I've been working on building software to aid research in proteomics. With the suggestion of my professor, I created a Ruby implementation of a pepXML to mzIdentML converter. Proteowizard already created one, but through the power of Ruby and Nokogiri, Search2mzIdentML is simple, clean, and easy to expand. There were some challenges that came along while creating this, but I believe that Search2mzIdentML is now a releasable program.
I know that my blog only has about two readers, but just in case there's someone out there who searches for an mzIdentML converter that happens to come across this, I hope they'll find it useful. What's more, if that person could give some input about any problems or concerns, that would be even better.