1

A method comprising:in a computer-based device, storing a data structure which includes information about a plurality of specific actions and a plurality of specific times, said information defining a likelihood that one of said actions will be carried out at one of said times;

using the computer based device for comparing a current time to said times in said data structure;

determining an action which is probabilistically likely to happen in the computer based device at an operation time, where said operation time includes a current time; and

taking an action on the computer based device, based on what is most likely to happen at said operation time, prior to said actually action actually happening.

View Prior Art for Claim

2

A method as in Claim 1, wherein said data structure is a feature vector that stores information indicative of likelihoods of plural different actions at plural different times.

View Prior Art for Claim

3

A method as in Claim 1, further comprising forming an individual data structure for an individual user of the computer-based device and updating said individual data structure based on actions taken by the individual user and times when those actions are taken.

View Prior Art for Claim

4

A method as in Claim 1, further comprising storing sets of information in the data structure that represents generic users and likely actions and times of actions taken by said generic users, and selecting one of said data sets as a set of information to use for the current time.

View Prior Art for Claim

5

A method as in Claim 1, wherein said data structure also includes information indicative of a time between a first action and a second action.

View Prior Art for Claim

6

A method as in Claim 1, wherein said action comprises determining that a program is likely to be started, and responsive to determining that the program is likely to be started, loading at least some part of said program into a memory prior to the program being actually started.

View Prior Art for Claim

7

A method as in Claim 1, wherein said data structure includes times that are spaced by different amounts of granularity depending on time of day, so that a first amount of granularity between times is present at a first time of day, and a second amount of granularity between times is used at a second time of day.

View Prior Art for Claim

8

A method as in Claim 1, wherein said action comprises determining a likelihood that how likely it is that the computer will be used at said operation time, automatically selecting an aggressive sleep state for a first likelihood that the computer will be used and automatically selecting a less aggressive sleep state for a second likelihood that the computer will be used.

View Prior Art for Claim

9

A digital video recorder, comprising:a recording part that includes a computer device that records a video program and which allows watching the video program, allows pausing the video program during recording, and rewinding the video program during recording, said recording part storing data indicative of the video program on a memory device; and

a detector, that detects an indication of activity, wherein said computer device operates to continually record during said indication of activity and automatically stops recording responsive to there being no indication of activity for a specified time period.

View Prior Art for Claim

10

A digital video recorder as in Claim 9, wherein said detector detects operation of the digital video recorder, determines a time since a last operation, and stops the recording when said time is greater than a specified amount.

View Prior Art for Claim