My 2 cents, I'm trying to keep the whole thread on my head so bear with me.
Graphically there's nothing wrong with Alleg. It's not shiny, but it is also the only game truely like it, and gameplay and user experience trump graphics in many cases. Providing the high res textures, UI overrides and keymaps that everyone considers "standard" these days, and updating the installer so that anyone downloading now doesn't have to go through several rounds of updates is a major step in that direction.
To that end, it's not PKK's job to release the new installer, though he might be able to update the package the fastest. The installer stuff iirc is in the repo, so anyone who wants to check it out and run with it should be able to. If PKK was somehow preventing an updated installer from being released through the download links, that's another issue.
Metz, many thanks for trying to spearhead an effort to address the above, but calling for PKK's head over it is unnecessary theatrics. When PKK was having site issues, I stepped up, got admin access to the server, and did what I could. There's more I could probably do, and might be in a position to assist you with deploying stuff. Did you ever request the authority to make the changes you wanted before you started badmouthing the admins? He seems pretty responsive for someone who's doing this as a labour of love.
Minor point, Free Allegiance isn't Alleg 2. Maybe 1.5, but it will address the issues of Windows requirements when it's done. I don't think being tied to MS's code is an issue, no one is going to come gunning for the game. Just having a client/server environment that runs on stuff other than windows, and that can easily be upgraded and updated gives us a great deal of wiggle room. I'm trying to address those issues, and more help is always welcome, but just having a linux server and/or client with the current art assets and a development pipeline that isn't as hostile as the current alleg source will go a long way.
The alleg source situation is not anyone's fault. The code is old, and the stuff it depends on is some combination of deprecated, difficult to use, and not usable with MS's free tool chain. The actual game code though, still solid. Salvaging what I can and rebuilding around established FLOSS technology has been paying off. I might even be able port FedSrv to linux once I'm ready to replace the networking layer.
Continuing to foster community development through the Friday night fights and Commander Wars is, I believe, the right strategy. We're kind of past squads, we just need quality PUGs and events with a community that is quick to welcome new players and quick to show them the ropes. Beyond that, it's just a matter of minimizing the steps required to go from download to playing online.
I'm going to keep working on my stuff, and hope it bears fruit before the community collapses. I'll probably keep working on it even if it does, because there's nothing to stop me, but the game and the community are sort of the same thing. It's just a shame that many of the people that kept me here have moved on for one reason or another. In the mean time, the current R7 release just needs some serious "Dog food testing". Look at what the experience is like for a first time user, try to identify where someone would be tempted to give up, and remove those issues if possible.
Growth is going to be hard for a game that requires a healthy population to get decent games going, for those who get in on a night that no one else is around, it's going to turn them away. I know the ad campaigns are trying to draw people in for the weekends, but to that end the time it takes to go from seeing the ad to having fun needs to be as short as possible. If someone decides to make a friday night of Alleg for the first time, they need to be online before they finish their first beer.
Since Turkey seems to have abandoned efforts on the client, I'll see if I can at least get a functional R7 debug client out of the current source to help pin down crash causes, but it may not be possible with Express Edition 2013, and all I've got on the professional edition level is VS 2005. If it had been easy to jump in with a build setup that "just works", I would have been working on the original source (Helping with R8) rather than my current project. But I think what I'm doing is something that would eventually need to happen to pave the way for the next chapter.
We can have a decent community around R7, just focus on what needs to be done now, not why it wasn't done earlier and who's fault that might have been.