Zvi responded to my post on tidying and interpretive labor, with a proposed tidying system that, on reflection, is solving a totally different problem from the ones I end up with. I figure it might be helpful to add some concrete details.
When tidying my room, I care about a few things.
First, clearing up workspace. The value of empty surfaces is huge - it not only allows me to much more quickly find the few items I've decided to leave out, but lowers the total burden on my attention when navigating the space. Sitting at my desk and writing is much, much lower friction if I don't have to clear a space for my laptop, can put down a cup of water somewhere, can pull out a book I need to reference and put it on the desk. Likewise, having a floor I can easily lie down on, walk across, temporarily put things down on, etc., is an important convenience.
I used to insist, when tidying, on an organization-first approach, arranging things on first principles, so that the last step was the visible reduction in clutter. I was doing this in order to avoid being the sort of person who just stuffs clutter into their closet, until their closet is overflowing, at which point they move to a place with two closets, and so on. Attempts to e.g. reorganize a bookshelf often ended after a few hours with me exhausted and a bunch of piles of books on the floor.
My life got substantially better when I started paying attention to the relative cost of having clutter in different places. It turns out that it's actually less costly to store a given amount of clutter in a box in the closet than on my desk - reallocating the clutter actually increases the amount of usable space. To avoid the problem of the overstuffed closet, I just have to ... pay attention to when I'm incurring an inconvenience from an overstuffed closet, and balance that against the cost of clearing it out. The person with the perpetually overflowing closet doesn't get there by doing honest short-term optimizations. They get there by dissociating from their environment, and trying to hide the mess from themselves. I can just ... not do that.
But, eventually, even if I sort things by kind, eventually I need ... less stuff. And that means making decisions about what to outright get rid of or put in long-term storage, especially things that might theoretically have sentimental or reference value in future, or what to do with old notes/papers. So I end up going item by item and asking questions like “If I were offered this for free would I take it?” or "Is this old greeting card worth the storage cost?", which is exhausting to do with like 50 things. Putting stuff away, anything that doesn’t fit a preexisting organizing schema, I have to decide whether to reconfigure the system, find an acceptable kludge location, or do something else.
Zvi's system is focused on the storage and retrieval of atomic objects. But in some cases, like paperwork and clothing, there are a lot of potential objects to store! Grouping things into natural kinds (e.g. ties, socks, legal ID docs, bank related stuff) makes search much, much more efficient.
Again, my life got a lot better when I realized that I could "cheat" and make ad hoc organizational decisions for some things without systematizing all the things in one go.
Sometimes my clutter is a specific message for future me, and resolving the clutter means resolving some other unrelated task. For instance, the floor of my room currently has three unpacked shoeboxes, because I knew I might want to return some of the shoes. I do in fact need to return one pair, so to clear the floor I need to first resolve the reference to an unrelated task. Another way to fix this would be to repair my task-tracking system and separately store the one needed shoebox in a less prominent place, but that’s a much bigger task to get through before tidying my room.
A friend suggested moving the shoes to the car to return later, and indeed my car also has a bunch of unresolved tasks stored as clutter.