Wow, thanks for the quick reply!
I'm using "window" to mean slots like "3pm - 4pm" and "budget" for allocations like "30 min / day". Let me know if KoalaSafe already has terms for these, so I can use your language!
Your question makes me wonder if I wasn't clear about my feature request, so: I don't think I need a finer-grained way of filtering activities. I do want to be able to limit daily use by two measures, concurrently: the window (when it can happen) and the budget (how long it can happen for). These filters should take effect simultaneously.
Replying to your suggested timer method: So far, I really like how the KoalaSafe doesn't demand user interaction - that feels like it would be less easy for the user, and for your product those users are probably quite young to be operating a timer. (I'm 40; I sometimes still forget to clock in and out of tasks on our work's time tracking system!)
And for this feature, I meant "watching videos" as just an example of activity, to be clear. You're already detecting what behaviours are engaged (Youtube vs Minecraft) fine for my proposed example.
A timer might do the trick, but I'd rather it was seamless/invisible to the user - that activities could be budgeted, and could be summed and closed off when they hit budget. This would need some sensitivity too - stopping the video two minutes from the end would be pretty hard on a young person, but that can already happen if the window comes around. I can see the appeal of timers too - they start to let you reach which user is on that device.
Some challenges with timer approach:
- I can imagine cases where one savvy sibling learns to swap devices with another when they sense the budget limit approaching
- I can imagine cases where a savvy kid learns to start their "wikipedia" timer and goes back to minecraft. So a "watching videos" timer may not be effective.
- Interfaces which encourage people to make "bad decisions" (like the above two) are to be avoided IMO.
If I can speculate wildly ... degrading service as the timer passes might seem "more natural" to the user. So when the YouTube budget limit is hit, reduce bandwidth to that device over the next few minutes to ease the user out of the experience. (This might also enrage the user. It feels pretty hostile, but I'm spitballing other approaches to "hard stop".)