chores geekery [pb]
teyli turned me on to ChoreBuster!
So you aren't disappointed, I'll say up front, they haven't actually found a way to reduce one's chore load magically via the Internet. It's actually a chore tracking and scheduling tool. You enter people (in my case, me), chores (difficulty and desired frequency and so on), and *pop*, a balanced schedule comes out. (This could be used for almost any scheduling task, not just chores, but that's their target.) The underlying geekery is worth some analysis, too.
Why is this useful? I mean, I can see when I need to do dishes, and I don't have any roommates. Well, I find that the cognitive cost of maintaining/producing a chore list is high. When left to my own devices, I do the same chores frequently and never ever ever do others. As a result, by the time I have to actually get the entire house clean, certain tasks have become monstrosities. It's also very nice for tracking certain things; for me, anything that needs to be done "occasionally," like disassembling the Roomba, is likely to get lost in the shuffle. Finally, just sitting down and coming up with a list of things that should happen, with frequencies, was informative.
Summary: It is one heck of a lot easier to do what you're told than to figure out what to do.
As for geekery: The effort is clearly in beta. The underlying scheduling looks to use fairly non-domain-specific constraint satisfaction. Generally it's pretty clever, but it's too easy to specify unmeetable constraints and break the scheduling. It doesn't crash, but it starts suggesting truly whacked-out schedules. Similarly, if there are really too many things to do for the human-time available (like there always are), the schedule doesn't degrade gracefully. (I actually ended up working around this by reducing the difficulty of each task, so I can sort out conflicts myself.)
The interface is pretty usable, which is impressive, but the display mechanisms and suggestions need work. (Suggesting "vacuum living room" is okay; suggesting "vacuum Brandon's room" is not.) There are a few settings and options that make the scheduling worse rather than better. Overall, it seems brittle - it's best to enter some chores and leave it alone to do the best it can, which is stupid, because it doesn't take advantage of my domain knowledge.
Summary: cute, provides quite a bit of value, but needs work to fulfill its potential.
Anyway, it's free to use, with a very small cost ($2+) to enable certain features. Worth playing with if you (a) care about the chores part or (b) are into HCI, AI planning, or task scheduling.
So you aren't disappointed, I'll say up front, they haven't actually found a way to reduce one's chore load magically via the Internet. It's actually a chore tracking and scheduling tool. You enter people (in my case, me), chores (difficulty and desired frequency and so on), and *pop*, a balanced schedule comes out. (This could be used for almost any scheduling task, not just chores, but that's their target.) The underlying geekery is worth some analysis, too.
Why is this useful? I mean, I can see when I need to do dishes, and I don't have any roommates. Well, I find that the cognitive cost of maintaining/producing a chore list is high. When left to my own devices, I do the same chores frequently and never ever ever do others. As a result, by the time I have to actually get the entire house clean, certain tasks have become monstrosities. It's also very nice for tracking certain things; for me, anything that needs to be done "occasionally," like disassembling the Roomba, is likely to get lost in the shuffle. Finally, just sitting down and coming up with a list of things that should happen, with frequencies, was informative.
Summary: It is one heck of a lot easier to do what you're told than to figure out what to do.
As for geekery: The effort is clearly in beta. The underlying scheduling looks to use fairly non-domain-specific constraint satisfaction. Generally it's pretty clever, but it's too easy to specify unmeetable constraints and break the scheduling. It doesn't crash, but it starts suggesting truly whacked-out schedules. Similarly, if there are really too many things to do for the human-time available (like there always are), the schedule doesn't degrade gracefully. (I actually ended up working around this by reducing the difficulty of each task, so I can sort out conflicts myself.)
The interface is pretty usable, which is impressive, but the display mechanisms and suggestions need work. (Suggesting "vacuum living room" is okay; suggesting "vacuum Brandon's room" is not.) There are a few settings and options that make the scheduling worse rather than better. Overall, it seems brittle - it's best to enter some chores and leave it alone to do the best it can, which is stupid, because it doesn't take advantage of my domain knowledge.
Summary: cute, provides quite a bit of value, but needs work to fulfill its potential.
Anyway, it's free to use, with a very small cost ($2+) to enable certain features. Worth playing with if you (a) care about the chores part or (b) are into HCI, AI planning, or task scheduling.