{"id":1292,"date":"2006-07-23T14:44:50","date_gmt":"2006-07-23T18:44:50","guid":{"rendered":"http:\/\/mjtsai.com\/blog\/2006\/07\/23\/quantitative-design\/"},"modified":"2006-07-23T23:01:22","modified_gmt":"2006-07-24T03:01:22","slug":"quantitative-design","status":"publish","type":"post","link":"https:\/\/mjtsai.com\/blog\/2006\/07\/23\/quantitative-design\/","title":{"rendered":"Quantitative Design"},"content":{"rendered":"

\r\nHumanized<\/a> has an interesting article about measuring interface efficiency using information theory (via Daniel Jalkut<\/a>). It’s an idea that deserves some thought, but for the moment I’m unconvinced. The thesis seems reasonable, but the model doesn’t capture the nuances of the real-world examples and therefore it makes predictions that seem intuitively wrong.\r\n<\/p>\r\n\r\n

\r\nThe natural language examples make me suspicious right off the bat. The\r\nauthor does draw a distinction between information and meaning, but\r\nsubsequent paragraphs sometimes conflate these ideas.\r\n<\/p>\r\n\r\n

\r\nWhy is efficiency, defined as the amount of information entered,\r\nimportant? There are so many other factors to consider (including time spent waiting, cognitive load, propensity for errors, and ease of error recovery). My first\r\nreaction is that it’s better to bundle all the known and unknown factors\r\ntogether and measure efficiency using the time to complete the task.\r\n<\/p>\r\n\r\n

\r\nIn the real world, we rarely start from scratch and try to set a\r\nspecific time. The initial state of the watch and the shape of the delta\r\nmatter. Usually, we want to adjust the watch a few minutes forward or\r\nbackward if it’s gained or lost time. Or we want to leave the minutes\r\nalone and shift the hours to account for daylight savings time or\r\ntraveling through a time zone. My digital watch lets me do the daylight\r\nsavings adjustment using three button presses. Does that make it more\r\nthan 100% efficient?\r\n<\/p>\r\n\r\n

\r\nAnalog watches also make it easy to make small changes to the time,\r\nalthough sometimes the only way to go backwards a few minutes is to go\r\nforwards nearly a whole day. Surely these factors matter, but the model\r\ndoesn’t capture them.\r\n<\/p>\r\n\r\n

\r\nThe article asserts that turning the crown of an analog watch represents\r\n9.5 bits because there are 720 possible times. The way this is presented\r\nseems like reasoning backwards to me.\r\n<\/p>\r\n\r\n

\r\nWhat if an analog watch had two knobs to turn, one for each hand. There\r\nare 12 positions for the hour hand and 60 for the minute hand. In the\r\nreal world, this makes it much easier to set the time because you don’t\r\nhave to go around and around to get to the right hour. But according to\r\nthe model, the efficiency has gone down because we’re still choosing one\r\nof 720 possible times, only now we have to choose between two knobs, too.\r\nAfter all, the digital watch was penalized for having two untouched\r\nbuttons while in the quasimode of advancing the hour.\r\n<\/p>\r\n\r\n

\r\nHere are two designs that the model predicts would be good:\r\n<\/p>\r\n\r\n