[ColinWright]

There are said to be two types of people - pilers and filers. I've always been a filer, but have found that as my interests spread, and my work has increasingly been to create connections, I don't know where to file things.

So for my paperwork I have a single "pile". Each document gets a unique reference number, it gets filed in alphanumeric order, and I keep a file with the number and every keyword I can think of that might be relevant. When I want something I do a keyword search on the file and that pops out all the related documents references, along with the other keywords so I can decide if it's actually relevant.

I'm doing a similar thing with texts, snippets, photos, PDFs, etc. I put everything in a wiki, migrate things freely, but retain with everything a reference number, and in an index file, a list of reference numbers with their keywords. To start with it felt like an effort, but it has paid off time and time again. Things tend to migrate to meet up with their friends, but the index means things are easy to find.

Search is fast, but having things migrate and cluster creates the opportunity for coincidences and serendipity to play a role.

Embrace, adapt, enhance, and do what works for you.

And have backups. Multiple backups. And test your restore process regularly.

[zuttton]

This process seems well thought out and worth adapting. Few things:

So, what does your file structure in your home look like?

Are you now basically dumping everything into this wiki and you reference files by a number system?

[ColinWright]

> what does your file structure in your home look like?

If you mean physical files, such as letters, books, and paper documentation, I have a filing cabinet and things are simply stored by giving them a reference number, and then filed in that order. Then in a reference index in the wiki I put a single line with the reference number followed by all the tags I might want to use to find that document. I use everything I can think of here, because it costs effectively nothing to have as many tags as I choose.

If you mean files on a computer then I don't know what you might mean by "in your home." I store snippets, etc., in pages in a wiki, and each item is either obviously already searchable, or has some tags associated with it as a comment of some description. Large files such as PDFs, raw radar data files, downloads of book drafts, etc., are given a sensible name and assigned a reference number. Where they are stored in the computer's filing system depends on a whim, but usually they go in a structure xx/zz/yy/filename where the xx:yy:zz are the initial six characters of the reference number. Then again, it's indexed, one line containing the reference number, filename, and tags. It doesn't matter if it moves because I can always find it by filename with a "locate" search (if I'm running locate) or just by a find command.

> Are you now basically dumping everything into this wiki and you reference files by a number system?

Pretty much, yes. I find that having a routine stops me from having to think about how and where I put things. I just follow the routine blindly, and it's flexible eough that I can find things when I want them. More, the fact that things can be moved about, taking their reference with them, means things end up clustering without very much extra effort.