Slashdot Mirror
How Do You Organize Your Data?
kpellegr asks: "After returning from a well deserved holiday, I was faced with an exploding inbox. While organizing and deleting my mail, I realised I was having trouble classifying each mail into one specific folder. I had the feeling I should be able to link to one email from several folders (e.g. product information should be linked to from the 'vendor' folder, as well as from a specific project folder where this product is used). The more I thought about this, the more I realised that trees (such as the Windows filesystems) are not really ideally suited for organizing data. On UNIX-like filesystems, symbolic links allow the creation of simple graphs for organising data, but I have the feeling data could be organized more efficiently. How does the Slashdot crowd organize their data? How do you manage files, email, contacts, meetings and all the relationships that might exist between them?"
... and grep
;-)
...organizing data was quite simple for Mac users. (All you Mac people out there have to admit: You're right with me on this. Don't lie!)
The process was simple:
- Save everything to the Desktop.
- When you couldn't see the background pattern anymore, create a new folder called "Desktop crap" or something, and move all the files into it.
- Move the folder on to the hard drive.
- Repeat.
:-DCAn'T CompreHend SARcaSm?
And when you delete a message from one folder, it's deleted from all of them!
If you're aware of the virtual folder concept, this can be very powerful. But is a first-time user going to expect it? Of course not, he thinks the folders work like everywhere else, and copy means make a copy, not just a link. So many emails were lost at the last job where I used it, for this reason.
BeOS used file attributes and file system queries to organize data. Longhorn's WinFS is built on this concept. The real question isn't how to organize your files, it's why does your data need to be in files? Why are folders so closely entwined with our computing experience? This type of grouping is best suited for your clothes in your dresser. In real life, tossing everything into a pool and pulling out what you need by characteristics ("attributes") is much more useful.
Magic prioritization:
Step 1: Reply to the email ASAP with a question. Emails are (at best) only half thought out to begin with, so this is typically necessary anyway. It is best if it is something the person will have to look up or follow up on, rather than something they will know right away.
Step 2: Delete the email.
If it takes the person two weeks to get back to you, you know that it is not important AND you just bought yourself two weeks.
If your phone rings 30 seconds after sending the email, it is urgent.
Everything else falls in the middle somewhere, but you get the idea. In my case, probably close to 5% of these never get a response (or get a quick reply that they will look into it, but no final answer), or having been forced to think about their request, they send a response that they need to work on the details of the request before I begin working on it.
You can take that a step further and use IMAP with fetchmail and procmail. Set it up once and fine tune as needed and have one set of folders and filtering available to any IMAP client.
The backup is easy if working with standard mailbox style folders because the format is text, readable by any viewer. You can tgz your mail directory to a file via cron. I back mine up on a rotating basis to a different drive. For things I know I will never need but want to keep anyway or for archiving important things, I create a new IMAP folders with my client, move the messages over to that folder, tgz it and move it out of the mail directory. If I ever need it, I can extract it back to the mail directory and view it again or I can add more mail to the archive file later with a few commands I am not familiar with the native format of any mail clients anymore because I have been using IMAP for years. I switched for two reasons, I got tired of always trying to convert proprietary mail formats everytime I wanted to change mail clients and I wanted access to all of my mail regardless of what type of machine or where I was coming from. I will never go back no non-IMAP. The fetchmail and procmail functionality are an added bonus. You get the most from IMAP when it is running 24/7 on a stable machine somewhere on your local/home network. If you don't have such a thing already in place, it might not be worth the initial effort.
Bad boys rape our young girls but Violet gives willingly.
The more I try to strongly 'type' my data, the longer it takes to deal with it. Big general buckets work the best for me.
I don't always succeed at that, but I do try. Sometimes I don't produce the same neural network or mneumonic-map that I did two years ago for the same datum, and then it gets lost. So the more general, the better.
That is 3 different results from the same user action! So how do folders work everywhere else?
Well, IMO the real problem is not whether one maufacturer or another has his own user interface rules, it is the fact that folders and documents were introduced as the universal metaphor for arranging data on a computer in the first place.
And now we are stuck with the restrictions imposed by that representation, which will often lure first-time users into believing that just because it looks like real-life a folder it will behave like a real-life folder. No matter how you then try to squeeze the concept of links, views, etc, into some kind of association with this rather limited concept, you are likely run into problems. What, really would be the real-world counterpart of a symbolic link, a virtual forlder (!), a view, etc?
The file/folder metaphor comes from an age where files were few and far between for the average user. Maybe we need a completely new user interface concept to deal with today's overload of data.