>Your example of writing a tool to do area
>generation is a perfect example. It would probably take me
>three times as long to get a tool done and in any sort of
>working state as it will to get the shifter revamp done. So
>far it's been about 6 months of my freetime. So in 18 months
> with no visible changes to CF), you can get a tool that'll
>help you write areas better/quicker. Which is more valuable?
>I'd argue the shifter revamp.

I actually spent some time thinking about this while working on my area. (Time I should have spent on the area itself, as it turns out.) What struck me as the most complicated part is that rooms needn't conform to normal 3d space, so you can't just map them onto a 3d grid. It has to be an arbitrary graph, which is a lot harder to represent visually.

One side thought: how much time do you estimate will be spent creating and/or editing areas before CF eventually shuts down? How much of that time would be saved if a graphical editor were available? If the saved time is more than the 18 months it would take you to build the thing, then in the long run it might be worthwhile.

The counter-argument, of course, is that your time (as an Implementor) is worth much more than the time spent by heroimms creating their areas. So it's not an apples to apples comparison.