Abstract: In one embodiment, a method includes identifying candidate content associated with a user of a computing device, selecting, from the candidate content, cache content to be stored in cache storage of the computing device for access by the user when the computing device does not have network connectivity, and storing the cache content in the cache storage of the computing device. The cache content may be based on information associated with a user node that represents the user in a social graph. The cache content may include entities liked by the user, friends of the user, and/or entities of interest to the user. The cache content includes web pages accessed by the user and/or web pages referenced by content created by the user. The cache content may include information related to past, current, and/or predicted actions of the user, such as social network posts, travel itineraries, and geographic locations.

Abstract: In one embodiment, a method includes inferring an intent of the target user to post to a social-networking system based on one or more information items. One or more information items may be assembled that are relevant to one or more of the information items from which the intent to post was inferred. A prompt to post may be sent to the target user comprising one or more of the assembled information items. One or more indications of one or more reactions of the target user may be received from a client system of the target user. The one or more reactions of the target user may be used in both inferring a future intent to post on the part of future target users and assembling information items for future target users.

Abstract: A system and a method are disclosed for analyzing health-related data associated with a user and suggesting goals to the user to improve health. The system and method monitor progress towards the goal.

Abstract: In one embodiment, a method includes identifying candidate content associated with a user of a computing device, selecting, from the candidate content, cache content to be stored in cache storage of the computing device for access by the user when the computing device does not have network connectivity, and storing the cache content in the cache storage of the computing device. The cache content may be based on information associated with a user node that represents the user in a social graph. The cache content may include entities liked by the user, friends of the user, and/or entities of interest to the user. The cache content includes web pages accessed by the user and/or web pages referenced by content created by the user. The cache content may include information related to past, current, and/or predicted actions of the user, such as social network posts, travel itineraries, and geographic locations.

Abstract: In one embodiment, a method includes inferring an intent of the target user to post to a social-networking system based on one or more information items. One or more information items may be assembled that are relevant to one or more of the information items from which the intent to post was inferred. A prompt to post may be sent to the target user comprising one or more of the assembled information items. One or more indications of one or more reactions of the target user may be received from a client system of the target user. The one or more reactions of the target user may be used in both inferring a future intent to post on the part of future target users and assembling information items for future target users.