codebase is written in Java (which I suspect it is based on the errors that occur on some API calls being typical ones from certain Java libraries) and sensibly organized (which I suspect it is, because Fizzer clearly knows how to code based on his CLOT repository on GitHub).
Are you looking at the coding for Warlight right now in order to know fully well how trivial or not it is.
No, because Fizzer hasn't open-sourced the codebase or made it remotely easy for a developer to work on. Most of the public APIs are much, much more restrictive than WL's private APIs, sometimes to the point of being near-useless outside some very specific situations (I'm talking about the Create Game API outside CLOTs, for example). If Fizzer open-sourced Warlight, not only would it allow for quicker bug-detection but it would also basically lead to a massive amount of development on his platform and almost certainly make it last longer.
To be honest, if I had access to Warlight's codebase, I'd easily do the changes I'm suggesting without charge even though I've got a ton of other obligations to worry about, just because they're almost certainly not that hard. I'm betting that they're easier than what a decent 17-year-old programmer would implement in 24 hours at a hackathon.
What's being seriously neglected? He's made consistently good updates to the game since I first arrived here. They just may not be the updates you WANT.
I apologize if I'm coming off as hostile, but I genuinely think those are the three possibilities. I understand that Fizzer clearly updates this game regularly, but I simply don't understand why he adds things like the Bomb Card over the vast majority of sensible, simple user suggestions.
So what I'm saying is that
there are people that would easily implement user suggestions for free if Fizzer just made Warlight developer-friendly. We wouldn't have to go through this uservoice "immediate roadmap" situation if people other than Fizzer just had the ability to actually do something to make Warlight better without expending the effort of reverse-engineering Warlight's APIs and then worrying about Fizzer finding out/caring that they did so.
Edited 7/10/2015 16:07:26
