Vivienne T Lee, 45Sunnyvale, CA

2009010, Apr 16, 2009

Oct 15, 2007

11/872064

Jonathan L. Conradt - San Jose CA, US
Vivienne C. Lee - San Jose CA, US

MICROSOFT CORPORATION - Redmond WA

H04N 5/445

725 46

Recommendations from social networks is described. In embodiment(s), communications data that is representative of communications between users of client devices can be collected. The communications data can then be evaluated to associate the users that communicate with each other, and a social network can be created that includes the associated users. Media content data that is representative of media content utilized by the associated users in the social network can be complied. A media content recommendation for a user can then be generated based on the compiled media content data for the associated users in the social network.

2014024, Aug 28, 2014

Feb 22, 2013

13/773666

- Redmond WA, US
Vyacheslav Leizerovich - Sunnyvale CA, US
Vivienne Catherine Lee - Palo Alto CA, US
Christopher Elliott Fry - San Francisco CA, US
Paul Cooper - San Jose CA, US

Microsoft Corporation - Redmond WA

H04N 5/63

348730

When a set-top box determines to enter a low-power state, the set-top box determines a wake-up time based on a recording schedule, a maintenance schedule, and a boot spread. Before transitioning to the low-power state, the set-top box sends a message to a schedule server with the wake-up time. The server provides services to many set-top boxes. The server uses a wake-up time schedule that includes the wake-up times for the set-top boxes that it services to determine if the wake-up time will create too much load for the server given the set-top boxes already scheduled to wake-up around the wake-up time. If the proposed wake-up time will not create too much load, the server may respond to the set-top box that the proposed wake-up is acceptable. If not, the server may propose a new wake-up time based on the schedule.