David Weinberger's explanation of the new digital disorder in his book "Everything Is Miscellaneous" illustrates just how effectively the internet has re-ordered our entire lives from the ground up. With the ability to communally tag and categorize content continually, we have completely restructured the organization of information online. This is not only a testament to the Web's democratization, but it also illustrates the power of community - a community which decides for itself how items should be categorized. While the pre-internet, "analogue" methods of categorization could rank items by title or by subject, metatags allow us to rank them by title AND subject AND date, or by length AND genre, and almost every combination of a work's characteristics. This allows users to search for material via multiple paths as opposed to a single, narrow search, allowing users to discover new works just as though they were browsing through a book store. However, this new method is not without its flaws, as human decision is completely subjective.To discuss this new method, Weinberger compares Amazon.com's model to the Dewey Decimal System. "Dewey created a single way to cluster books; Amazon finds as many ways it can. Melvil Dewey took the design of the system upon himself; Amazon lets anyone create her own category, give it a fun name, and publish it" (p. 61). But what happens when categories (or tags) are chosen in haste, or only few are added, or they're misspelled, or arbitrarily chosen? Who monitors them?
The new system, as rounded, encompassing, and editable as it is, still does not always provide the most accurate results. To test just how accurate this new tagging system is, I tried a very amateur and probably quite flawed experiment. Using Flickr.com, I randomly chose 16 words, and searched for them as "tags only" to see how the first picture that showed up on the page correlated with the tag. First, I searched each tag by "most recent," then I searched by "most interesting." And my results were most interesting.
The first thing I noticed was that the results ranked by "most interesting" had substantially more tags than those that were "most recent." Not surprising. But, most of the tags were added by the author, showing me that the owner probably often goes back to his or her photos and adds tags as needed over time. Secondly, I noticed that while other users added more tags to the pictures that were "most interesting" than those that were "most recent" (for obvious reasons), the difference wasn't all that much. This tells me that users rarely tag others' material. And if they did, it was usually only one other tagger. A third characteristic I saw was that the "most interesting" pictures had many more "odd tags" that had nothing to do with the picture than the "most recent." This tells me that even though people continue to add tags to their photos, they can be very poorly categorized at times. This leads me to my final observation. The "most recent" photos and the "most interesting" photos had nearly the same number of photos that did not match their tags at all. For example, when I typed in the tag "world," my first result by "most recent" was a picture of girls at a bar with the Jonas Brothers. Or, when I typed in "telephone," my first results by "most interesting" was a photo of guys hanging out at a pool. This happened about a quarter of the time for both results.
So, while this experiment was by no means "scientific," it did illustrate the fact that human error still abounds in the world of digital tags. While most of the photos had several tags, no tags were more "important" than the other, and no tags were denied access because they were not an accurate depiction of the picture. But that could be the author's decision. Because these models are completely subjective, it's up to the community to correct any problems, and perhaps Flickr.com users don't frequent others' pictures enough to monitor that. Flickr.com is just one site that allows tagging, and perhaps with time, the system can become more reliable. As Weinberger states in the book, "such a cluster of photos is not a true case of a family resemblance, because all of those photos do indeed have one characteristic in common: Someone has tagged the 'italy.' But like a family resemblance, there is no single explanation of what makes 'italy' an appropritate tag" (p. 197). And we realize that dealing with the internet is "sort-of, kind-of," and that is fine because it's a "third-order mess."
Obviously, the new digital order has opened doors in terms of discovering new material and searching data that have never been opened before. It provides a way for people to organize the constantly updated information on the web, but it also allows that information to be as messy as ever. That is the beauty of the internet, and it's what Tim Berners-Lee set out to do - make it a "permission-free zone" (p.189). Despite our inherent human nature to have order, we are sitting amidst orderly chaos because the information is continually changing. When we collaborate on the Web, we open ourselves to more than just what we believe to be true, even if our photo of a tree is tagged as "Chris." We should appreciate more that someone cared enough to add a tag on our photo than whether or not the tag fits our ideas.
No comments:
Post a Comment