Abstract: A social data feed, comprising a set of data items submitted by individuals in a social network, may be presented to the user, but the frequency of new data items appearing in the social data feed may be overwhelming, and may cause the user to miss data items that may be of interest. Instead, potentially interesting data items may be selected and peripherally presented to the user, e.g., as a tiled set of thumbnail images presented near the right edge of the display, adjusted for low contrast with the display background. Peripheral presentation may suggest potentially interesting data items to the user, thereby promoting serendipitous discovery of interesting data items, while reducing the distraction of the user while viewing the social data feed. Metadata of an interesting data item may be displayed upon user selection (e.g., pointer hovering), and interesting content items within a selected data item may be highlighted.

Abstract: The claimed subject matter relates to architectures for facilitating network-accessible or local backup or restore features in a manner that leverages event-oriented contextual information associated with one or more users of the data to be backed up or restored. In particular, a first restore-based architecture can interface with an associated second backup-based architecture that hosts or maintains a backup data store in order to retrieve a version of the data file that is desired by a user. Additionally, the first architecture can obtain a log of contextual event descriptions that can be aggregated by the second architecture or obtained independently from, e.g., a social networking service or a calendar application associated with the user.

Abstract: Users of various devices may participate in communication sessions, but only if the interfaces and components of such devices feature a minimum set of capabilities involved in the communication session type; e.g., in order to participate in a teleconference, the device of every user has to be capable of sending and receiving audio. Instead, a communication session server may automatically convert session items received from each user into interface-specific representations for each other user, based on the capabilities of the interface and device of the other user. For example, speech recognition may transform speech to text output; speech synthesis may transform text input to speech; and video recognition may identify individuals and gestures in a video stream and represent this information as text. The server may therefore host a “mixed” communication session, where each user may fully participate irrespective of the capabilities of the chosen device and interface.

Abstract: A portable computing device may be configured to allow a user to interact with one or more displays in a retail environment. The portable computing device may establish a communication link with a communication device in the retail environment and the user may be able to use the portable computing device to navigate through retail data and have it displayed on the display device in the retail environment.

Abstract: The claimed subject matter relates to architectures for facilitating network-accessible or local backup or restore features in a manner that leverages event-oriented contextual information associated with one or more users of the data to be backed up or restored. In particular, a first restore-based architecture can interface with an associated second backup-based architecture that hosts or maintains a backup data store in order to retrieve a version of the data file that is desired by a user. Additionally, the first architecture can obtain a log of contextual event descriptions that can be aggregated by the second architecture or obtained independently from, e.g., a social networking service or a calendar application associated with the user.

Abstract: The claimed subject matter relates to architectures for facilitating network-accessible or local backup or restore features in a manner that leverages event-oriented contextual information associated with one or more users of the data to be backed up or restored. In particular, a first restore-based architecture can interface with an associated second backup-based architecture that hosts or maintains a backup data store in order to retrieve a version of the data file that is desired by a user. Additionally, the first architecture can obtain a log of contextual event descriptions that can be aggregated by the second architecture or obtained independently from, e.g., a social networking service or a calendar application associated with the user.

Abstract: Embodiments extend functionality of a web page using cloud-based page applications. A page application provider, executing as part of a web browser, receives notification of a web page being accessed by a user of a user computing device. One or more page applications associated with the web page are identified. The identified page applications provide functionality related to the web page and, in some embodiments, interact or share data with the web page. The page application provider presents the identified page applications to the user. Based on a selection of at least one of the presented page applications, the selected page application is provided for execution by the user computing device. In some embodiments, the correlations among the user, web pages, and page applications are available from a cloud-based page application service.

Abstract: The claimed subject matter relates to architectures for facilitating network-accessible or local backup or restore features in a manner that leverages event-oriented contextual information associated with one or more users of the data to be backed up or restored. In particular, a first restore-based architecture can interface with an associated second backup-based architecture that hosts or maintains a backup data store in order to retrieve a version of the data file that is desired by a user. Additionally, the first architecture can obtain a log of contextual event descriptions that can be aggregated by the second architecture or obtained independently from, e.g., a social networking service or a calendar application associated with the user.

Abstract: The claimed subject matter relates to architectures for facilitating network-accessible or local backup or restore features in a manner that leverages event-oriented contextual information associated with one or more users of the data to be backed up or restored. In particular, a first restore-based architecture can interface with an associated second backup-based architecture that hosts or maintains a backup data store in order to retrieve a version of the data file that is desired by a user. Additionally, the first architecture can obtain a log of contextual event descriptions that can be aggregated by the second architecture or obtained independently from, e.g. a social networking service or a calendar application associated with the user.

Abstract: Queries may be issued against an image store to produce a set of image instance relating to images in the image store that relate to the query. The relevance of the images to the query may be depicted by scaling the image instances according to the predicted relevance of the image to the query. The image instances may be further positioned within the image instance set query result, e.g., by clustering according to image relatedness or by similar predicted relevance of the images to the query terms of the query. The image instances may also be presented as smoothly zoomable images, such that the user may zoom in on the images in an efficient manner that facilitates realtime, gradual zooming with reduced resampling inefficiency.

Abstract: When presenting a news feed (e.g., statuses of contacts in a social media network), it may be inefficient to display the data items in a single column according to a single filter, such as a designated group of contacts. Rather, from the data feed, a set of at least two data subfeeds may be generated based on different filters, and the at least two data subfeeds may be currently presented to the user (e.g., as two or more vertically aligned columns, each presenting a different set of data items based on a different filter.) The user may prefer the concurrent presentation due to the contextual separation of the data items according to different filters, and may more easily administrate the filters and groupings; e.g., dragging data items from one column to another column may alter the filters to achieve a more preferred sorting of data items in future presentations.

Abstract: Data items of various data feeds (such as articles posted to a website or entries in an RSS feed) may be associated with various topics, but a user may be unable to monitor a large number of data feeds, and to avoid previously reviewed data items while searching for new data items. An aggregation service may monitor many data feeds, perform an automated topical evaluation of the data items, and generate a set of topical data feeds. Additional services may also be applied, such as filtering the topical data feeds by various criteria, translating data items from a native language into another language, and removing equivalent data items, such as articles redundantly covering the same news story. A centralized or cooperatively distributed service may scale for improved efficiency and value, since each data feed may be monitored and each data item received and evaluated on behalf of many users.

Abstract: A computing device may be configured to allow a shopper to interact with one or more displays in a retail environment. A perception device may detect human actions in the retail environment and the shopper may be able to use human movements to navigate through retail data and have it displayed on the display device in the retail environment.

Abstract: Embodiments extend functionality of a web page using cloud-based page applications. A page application provider, executing as part of a web browser, receives notification of a web page being accessed by a user of a user computing device. One or more page applications associated with the web page are identified. The identified page applications provide functionality related to the web page and, in some embodiments, interact or share data with the web page. The page application provider presents the identified page applications to the user. Based on a selection of at least one of the presented page applications, the selected page application is provided for execution by the user computing device. In some embodiments, the correlations among the user, web pages, and page applications are available from a cloud-based page application service.

Abstract: A portable computing device may be configured to allow a user to interact with one or more displays in a retail environment. The portable computing device may establish a communication link with a communication device in the retail environment and the user may be able to use the portable computing device to navigate through retail data and have it displayed on the display device in the retail environment.

Abstract: Data items of various data feeds (such as articles posted to a website or entries in an RSS feed) may be associated with various topics, but a user may be unable to monitor a large number of data feeds, and to avoid previously reviewed data items while searching for new data items. An aggregation service may monitor many data feeds, perform an automated topical evaluation of the data items, and generate a set of topical data feeds. Additional services may also be applied, such as filtering the topical data feeds by various criteria, translating data items from a native language into another language, and removing equivalent data items, such as articles redundantly covering the same news story. A centralized or cooperatively distributed service may scale for improved efficiency and value, since each data feed may be monitored and each data item received and evaluated on behalf of many users.

Abstract: Users of various devices may participate in communication sessions, but only if the interfaces and components of such devices feature a minimum set of capabilities involved in the communication session type; e.g., in order to participate in a teleconference, the device of every user has to be capable of sending and receiving audio. Instead, a communication session server may automatically convert session items received from each user into interface-specific representations for each other user, based on the capabilities of the interface and device of the other user. For example, speech recognition may transform speech to text output; speech synthesis may transform text input to speech; and video recognition may identify individuals and gestures in a video stream and represent this information as text. The server may therefore host a “mixed” communication session, where each user may fully participate irrespective of the capabilities of the chosen device and interface.

Abstract: A social data feed, comprising a set of data items submitted by individuals in a social network, may be presented to the user, but the frequency of new data items appearing in the social data feed may be overwhelming, and may cause the user to miss data items that may be of interest. Instead, potentially interesting data items may be selected and peripherally presented to the user, e.g., as a tiled set of thumbnail images presented near the right edge of the display, adjusted for low contrast with the display background. Peripheral presentation may suggest potentially interesting data items to the user, thereby promoting serendipitous discovery of interesting data items, while reducing the distraction of the user while viewing the social data feed. Metadata of an interesting data item may be displayed upon user selection (e.g., pointer hovering), and interesting content items within a selected data item may be highlighted.

Abstract: One or more techniques and/or systems are disclosed for presenting members of a shared online space to a user of the shared online space. A relationship value is determined between the user and a member of the shared online space by: determining an interaction value for the user and member relationship based on the user's and member's co-presence in the shared online space; determining a social network value for the user and member relationship based on a number of social network connections between the user and member; and combining the interaction value and social network value. The relationship value is associated with a specified visual representation of the member used in the shared online space, and two or more visual representations of members are scaled in the user's display in the shared online space in order to fit an available screen space, based on the relationship value.

Abstract: When presenting a news feed (e.g., statuses of contacts in a social media network), it may be inefficient to display the data items in a single column according to a single filter, such as a designated group of contacts. Rather, from the data feed, a set of at least two data subfeeds may be generated based on different filters, and the at least two data subfeeds may be currently presented to the user (e.g., as two or more vertically aligned columns, each presenting a different set of data items based on a different filter.) The user may prefer the concurrent presentation due to the contextual separation of the data items according to different filters, and may more easily administrate the filters and groupings; e.g., dragging data items from one column to another column may alter the filters to achieve a more preferred sorting of data items in future presentations.