These days I have been designing my J.D structure and now I’m at a point where I want to really give it a try. I have an AC.ID system (I decided against projects, because I have many stuff that isn’t easily categorized into a single project and I don’t want to create a mess of symlinks). I’m not sure my scheme is optimal and I may be making some wrong decisions, but that doesn’t worry me a lot, because I’m pretty sure even if it’s suboptimal, it will still be much better than the writhing mess of primal chaos that I currently have.
So I was going to start moving folders around, when I thought the following: do I really need the area folders “10-19 Personal”, “20-29 Research”, “30-39 Teaching”, etc. to be filesystem folders? Why not a flat folder structure, directly with
21 Research ideas
22 Papers
23 Presentations
…
28 Grant requests
31 Course materials
32 Course organization
…
After all, the hierarchy is already implicit in the numbering (i.e., I know that 2 is an area, in this case research, and 3 another, in this case teaching - the numbers themselves are giving me the information that those categories belong to different areas, even without explicit nesting).
The cons I can see are:
- The names of the areas aren’t directly seen when browsing the filesystem (but this doesn’t look too bad because there are only 6, so they’ll probably become ingrained very soon. And even if I see it as a problem, it could be averted by renaming the categories, e.g. 21 Research.Ideas, 22 Research.Papers, etc.)
- More scrolling. But I don’t think I’ll have much more than 30 categories, and the ordering and numbering provide a good idea of how much to scroll, so it’s not the end of the world.
The pros would be:
- Less clicking.
- For a few categories that can be a bit ambiguous as to what area they belong to, it can be easier/faster to make a visual search, rather than having to go into the different areas.
Any thoughts about this? I’m quite convinced of trying it but I wonder if there’s an obvious pitfall I’m missing, or if someone tried this and failed.