[jacquesm]

That's very interesting. I foresee a problem with capacity. If you keep everything around eventually you won't be able to move anymore. I'm so happy that I'm beginning to forget some of the stuff from my childhood, to keep carrying that along forever seems like a huge weight. Forgetting is just like remembering: a blessing and a curse at the same time. It would be really nice if our minds could actively forget something (like a file delete on a file system), right now it seems to be mostly governed by some caching rule where the last stage of the cache is /dev/null.

By committing it all to paper (and presumably, by re-reading it) you fire up all those neurons periodically which will lead to you thwarting the garbage collection process. As a consequence eventually you'll either run out of room or possibly end up with mental issues (inability to acquire new stuff or inability to focus due the large number of associations running out from each thought you have).

There is a cost associated with this map making, still, I'm very much tempted to try it to see what the positive effects are.

[arh68]

>inability to acquire new stuff or inability to focus due the large number of associations running out from each thought you have

It's exactly this. In the active mind (hyperactive, perhaps), the dataset is ever increasing and the memory banks are thrashing. On the best days, the cache is very deep. You never really know how deep the cache goes, though, so what might be the in L3 one day silently becomes /dev/null. This is a problem. (It's also, as you point out, a blessing)

A permanent store is the solution, but the brute force (write everything down) doesn't immediately solve the issue. There is no metadata in writing, so searching through papers involves full-text O(ludicrous) kind of searches. Every permanent store needs an index to be useful. It looks like the author has developed an extensive, coherent(?) way to colocate and index written information.

Considering how slowly most people can write, I'd be very curious to see how good mapping software could be used with ~300wpm stenographer-paced typing (Plover). Typing as fast as I can think into a searchable, sortable, arbitrarily small/large/nested/rearrangable/duplicable/etc medium is intriguing, to say the least.

EDIT: One more thing. The 'immobilization' that some ridicule is inherent in the medium: ink on paper. A written page is immutable and cannot show a diff. This is a bit like Purely Functional Data Structures. The 'structure' is at best just a single page, rewritten fairly quickly. As the ideas grow, sheet by sheet, editing a single page requires rebuilding the now-12-page idea. This is refactoring with pen & paper.

The only way to be 'mobile' is to either not write it down in the first place or to archive and never look at again. You're right in that rehashing old ideas thwarts gc: it's like burning an image into your screen, right into the phosphorus.

[jayrobin]

Interesting perspective. I often get annoyed at myself for remember very little from my school years that's not related to my current career or hobbies (e.g. history, geography, English lit) and almost feel that I should learn it all again, but I guess if I don't actually use that knowledge I'd just be forcing myself to memorise facts for the sake of it.

Perhaps it is best that I just let it go and instead focus on learning new things that I really want to learn or that'd be useful for what I do/want to do?

Now if only I could selectively delete memories I could finally forget the time I ended a call to a client with "Love you, bye!".

[cgore]

I have been keeping journals for years, and have most of it all the way back to around the age of 10 (I'm 34 now.) Generally what happens is that my really old stuff is just sitting in big plastic storage boxes, and I don't ever look at it except every few years when I feel nostalgic, so that doesn't become a problem.