Dear Peter,
Can I suggest in the strongest terms that the Jaina prosopography project uses the
Pandit prosopography system for managing its data? Many of us are interested in the same kinds of question, and it is critically important that we share data. If we all go off in corners and re-invent the wheel, we will set back our field by decades.
Pandit isn't perfect. It may not even meet all your needs, although I suspect you will be surprised by how much it does do for a project such as yours. However, Yigal and the team who created and maintain Pandit are actively developing it and they are very open to technical suggestions (and money!). I am in such a dialogue myself right now, with one of their system programmers who lives in Brazil.
It is easy to underestimate the difficulty of creating a sound tool for serious prosopographical work. Pandit is itself the third iteration of the project. It started as a Windows-based Filemaker Pro project. That became unusable after a certain volume of data was added. Then it was ported to MySQL running on a Linux base. That was more robust, but also had drawbacks due to leftover features from the earlier Windows mess. Finally, Yigal had the whole thing rewritten again in the light of everything we had learned and ported the very considerable volume of data forward to the new system. Since then, even more author/work/manuscript data and bibliographical material has been added.
Looking at your Jaina prosopography paper, all the "methodological issues" you raise in sections 3 and the onomastic issues in section 4 were faced and mostly solved by the
Philobiblon project, many years ago, in the context of Iberian prosopography. A major part of that thinking informs Pandit.
The basic fact is that systems such as this become exponentially more useful as more data is added. What you add links to what has been added before. New relationships are discovered, time is saved by not entering the same data repeatedly and this, in turn, leads to a major gain in accuracy. (A public genealogical system that demonstrates this kind of crowdsourced cooperation-gain superbly is
geni.com.)
If you decide not to put your project's data in Pandit, then I hope you will produce a compelling document explaining why not. It will be very useful to the Pandit team for their own future consideration.
Best,
Dominik