On my TO DO list (close to the end somewhere), is an @INC searcher that finds all the installed copies of a particular module, not just the first one. From time to time I need to clean up a client's installation, but they are using different values in PERL5LIB and so on, so everyone is using different versions. At one location, Randal and I found 300 copies of cgi-lib.pl on the same box. I started something with Module::Find to find out someone beat me to the name by a couple weeks, and then I didn't do any more work on it. :\

I'd also like a module to fetch, cache, and parse the CPAN meta files without using CPAN.pm. It might look in the CPAN.pm or CPANPLUS configurations to get preferred mirror locations, but it doesn't have to.

I started Test::WWW::Accessibility, but never did anything with it. Lots of government organizations would love to have it do something. No patches necessary: you could own the module.

I had an idea for a module to set the computer clock time zine automatically when it connects to the network. This means it has to figure out which time zone its in. With Geo::IP you can easily do that ($20 for 50,000 web services accesses). I just haven't done it, yet.

Mac::PropertyList needs to be gutted and completely redone. I know you're not looking to patch things, so start from scratch and come up with something that can quickly parse a 100 MB iTunes Music Library file.

I'd really like a Perl interface to Sourceforge. Unfortunately, that means screen scraping, and that sucks. Module::Release does some of it, but there's more to do.

I could also use a Perl interface to a Javascript library so I can execute Javascript within a Perl script, especially for web testing. I think Andey Lester said he'd include that with WWW::Mech if someone wrote it (as a separate project).

If Perl knew about the various formats that Garmin uses in its devices, I could use Perl to play with them. Mmmmm, graphs!

I've recently thought about writing PDF::Graph, which would be a knock off of GD::Graph. Most of the work is already done. I just need to switch out the GD underpinnings for PDF. Beyond that, it's mostly relative coordinates. You can probably come up with an adapter that just makes it work.

Speaking of PDF, a higher level interface to PDF::API2 (maybe like PDF::Report on steroids) would be wonderful.

I also wish I had a Perl interface for Authorize.net. They have three file formats each slightly different. Sometimes I have to join all three of them. Even if the thing you come up with is to simply download and import them into something else, that woudl be nice. The interface to sign on, etc, and downlaod the right thing is a good start.