This is an excellent rule for cleaning up (a room, a house, an office...): touch any given thing once at most.

In extreme cases, it doesn't work because you don't actually know where things are going to go, in which case a two-round approach is good: everything gets pulled out and sorted in round one, and things go to their destinations in round two.

It's interesting, I use precisely the opposite approach. I work in passes, each one intended to make the room cleaner as a whole but not necessarily finish any part of it.

For example, let's say I've had a big party and the kitchen is a mess. Step one might be to put the trash can in the center of the room and dispose of all visible debris. Step two might be to move all remaining small objects to one side of the room to improve visibility, then repeat the trash pass. Step three might then be to wipe down the counters, which are now clear of objects. Etc.

I tend to work in a similarly incremental way on programming tasks. Do you have a more "touch-once" approach there?

That's how I clean - pull everything out, assess, and put it all away. It's very efficient if you have enough space, but it drives people nuts because they see it as causing more of a mess.

An issue with this approach is it turns 'cleaning' into a time-consuming task that needs to be completed to avoid making the situation worse. That can cause some folks anxiety and require a building up of sufficient willpower to even begin the task.

Contrast that with a transaction-based approach to cleaning where an individual item is moved to it's destination place/state, then you move on to the next item. No middle state where things are worse, and you can stop at any time yet still have gained the benefit of any work completed to that point.

However, with the second approach it's key that everything has a suitable destination 'home' to be moved to. Often, that 'home' is the trash can :)

My wife attempts to use this method, but usually gets distracted at peak-messiness and things just end up worse off than they started.

No, you can't blame your wife for that method.

It was my method first.

She must have stolen it from me. ;)

I get scorned for doing it this way all the time!

Heap sort vs insertion sort.

Have also built stone walls in the Cotswolds, and I wouldn’t agree with you about not putting a stone you picked up back. Sometimes when you pick up from the pile you didn’t see part of the stone that makes it entirely unsuitable.

It’s better to pick the best stone to complement the space than try and jam something in that sorts of fits.

Have done it in the winter in a blizzard on the side of a hill - that was a painful experience, summer was much more enjoyable.

Edit to say: I realise now from your username we know each other - hope you are well! ;-)

Good to see you here, am well thank you and I hope likewise!

To your point, that's just the tradition I was told by the stonemason who built the walls I helped on.

There's a subtle art to pre-examining the stones before picking, it wasn't a case of if you touched it, it had to be picked up. It was more if you committed to this stone then make it work in this spot or another. There's always a place :-)

I see that rule as avoiding the search for perfection - otherwise you might spend hours picking up different stones and contemplating as if it were a jigsaw puzzle, when in actual fact there are 'solutions' everywhere.

"Perfect is the enemy of good."

I always saw this type of advice in the lens of: "this thing is heavy. I only want to pick it up once otherwise I will exhaust myself fairly quickly." I learned that from laying sod on a golf course one summer...

Laying a brick is like a chess move! Thank you!

This seems like it could be a (very) loose analogy for type-first programming. Before you start manipulating data (passing values to functions) you have to give careful thought to what data types you will need. If you do it right, there's far less rearranging (rummaging through the pile) needed later.

I use the same idea when stacking firewood.