Resistance and Liberation (RnL) was a WWII game that first penned itself as a "Infantry Simulation" and not even a FPS. There were a lot of assumptions made. The one thing I did correctly predict was the rampant TKing. In fact the closed beta testing went fine. But the first weeks of the Alpha were almost purely being TKd in your own spawn and seeing it later on Youtube with lots of laughing commentary. Things were okay only when you had competent admins on. Next patch added a TK limit on by default which banned players for a week on that server. That didn't stop the team wounding though. And some people would just disconnect and reconnect. In fact, there was one guy I still remember. The first thing he'd do is ask if there was an admin on. If some idiot admin responded or some player had clan tags for that server, he'd immediately disconnect. If not he'd kill four people and reconnect. Kill four more and reconnect. And people would revenge TK him and get autobanned themselves. He came back again and again and again for years. There's no telling the amount of damage him and other trolls did. Admins were not there when they needed to be and were not doing a good job. Trolls are going to be a huge problem and anyone who wants to run a server has to be ready for it.
I'm not sure what it is... but realism folks always seem to not police their servers very well. I think it's the retreat into closed servers events and then leaving their public servers mostly abandoned because the regular gameplay will pale in along side events. You'd see 32 guys in a 32 person server with clan tags on and then it'd empty out and there'd be two or four left, none of them admins, after the drill.
A lot of trying to reason with trolls also, but that might have had to do with the size of the community, WoR's will probably be bigger. My server worked out pretty well because I had a 'no warnings' gestapo style admining where you just made bad people disappear before they even knew they were being watched. 1500 bans and about 1700 people in the steam group and it was the most popular servers and everyone who played that game knows me. The trolls drive away players more than anything else.
A definite positive I see in WoR is that team kills are shown in chat. Death messages are very stupid and unneeded but ones for TKs and team wounding should be shown (to admins at least). Another one is that the game costs quite a bit right now. One problem with the Source engine was that it became free and the mod was free and people would come on with a new steamid in 2 minutes if you steamid banned them. There were even tools that would generate a new steamid and allow them to be back seemingly infinite times to mock the admins. I don't know if this engine is entirely 'free' on it's own yet but I'm glad the game is not altogether free and never will be.
I'm not sure if either of those would deter meelee rushing. The game is the game and people are going to play it. If stabbing is faster and more clean-cut, a lot of people will prefer it entirely. I used to think that was the case with spawn times but I found increasing spawn times only has the main effect of increasing the reward on tactical successes. When you kill a bunch of enemy, you get more time without them on the field. When spawn times are very low you'd see a lot more bloody stalemates as the two sides bleed each other out instead of trading ground. The defender could be getting massacred but unless there was a concerted effort to cede ground they could plug holes in their front as fast or faster than the attackers. It was a pretty consistent effect of lowering spawn times. With longer spawn times eventually the map would progress (to a different objective) but with short spawn times the two sides would bleed each other out at the first objective. The only other things I noticed with longer spawn times was more camping, less role-playing, lots of complaints from players about spawn times, and the best players were extra effective as they spent a lot more time alive and killing than rookies who spent a lot of their time dead and moving up to the line. I don't know if there is an ideal spawn time but a dynamic one that starts at some base will have an effect on how things go for the teams. It's the easiest way to alter the flow of a map is the number of players each side gets and how close to where they are needed.
As it is on some of the maps (like the cornfield) it's a serious delay to get back into the action even after spawning. Spawning on a literal flag somewhere close to the action would eliminate that hardship and might encourage more impulse even if the spawn time is much longer than it is now. I know I'd love to be able to stroll to the fridge after I die rather than taking a virtual stroll. I'm imagining facing a 20 man company in line with a flag bearer and all the sudden their flag bearer poops out 12 more guys. Or better... you're chasing a flag bearer and he poops out a bunch of friends. You'd have to take into consideration that you're making the flag bearer a critical asset (and therefore also subject to severe troling) on top of odd placement of the colors and making him (or it laying on the ground?) a prime target. Progressing zones of control with spawn points might be more sensible than making a human player the focal point of the team's ability to spawn. I've already had a Union flag bearer join our CSA line on Burnside's Bridge.
Anything that can be abused will be, especially if it's easy. I'm assuming you would have a proximity trigger to stop the spawning when the enemy are close. That would only make sense. So the common sense tactician in me says get close! Stop his spawning. That probably means meelee. Hold my own flag bearer back so I can spawn players and charge his to block his spawn. Just like that he runs out of players in the area and I keep getting them. If you have multiple set areas where a team spawns (like in a cornfield) that can also be blocked by proximity of enemy, expect there to be individual team players hiding and blocking those spawns.
As for reinforcement counts, RnL had a server-side option that let you hide them from the opposing team. That did not stop questions of "How many tickets do you guys have left?" And some people answered. Though you're still going to get "How's your team morale doing?" You're also going to have to figure that both teams could bottom out around the same time. So will there be draws or will the decision be immediate? Will teams who run out of morale still be able to defend until the timer goes out even though they are in the process of being killed off or will McClellan give them another few minutes before calling off the whole operation?