Kurt McKee

lessons learned in production


Hey there! This article was written in 2006.

It might not have aged well for any number of reasons, so keep that in mind when reading (or clicking outgoing links!).

The shift, unexpected

Posted 24 September 2006 in data, information, and metadata

I have been in many respects a digital pack rat. I've meticulously kept track of as much data as I can, going so far as to engineer bizarre directory hierarchies in which to store my Precious, my data. When I was younger, I even envisioned one day creating a librarian to help me manage my Alexandrian archives.

Meh. I've gotten over that to some extent, and have become more concerned with information. Why keep a closet full of data when you can have neatly-organized information on a bookshelf? Unfortunately, I'm running into a new problem.

My metadata is growing unwieldy. This is for the most part a failing of my tools, but an unexpected failing nonetheless. My photos have brought this problem to light: while I've removed a number of darkened, blurry, or otherwise unusable photos, I'm now fighting an avalanche of tags. The worst two offenders are People and Places with almost 150 and 50 tags, respectively.

The other area where I'm feeling a crunch is my private bookmark collection. Did I ever mention I went on a hunt to find the websites of everyone I knew? I was very successful, but today I wiped out almost all of those bookmarks. My bookmark count dropped by about half, but I only lost 3 or 4 tags in the process (!).

Perhaps you're wondering what the big deal is with bookmarks and photos?

The big deal with those two types of data are that they almost cannot be searched for. A bookmark is a title and a URL, which is a scant amount of text to search for. Photos? There's absolutely no text at all. I have to provide all of the information myself if I'm to find anything 10 years from now.

I once attempted to assign context and meaning to a photo or bookmark by storing it in a specific folder, but the system was too limited and I ended up with quite a bit of clutter. The innovation of "tagging" data has allowed me to turn my data into information. Unfortunately, those tags have themselves turned to clutter.