Re: I don't want the .git directory next to my code.

From: Junio C Hamano <gitster@pobox.com>
Date: Jan 16, 2008, 19:23 UTC
Message-ID: <7v8x2pr2w4.fsf@gitster.siamese.dyndns.org>
In-Reply-To: <alpine.LFD.1.00.0801161000310.2806@woody.linux-foundation.org>

Linus Torvalds <torvalds@linux-foundation.org> writes:

Show 16 quoted lines

> Let me do a few examples of why this is a good idea:
>
>  - the whole point of development trees and SCM's (and that's *especially* 
>    true with git) is how you can try things out, go backwards in time, and 
>    generally just do *development*.
>
>    If you do that in what is your public deployment area, you're already 
>    very limited. Not only may you not want to make that .git directory 
>    accessible to others (while you *do* obviously want to make the 
>    deployment itself), you also end up exposing things like your 
>    management scripts and source code along with "generated files" etc 
>    that are the things you actually want to deploy.
>
>    Yes, it's certainly quite possible that you simply don't have any 
>    management scripts etc, and that you don't generate any files, and you 
>    simply want to just deploy the exact files that you also want to track.

Even without any management script, you cannot do any _development_ in such a tree. By definition, mucking with the files in the deployment area means all of your changes are immediately visible by the clients.

One reason (admittedly misguided) people mentioned why they want to do this is because they want to be able to "git-add" files generated on the deployed server by client actions (think of a Wiki that drops new contents in an area writable by the webserver). If your deployment area is _not_ managed by git, they instead need to write a Makefile target in their development area that takes new/modified files back to the development tree and "git-add" those copies.

The right way to manage these client-generated contents would of course be to commit them to a branch separate from the sources you develop (otherwise your history will be a mixed mess between the true development and client content changes), so the argument is very weak and is not a good justification for wanting to have deployment and development tree as one.

But I can well imagine that it is a tempting way to work for people who have not thought about the reason why history matters, especially for the ones who still suffer from CVS braindamage that makes separate branches inpractical.