And you would have been sooooooooo right. Since I wanted to use sh only and not even Bash, I don't even have some useful Bash features such as arrays, so I use temporary files for many many things. Plus invoking a git process for each meta information, plus an awefull lot of sed processes for the templating system… I have a few users with a growing number of blogposts who are complaining about the speed when regenerating their archives page and RSS feed. But now I don't feel like rewriting it in Perl. Plus I had a lot of fun coding in sh.
If you or anyone feels like writing a fugitive clone in Perl, I can assure you a few users as soon as its working!
I think the only thing that I might have done differently would have been to allow myself to use Perl since it is a dependency of git anyway.