It's a bit strange that most these projects seem to look and feel and sound almost exactly like quake 3, maybe with just some odd bits sticking out.
Maybe it's impossible to do coherent art direction in nebulous volunteer efforts. I don't believe the technology limits the style so much. Certainly that is not the case with something as basic as sounds.
The previous versions were more TRON-inspired graphically. I have not seen the project since a few years and was astonished by the changes in presentation.
What differs the most from the other Q3-likes (direct clones like OpenArena and more distant ones like Nexuiz) is the core of the game: while the others merely handle framerate-bound physics, bunny hopping, rocket- and strafe-jumping as gameplay side effects, this one is designed from the ground up to integrate such rule-abusive gameplay into its core. Jump features have been added (bounce-jump on walls), physics has been tweaked to provide less hackish strafe jump, configurations options abound to make the game tunable for enthusiasts (like coloring walls and floors instead of simply disabling textures and lightning). It seems this version goes even farther in configurability, with configurable coloring and extensively configurable/scriptable UI.
I would think that's because Quake3 is held up as one of the best multiplayer FPS games of all time. It really had a buttery smooth feeling to it that's never present in modern games like BF3 or CoD.
You may be grouping "arena shooters" all together (one of my favourite genres, btw). Examples include Quake 3 / 4, Unreal Tournament series, Serious Sam, Open Arena, Nexuiz, Xonotic to name some popular ones. It's somewhat like calling the Battlefield series a clone of Call of Duty.
When this game was founded, about 7 years ago, Quake 3 was at a pretty big peak. The original goal was to make the next big competitive FPS. Though that didn't last too long, even though the scene is still largely very competitive players, there's no big tournaments for the game. Another thing to note, is that Warsow's engine is actually based on idtech 2.
Maybe it's impossible to do coherent art direction in nebulous volunteer efforts. I don't believe the technology limits the style so much. Certainly that is not the case with something as basic as sounds.