when working on coding projects, lately, I have been following the convention of checking code into a folder with an extra "wrapper" folder. example:

"git clone git://git.foo.com/projectname.git projectname/projectname"

I do this becuase then I have a folder one level up to place things I don't want under revision control (like downloaded doumentation for libs i may be using, etc). I got the idea from how every repo in gitorious is in a project folder. I like it and I don't. Alot of times I don't start out needing this extra layer (sometimes I never need it) so it may sit empty(minus the second folder of course). Anyway I was just wondering what other people do. Is there a common convention?

Another idea I have had is to just put the folder "gitignore" in my global gitignore file that way I can create it if needed and drop things in there.

I also create repositories into subdirectory, to store additional data out of it. If you don't like "empty" directories you can first clone project into main directory, and later if you see that you need additional files, you can move repository to subfolder.

I have my own directory conventions, for different purposes, but for determining what is and is not included in a commit on my own systems I use the .git/info/exclude config. That places no pre-defined constraints on a project and gives me control as needed with any required granularity.