i generally just lay down rooms in some path i think would be interesting, then go back and wall everything in how I see it. This way I try to avoid stumbling with the layout, and can work detail etc around the path. But, if I ever do get stuck on something like that, I'll try to just slap more rooms down and if I like it i'll keep it, otherwise I'll start over and keep trying until something sparks an idea. 

Often I find I have a vague idea of what I want to acheive in a particular area, but I can't visualise it fully. I usually slap down a load of rough 'n' ready placeholder brushwork, with the intention that it'll kickstart my creative process. After this, often I still don't have a clue as to how I want the final area to look, so I move on to something else. Eventually, I'll have the creative spark that allows me to go back and flesh out the area properly, but sometimes it takes a while. 

The best method is to actually plan the map out first. Have a firm physical (sketched on paper) or virtual (extremely basic set of brushes representing physical space) concept before you even start the map. Then block it out as bare rooms or areas, and only then can you add details.

Now, as for creative blocks when working on the details in a situation like that, there's not really much to do except throw together lots of different things trying to find something that looks good. (Note that once again you can use concept sketches to your advantage, even if you're not good at drawing.) Go look at similar maps, or even dissimilar maps that you think have good detailing. Check stuff on conceptart.org. Run a Google image search for keywords associated with your detailing and architecture style. Maybe check a few movies, or comic books. Or, hey, just read a regular book and take a short break.

If all that fails, then get one of your mapping buddies to take a look at the map and offer ideas for the room(s). The ideas don't have to be extremely detailed, just something like "Hey, that room is near the top of the map, and it just screams to have a window over-looking a mountainous area." Or "That room is in the middle of a tight and stuffy area; maybe put some fans along the short edge." Look for ideas you haven't included in the map yet.

And good luck. 

If you have defined a certain portion of the layout, but haven't invested a lot of time in detail etc. rebuilding from scratch can be a useful thing to do. It sounds like throwing the baby out with the *bathwater, but actually it makes you think about the overall design more and grasp it mentally, instead of relying on the editor 'remembering' it. By seeing clearly how you got the map to where it is, it can help you see where it wants to go.
In the meantime, you're still laying brushes and maybe laying them more neatly and efficiently than before. 

I usually give up.

No hey, try spending time away from it for awhile. Really. 

try spending time away from it for awhile

Like, a year or two? :P 

Not that extreme. But sometimes clearing your mind helps. 

Spending a few days doing something else (but related) like QuakeC or whatever your secondary speciality may be (textures/skins) often helps to rekindle the mapping spark. 

from #8,

addendum to RPG's comments

It really helps to know how to draw when you are dealing with any medium representing physical space. Often times you are blocked simply because you are having trouble visualizing the idea that is already there.

Noodling (drawing without purpose) can also help push you towards understanding the underlying ideas better.

Keep a notepad. If you have a vague idea of what you are trying to accomplish, writing it out can help make it a more thorough and content rich idea.

Study entity sets -- examine entity effects by building test levels for the sole purpose of learning how to replicate those effects. Good ideas will come if you work at it deligently. 

Good ideas will come if you work at it deligently.

I sound like a fortune cookie. See why I am only earnest when someone ask for help? 

by other mappers to rekindle the spark 

Good ideas will come if you work at it deligently.

Only worthwhile advice here. 99% of "mapper's block" is laziness. Just get working, whether by drawing a layout or building random shit, and the map will come. Ideas are a dime a dozen once you have something to start with. 

Just start working on another map, but keep the mapper's blocked one open so you can switch to it if designing a new map opens new ideas. Or you could take a couple hits of acid. Try both. 

I'll try not to let it go to my head 

in the sense of not knowing what I want to do. Usually my only mapper's block is just being burnt-out or otherwise unenthusiastic about working on anything.

Sometimes it's just as simple as taking a short break instead of continuing to bang your head on the keyboard. Even *gasp* going outside...

rebuilding from scratch can be a useful thing to do. It sounds like throwing the baby out with the *bathwater, but actually it makes you think about the overall design more and grasp it mentally, instead of relying on the editor 'remembering' it. By seeing clearly how you got the map to where it is, it can help you see where it wants to go.

This actually worked really well for me one time. I lost the map source, and *had* to rebuild, and the map ended up far cooler than it would have been in its original incarnation. 

This actually worked really well for me one time. I lost the map source, and *had* to rebuild, and the map ended up far cooler than it would have been in its original incarnation.

Also, recalling an older story I always tell when the topic of rebuilding comes up... the q1dm classic aerowalk had to be fully rebuilt 3-4 times, once from lost source files, and the rest for rebalencing/rethinking the map's gameplay. Reimagining the work after the fact can be very healthy for creating something cool. 

for when GtkRadiant decides to arbitrarily turn your .map file inside out and throttle it.
>_<