The Picture Project Surfaces

There were only about two or three things I wanted to accomplish today. I so happened to get none of them done, but such is life.

Amongst other things, I been thinking about my Everlasting Picture Project lately. I am slowly trying to widdle it down. This pet project of mine tends to have boundless scope. But, talking with Brad last week, we came to a few conclusions which should help make this project happen.

1. Not only is a multilevel tagging hierarchy difficult to implement, it would be difficult to use. As such, the requirement is being dropped. This means that if you're looking for every picture taken with a DSLR, tag it with DSLR and not just D40, or whatever. Ok? Ok.
2. It will be a desktop application. True, the desktop app is dying, but for us old farts, a little Windows shell integration (right-click -> Image Library -> Add Tag type thing) would go a long way.
3. Weighted tags are feasible, so long as we can work out a decent UI for it. Also, numerical weight values should probably be hidden. The user doesn't care what the number is, just how heavy/light something is in comparison to the others.
4. C++ (in all its glory) is probably not the platform to write this on.

I'll try hashing out the requirements again.