in response to 3, I think there's some "picking the best out of the maps that shipped" syndrome too. Generally, most people only play retail maps online so its only logical that one or some of them become regular competitive maps. And generally, most games ship with 90% crap and 10% ok to decent maps. 

These questions extend beyond just multiplayer FPS games. As a designer currently working on an MMORPG, I am confronted by the fact that I am not a hardcore MMO guy, and have never hit the level cap in an MMO, so I don't really understand the endgame gameplay (luckily everyone around me does :)

What I've realized in designing gameplay encounters is that what makes the encounters interesting and fun is that they pose problems that requires players to develop strategies in order to solve them. These strategies are understood by the designer ahead of time, so the designer is actually leading players through a progression of more advanced strategies as past solutions fail to apply to new problems.

If you've played Antediluvian, and remember some of the shambler encounters, you might see some of this at work. I had known strategies for dealing with shamblers, from the easy to the more advanced, and designed encounters that made easy strategies fail, so that players would have to use the more advanced stragegies. For players who know them, it's fun to excercise them. For players who don't, hopefully it's a learning experience.

The point is, that I couldn't have done that without having developed these strategies as a player. 

This is pretty much pure luck. DM2, DM4 and DM6 are played in competitively in 1on1 QW and DM3 is played in 4on4 teamplay. But ID could not have possibly been fully aware of what was needed to make these maps good for competitive play because NetQuake gameplay is very different to QuakeWorld gameplay and even QW today is played rather differently from how it was played 7 years ago. So I say it was pure luck. 

you took my comment a bit out of context. But nice to make a thread. :)

First of all, I said it should be taken with a grain of salt, and secondly, I said that one way to avoid that is just to make a lot of maps. For example, nobody plays dm1 or dm5 ever... Of all the episode maps, prettymuch only e1m2 is accepted for 4on4. If you've watched a good div0 4on4 on it, you're just amazed of the possibilities and nice situations it makes possible with the delicate area interlocking and obstacle placement. You can down to to water near gl, take sng and shoot the enemy from there just when he is taking the sng! Etc. That was probably not planned, but is a product of the layout nature of the map.

And the guys at id DID play those dm maps against each other. You can read it from the wired article that tells of the making of quake. I don't know about episode maps.

I don't think the id maps are too good for duel, but dm2, dm3 and e1m2 are good in 4on4. Mostly. We really need some real players to tell here what are their good and bad sides.

Many custom maps might be playable too, it's sad that more has not been tested. Yet, when you run around in "the big three" maps, they have somehow the right scale and complexity level. It's partly just getting used to, but often bigger maps with more open areas just make rocketspam (cmt3 bridge) and small places generate flak (e3m7).

Yes, I can think an average player is very capable of making a good map for the highly skilled players, especially if he/she is observant and willing to take feedback etc etc, and sometimes you just do it by pure luck and inspiration etc.

Sometimes mappers don't understand much of the game and make big errors in item placement or architechture, say, mega, rl and ra really close to each other in a one-ra map, or ceiling too low for bunnyhopping or you get stuck into places where only skilled people can get to, too many clip brushes, etc...

"Competitive play" often ain't that special at least in qw, it's just teamplay.
You have to check that the map doesn't have one easy tactic that you can repeat forever to win, or that the map is not 100% lockable so that you can control all armors, powerups and weapons from one small area, or that the map is not too balanced where both teams just hang out at their "base" and then run for the powerups for a "ffa" game (in qw with shaft in cmt4).

You always discover ugly things when testing... "oops, this area is impossible to take over, it has health, rockets, red armor and the only rl in the game quite close to each other" - happened with enraged, so it was changed.

To sum:
You can do it by luck and by trying, but some experience increases your chances, and you can do a good map without being "pro", most probably if you take feedback. And then there's the conservativeness of the players, which is partly stupid, partly understandable. ("You don't change the football field in every game do you?") There has to be some pretty rigid standard by which people can then make the pecking order. Yeah, it's testosterone and all that stuff, it's partly fun too. :) But let's not go into that direction please, ignore this end. (the post icon is true for tonight.) 

okay i am drunk.
In e1m2:
go down near grenade launcher, take supernailgun in the water, go to under yellow armor (you can't get up from there). Shoot the enemy when he is just taking the ya (he must be a guy with a rocket launcher since the teammates chose him to have the most priced armor in the whole map), and he drops from the delicate platform stairy thingy into the water - he'll be pissed for not getting it, and perhaps even your teammates might get it, maybe even turning the map around.

For the above to happen, the map has to be reasonably complex and obstacley and interlocked and hard-to-navigate and can't be just a quake3-ey smoothy platformy airpushy thingy. But please not too much stairs or holes in the ground to slow you down all the wya.

k me out. 

but there's also a large element of luck, generally I think if you're serious about making a good mp map what you need is serious testing and to not be afraid of making big changes. I have a feeling that too many people go into the testing phase when actually they've already decided they're virtually done and will only making minor changes (mostly to items) when what the map might really need to become good is a large rebuild of some part(s). 

board ate my post that I spent 1/2 hour on

not that it was of any significance but METLSLIME maybe fix that shit already ? 

How did it get eaten exactly? If you can, please reply in the site help thread to keep this one on topic.