|
|||
The most blatantly false statement is this one: The diary metaphor has some clear advantages over the desktop metaphor. It is based on the notion that what we have created, modified or even looked at most recently is probably still most important to us. And, Scopeware's inventor maintains, our sense of time is a strong organizing principle that can help us locate a file simply because we remember when we used it last. Actually, the opposite is true. Except for the recent past, humans have very poor memory for sequence or time. We remember things based on content; our memories are associative. We don't say "I want to work on the document I last changed July 7", we say "I want to update that description I wrote about a car I saw." As a baseball fan (and a fan of Sammy Sosa) I have a very vivid memory of a game between the Cardinals and the Cubs where Mark McGwire hit yet another home run, and Sosa ran in from right field to congratulate him. But I can't even tell you what year it happened. Remember that favorite toy you got for your birthday as a kid? Remember how you loved it so much you wanted to take it to bed with you, only your parents wouldn't let you? Which birthday was it? Odds are that the only way you can figure that out is by reconstructing it: "Well, it was when we lived in the old house up in the north part of the city and I moved away from there when I was 8, so it must have been before that. And some of the kids at the birthday party were from school, so I must have been at least 6." We only remember time when it's actually part of the content of the event. (For example, you might remember saying to your parents "This is the BEST 7th birthday I've ever had!!!" and know it from that.) We don't index by time as such. Our memories are not organized linearly. We temporarily keep track of sequence over the course middle term memory, out to maybe two-four weeks; after that, our association of time with memories progressively weakens. There are memories where the time is indelibly fixed as part. I know for certain that the Colombus Day Storm was in 1962, for example. But most memories don't include a time. The more trivial an event, the less likely you are to be able to access it with a timestamp. However, some researchers in the field of human-computer interaction think it's time to throw out thinking about "metaphor" altogether—after all, it hasn't gotten us too far since the 1970s—and to begin designing devices that have no metaphor, no real-world analogy. It's not the desktop metaphor that's holding us back, they say; it's the whole notion that we need to make computers act like something other than what they are. Metaphors for the sake of metaphors owe more to marketing than to solid human engineering. It's easier to raise funding if you can say "We're going to make it the computer equivalent of a can opener" then to try to explain what it actually does. (Well, in the case of a can opener, maybe not.) But eschewing metaphors entirely is not wise; it's not so much that a metaphor is valuable as that the study of real world behaviors tells us what we are and are not good at. Humans have two main kinds of memory: associative and spacial. We are extremely good at remembering where we put something so that we can find it again, and that's the primary power of the desktop metaphor. It's not the fact that it's a desktop that made it successful, it's that it used spacial memory as an access mechanism. In that sense, the desktop metaphor is not ideal because it's relying on a psychological "where" rather than a physical "where". The directory I put a given file into won't change in its virtual position in the disk structure unless I move it, but where it appears on the screen will be modified routinely, both by me and by the system. Another difficulty is the relative sparsity of visual cues. When all the files of a certain type look the same, and when their positions dynamically move, then spacial memory is largely defeated and we have to fall back on other mechanisms. The original memory that computer UI's relied on, and the one which still makes up the foundation of UIs, is associative memory. We remember things because they're relevant; we access based on content. You remembered that toy from childhood because I asked you about favorite toys. But if I'd asked you "What toys did you get on your 9th birthday?" you probably couldn't tell me. In computer UIs, the most common form of associative memory is the filename. We use it to attach a mnemonic term to a file, which we then recognize later. The name of the file will be related to its content in a way which makes sense to us and may not make sense to anyone else. But this is not a very good use. So what's coming? What will be the new interface? That question is on the face of it faulty. The desktop metaphor is now essentially permanent. We're out of the revolution stage and into evolution. This happens sometimes; interfaces take on a life of their own. The QWERTY keyboard was designed in the 19th century, and though the physical form of it (switches instead of levers on a typewrite 
| |||