![]() It seems simpler in the end just to keep my files in a hierarchy where I’ll remember where they are and can get to them when I need them. Am I really going to go through all my thousands of MP3s and photos to tag each and every one individually? The labor involved would be enormous. Curiositry points out several options such as Supertag and TagSpaces but singles out TMSU as the most promising one so far.Įven if we had the tools, though, we’re still in a bit of a bind. In lieu of that we have to rely on third-party tools that fall back on one of the workarounds like putting the tags in the file name. There is no technical barrier to this solution - it just hasn’t been implemented, at least not in a way that’s well-supported and portable. This is where tagging can come in handy, but where to put the tags? In the file name? In a separate database? Or in a “sidecar” file that accompanies the file it describes? The proper solution to this problem are file systems and user interfaces that allow us to add any metadata we like to the file itself. If I delete the link, should the destination file get deleted too?ĭirectories are hierarchical, and life is not. If I want to see what other files link to this one, I can’t. ![]() ![]() ![]() If I move the destination file, the link breaks. Ted Nelson’s illustration of the structure of information: “the perplex”.Ĭuriositry briefly floats links or shortcuts as a workaround for this problem, but they’re a hack too. I want to put it in a folder with the other songs released on the same album, but the same song was also released as a single and that should have its own folder too. It’s punk so I want to put it in a folder with my other punk music, but it’s also goth. From my own experience, let’s say I have an MP3 file of a song. He’s not wrong about hierarchy being a hack. Ted Nelson, The Future of Information (1997), p. The fact that projects overlap and categories overlap is not reflected, as it should be, in the system of filing. Hierarchical files were a cheap hack of an idea around 1947 that has come to be treated as God’s Will. Computers have been very carefully and intentionally set up with hierarchical files because their guardians haven’t come up yet with a better conceptual structure, and believe that is the right way for them to be. They tell beginners that computer files are necessarily hierarchical. One of his many pet peeves about the modern computer ecosystem is its hierarchies: ![]() Ted Nelson is a designer who worked on computer document systems for decades before Tim Berners-Lee invented the World Wide Web. In Curiositry’s example, should “song lyrics” go under “writing” or “music”? The hierarchical nature of the file system inherently means that the same file may belong in more than one folder. Our first instinct might be to sort everything out by sticking it in the correct folder in the file system’s hierarchy, but that’s not always so easy. Many of us have gigabytes of MP3s in our music folders, gigabytes of photos, and a folder for downloads that’s stuffed with everything we thought we’d look at but haven’t had a chance to for the last six months. In their blog post Taxonomy is Hard Curiositry writes about some common problems with computer file systems. ![]()
0 Comments
Leave a Reply. |