Hey there, sorry it’s been a while.
I see where you’re coming from with your inverted-projects approach, and there may be some merit to it: but the concept of a ‘project’ in my mind is of a thing that is a self-contained bunch of work. I could equally well – and have thought about – using the word ‘system’ across the site instead of ‘project’.
There is a recognition, of course, that many projects are similar. This is why you should try to re-use structure where possible. I’ve talked here and elsewhere about this sort of approach.
So in my mind, if you always deal with ‘finances’ as part of a project, then I would standardise on a ‘finances’ category and always have PRO.13.ID as that category (or whatever).
The GTD terminology of an ‘area’ does confuse matters here. My ‘project’ might be more like their ‘area’, if you have multiple projects.
The way I feel about (JD) projects is that they should be self-contained. Let’s pretend you’re at work and someone new starts, but they’re only working on one project. No problem: give them access to the project JD index and file structure and mailbox and away you go.
Whereas if you’ve got different project artefacts scattered throughout, deep in the structure, you just can’t do that.
Similarly at home. If you start a new project and it’s big enough that it warrants its own PRO – and be careful, fit it in an area if you can! – but if it does, then that’s an isolated thing. If you drop it later you can just drop that whole project structure and the associated file system. Again, you can’t do that if you’ve scattered it throughout your system.
That said, I’m always interested in how people do things differently. This is a work-in-progress. So if something works for you, please do it and let us know how you go. I don’t want this thing to be prescriptive.