Structure of Information (SoIII) Part I

CategoriesMethods & Tools, Personal Development
Let’s talk about how I structure files and data

Disclaimer: I’m a technologist so some tools I use may be a bit uncomfortable.

Continuous Backup

The idea is that losing any hardware should not cause data loss. I use Dropbox and it works great. My main folder on my laptop where I keep everything that isn’t otherwise archived, is synced to Dropbox, so if I lose my laptop (as has happened, for example, reinstalling the compiz library destroyed my OS) – I actually lose no data at all.

I’m fortunate in that I don’t really deal with large files. A media professional, for exampe, may create gigabytes or terabytes of data over a weekend, or a workweek. They have to have specific tools to address the data persistency problem. Having two copies of critical data, and continuously buying storage (terabytes of solid-state drives, 1Tb/week purchase, or monthly purchase of 1Tb of hardware space at the cost of $100/mo, etc). So, I guess one of the main things they do is continue buying storage hardware, although again I’m not an expert on this aspect because most of my data is text and I don’t deal with terabytes of generated data per week/month.

My code projects are quite simply in git, and on either bitbucket or github. It’s super difficult to lose git projects.

I also have several s3 (amazon storage) accounts for “large” (gigabytes) data that I don’t want to lose. It takes a while to upload things to s3, and it may be difficult while you are abroad, but again, you don’t lose data when you lose hardware.

You have to specifically worry about access keys. Don’t put everything in one bucket and then either lose the credentials or have them stolen. Ideally, distribute the data across multiple buckers and use IAM to manage permissions, and disable the root account – although admittedly this is more complicated than just keeping data in one place. It depends on how many people are interested in you, how likely you are to come under attack, and how valuable your data is.

Archive folder

From “cool uri’s don’t change” – it’s nice to not rename files, and always know where your stuff is. It’s nice to standardize on file naming conventions. Linux has already done it a long time ago, and individuals should be consistent with their files as well. I structure the archives as follows:

/archive
/archive/2018/
/archive/2018/Clients
/archive/2018/Clients//{sow, done, trash, invoices }
/archive/2018/Canon//{src, raw, fin, src2, fin2, thumbs}
/archive//{Meeting_Minutes, Wacom, UIUX_Mockups, Print_Material, Legal_Screenshots, Clients, Content, Canon, WasyaCo, ... }
/archive//hunter # even though there is quick-access "/hunter/", there are archives per-year of 
/hunter/{finances, taxes, forms, job-seeking, done, trash }
/hunter/selfies//...
/--workbench/ # current active client's files
/lib/{images, fonts, forms, games, music, mac_os_x, wordpress, ... }
/doc/{economics, mba, spanish, ... }
/done
the same
/{done, trash, Meeting_Minutes, ...}
/archive//{Meeting_Minutes, ...}
/archive//Print_Material/{Project-1, Project-2, done, trash, ... }

I historically keep very personal stuff in folder “hunter/” – for example, a copy of my drivers license, excel spreadsheets with people’s contact into, some financial data, etc. Note how the name is not very semantic and doesn’t mean anything in particular.

I keep folders “done/” and “trash/” in pretty much every folder. I’m used to having them and they’re not an eye sore. Trash can be destroyed at any time, and “done/” can also be destroyed, but I keep it for archiving and if I need it in the future. Since folder names have rolling date in their names (e.g. “archive/2016/meeting_minutes/201602_meeting_minutes/done/”), it’s not just one big folder “done/”, it’s multiple such folders and you can keep track of your datas somewhat easily.

I don’t like spaces in filenames. I use underscore (_) for spaces pretty much everywhere.

If something is important, capitalize the folder name. If it’s not important, don’t capitalize it. A capitalized folder name often means I spent significant time on it, it’s a standalone project, I’m collaborating with other people on it, or something else significant.

Don’t be afraid to reuse the folder name on several occasions. For example, I have multiple places where I keep forms such as NDA’s, and multiple folders “meeting_minutes/” and “hunter/”.

Leave a Reply