Tag.....

I think that tagging might be a solution in search of a problem. Its a nice idea, but I can't see it being something that 'normal people' would routinely use. Then again, that was one of the original comments about the telephone, I think.

Tagging is the concept of supplying information that describes something in a way that it can be used by a search engine. The geek-speak phrase is metadata (I will resist the chance for a horrible pun). Metadata means 'data which describes data'; one of its manifestations are keywords, aka tags. Documents with keywords are 'tagged'; that is, you can search for the document by using the keywords (and of course retrieve all documents with those same keywords, not to mention, if you've got a Bayesian search engine, documents that are tagged with words that often appear in conjunction with those words). Tagging is particularly of use when you are filing something that cannot be easily searched by a text search engine (as all are, I think; at the least, they are all character search engines). For example, I might want to ease retrieval of a photograph of myself by using tags such as Bill, Studly, and Hunk. (Though the person getting that picture back, having entered Studly as a keyword, might complain at the result. Smart search engines can (I think) bias their returned results based on that complaint.)

What brings me to all of this is an article in the current issue of The Economist which talks about the growing power of desktop search software such as the Google Desktop. And it really is powerful: I put it on my home PC and did a web search for a certain phrase; I was surprised to see it also return documents on that PC that contained the phrase. I didn't know those documents were there. But the article goes on to hyperventilate about tagging as the reason why the current directory/file structure is antediluvian and should be scrapped; just tag everything, the article (or actually Marti Hearst, a 'computer scientist at the University of California at Berkeley's School of Information Management and Systems) says, and your powerful search engine will find it -- no need to file it in a specific place. What the article doesn't say is that the current structure forces you to be methodical in where you put things; tagging doesn't, so when I tag that photograph of myself with the phrases Angkor Wat, Transistor, and Kumquat, that picture is gone -- I'm never going to find it again. If its in a folder marked Images, though, I at least have a chance of retrieving it.

I'm being picky, of course; tagging is done manually now, but software can and most likely will improve it (as an example, the article mentions that digital cameras automatically put the time and date, as well as exposure information, into the image file (my wife said, wryly 'Assuming you set the time and date right!') -- those values can be used to do some rudimentary tag searching.) I suspect that text creation tools will begin to have the ability to create their own keywords. I've mentioned already, I think, how I like a tool called Personal Knowbase, from Bitsmith Software; its a note taking tool that allows you to file information based on keywords that can be used for easy retrieval. One of my minor complaints is that it won't scan the text and generate its own suggested keywords. But that's okay -- no other text creation tools do automatic tagging, either. When that happens, tagging will take off. But not quite yet

Which reminds me of a story, one that I believe to be true. A man moved to Montana with his German born wife and when registering her car wanted to give her a license plate with the German phrase Guten Tag -- but that was too long, so he settled for GUTNTAG. Whereupon the fellow at the license bureau nodded and said 'She's a hunter?' At the mans bemused denial, the license fellow pointed to the desired phrase 'Gut? And Tag?'

Course, there is this, so maybe not.

Tag....you're it.