Microsoft Technology Licensing, LLC

Facebook, Inc.

Greg D Schechter

Shanku Shivabrata Niyogi

Leonardo Esteban Blanco

Adam M Smith

Sriram Subramanian

Joseph S Beda, Iii

Michael John Hillberg

Gregory David Swedberg

David J Kurlander

Matthew E Gibbs

Daniel Ryan Lehenbauer

Jevan D Saks

Chun-Fu Ricky Yeung

David Michael Miller

Gerhard A Schneider

Calvin Patrick Grunewald

Edward Kenneth O'Neil

Matthew W Calkins

Daniel Tomko

Rajat Goel

Kristen Nilsen Nye

Chris Fortier

Owen H Coutts

John D Rinaldo, Jr

Matthew Jay Kotler

George G Robertson

Michael Dean Stokes

Glenn Frederick Evans

Curtis Glenn Wong

Bongshin Lee

Ashraf A Michail

Oliver H Foehr

Jerry Jay Dunietz

Zhaowei Charlie Jiang

Roland Lucas Fernandez

Danyel Aharon Fisher

Kenneth S Reneris

Murali R Krishnan

Donald D Karlov

Serge Smirnov

Brendan Joseph Clark

Salim S AbiEzzi

Sam J George

Gurpratap Virdi

Graham Arthur Wheeler

David Ornstein

Vijaye Ganesh Raji

Joseph Ternasky

Donald B Curtis

Daniel Emerson

Dmitry Robsman

Wenbo Zhang

Kevin T Gallo

David Francis Gainer

Namita Gupta

Michael Marucheck

Yann Christensen

David C Brown

Kenneth Lawrence Young

Eldar Azerovich Musayev

Wei Zhu

Lawrence Brian Ripsher

Mikhail V Leonov

Adrian Secchia

Jason Scott Sobel

Colin L Campbell

Gilles Khouzam

Christian Olaf Abeln

Federico Schliemann

Michael R Harsh

Morten Holm-Petersen

Stella Yick Chan

Vladimir Fedorov

Jun Li

Eric S Vandenberg

Akshay Johar

Sameer Avinash Nene

Kurt B Jacob

Sameer Bhangar

Joe King

Prashant Ratanchandani

Benjamin L Chronister

Andrey E Arsov

Todd B Knoblock

Joseph Keith Croney

Benjamin Golub

Conal M Elliott

John Gossman

Lawrence Wayne Olson

Ian Hanschen

Severan Sylvain Jean-Michel Rault

Regis L F Brid

Ian Michael Dunmore

Alexandre Gorev

Weibing Zhan

Regis Lucien Francis Brid

Patrick G Mullen

Loic Deniel

Elizabeth K Nelson

Marc Benjamin Phillips

An online system receives an advertisement ('ad') request identifying an objective for presenting the ad request. The objective specifies a type of interaction with the presented ad request. A threshold battery level is associated with the type of interaction in the ad request that specifies a minimum battery level of a client device to complete the type of interaction. When selecting content for presentation to a user, the online system obtains a battery level of a client device associated with the user and compares the obtained battery level to the threshold battery level. Based on the comparison, the online system determines if to present the ad request to the user. For example, the comparison may prevent presentation of the ad request or penalize a value associated with the ad request if the obtained battery level does not satisfy the threshold battery level.

Accounting for battery level of a client device when selecting content for presentation by the client device

In one embodiment, a method includes receiving, by a server computing device, a notification which includes: information associated with the identity of the user associated with the user-initiated event or with the identity of the client computing device; a variable, pre-determined string of symbols for the user-initiated event associated with the notification; a variable number describing a value of an attribute associated with the user-initiated event; and a variable parameter describing an attribute of the user at or near the time of the event or an attribute of the variable number. The method includes correlating, by the server computing device, the user-initiated events with a user; and determining, by the computing device, a description of the user-initiated event associated with the notification based at least in part on the string of symbols.

Electronic notifications

A data set may be represented by samples of multiple resolutions, such as an image represented by pixels of various resolutions. When such multiple-resolution samples are potentially overlapping, it may be inefficient to render both a lower-resolution sample and also the plurality of higher-resolution samples that overlap the lower-resolution sample. Conversely, it may be more efficient to determine, before rendering a sample representing a particular data unit, whether the sample is overlapped by available higher-resolution samples that represent the same data unit. Techniques are disclosed for making this determination in an efficient manner, and for acting upon the determination to achieve improved efficiency in the rendering of the data set.

Rendering of data sets comprising multiple-resolution samples

An online system stores information identifying interactions by online system users with content provided by a third party system. The third party system includes a tracking mechanism in content that specifies one or more interactions with the content. When a user performs a specified interaction with the content via a client device, the tracking mechanism communicates information describing the performed interaction from the client device to the online system. The online system retrieves information identifying the user from the client device and stores the information describing the performed interaction in association with the information identifying the user.

Maintaining information describing interactions performed by users of an online system on third party systems on the online system

An online system enables developers of third-party applications to specify online system users to receive advertisements. The online system generates and sends a user-specific identifier for online system users that use a third-party application when an advertiser associated with the third-party application requests a user-specific identifier from the online system. For example, the online system retrieves user identifying information associated with a client device identifier provided by an advertiser and generates a user identifier by encrypting the retrieved user identifying information. The advertiser may include one or more user-specific identifiers in an advertisement request sent to an online system to present an advertisement in the advertisement request to online system users corresponding to the user-specific identifiers. The online system determines whether to present the advertisement to users based in part on whether users correspond to user-specific identifiers included in the advertisement request.

Targeting advertisements to customized groups of users of an online system

A dynamic thumbnail corresponding to an application window is displayed that is capable of being modified based on modifications in the application window. The modifications of the dynamic thumbnail may be accomplished in real-time. In some embodiments, a dynamic thumbnail is displayed in response to a user hovering a cursor over a taskbar button for a corresponding application window. In some embodiments, modifications to the dynamic thumbnail are reflected in the corresponding application window.

Manipulating thumbnails of windows

Identifying software application events

Providing context to a target minimizes the amount of information that a user must input. Context transfer pages receive context and reformat for the target. Selection of links to such pages provide context which is then reformatted and provided to the target to pre-populate information for the user. A return link can be specified to enable the target to return further context upon user interaction completion. The return link can specify further context transfer pages which can use the returned context to direct the performance of convenience actions, including invoking other applications on the user's computing device and entering information into them. The context transfer pages can themselves collect information from the user to provide appropriate interfaces without requiring resource investment from the target.

Automated target specific format conversion of context information from a user query

Described is a technology by which code, such as an untrusted web application hosted in a browser, provides content through an interface for playback by an application environment, such as an application environment running in a browser plug-in. Content may be in the form of elementary video, audio and/or script streams. The content is in a container that is unpackaged by the application code, whereby the content may be packaged in any format that the application understands, and/or or come from any source from which the application can download the container. An application environment component such as a platform-level media element receives information from an application that informs the application environment that the application is to provide media stream data for playback. The application environment requests media stream data (e.g., samples) from the application, receives them as processed by the application, and provides the requested media stream data for playback.

Media streams from containers processed by hosted code

Various technologies and techniques are disclosed for controlling a shader declaratively. A declarative programming model enables declarative statements to be used that control an instantiation of a shader in a declarative presentation framework. A shader-based effect that was specified declaratively is instantiated programmatically for rendering graphic effects for a software application. Declarative statements are sent to a shader processor to enable a shader to render the graphical effect customizations for the software application. A method is also described for utilizing a declarative programming model as a multiple pass effect that controls and encapsulates a set of shaders to be run during a multiple pass operation.

Shader-based extensions for a declarative presentation framework

Browser application programs may be prevented from performing cross-domain retrieval of anything other than image files. To enable browsers to pre-fetch resources utilizable by code executing in a browser-based programmatic environment, without having to wait for the environment, and the code, to be instantiated and make the cross-domain request themselves, the cross-domain resources can be transcoded into image files. Once retrieved, these 'image' files can be retained locally in a browser cache and provided to the code executing in the browser-based programmatic environment from there much more quickly. Then the code can decode the 'image' and extract the original resource file, which the code can then utilize. The transcoding of resources into image files can be performed by first determining the height and width of an image file that will result in the least amount of wasted space when the resource is transcoded into it.

Cross-domain browser pre-fetching through data transcoding

The claimed subject matter relates to an architecture that can construct a map for summarizing analyses with respect to data included in a database. In addition, the architecture can display the map in a special canvas area. Generally, the map is a hub-and-spoke-style map in which a hub is associated with an entity (e.g., a related set of records) included in a database. Likewise, the spokes are typically representative of operations (e.g., filter, join, transform) that act upon the hub/entity from which it extends. The map can aid with open-ended analysis on complex databases by recording and ordering competing hypotheses and can also further collaborative efforts with respect to analysis.

Canvas approach for analytics

A computing system translates two-dimensional (2D) graphical input by a user who is selecting one or more 2D images in a three-dimensional (3D) scene of 3D models created by model 3D objects. The computing system comprises a viewport module, a retrieve module, set ray module and, a hit detection module. The viewport module defines 2D boundaries of the 3D scene and a view point location in 3D space for viewing the 3D scene. The retrieve module retrievies a selection point location in 2D space for the 2D graphical input. The set ray module sets a pick ray in 3D space based on the view point location and the selection point location. The hit detection module detects a hit by the pick ray on a 3D model in the 3D scene. In this computing system there is a method of processing a hierarchy of computer program visual objects for detecting a hit by 2D input on 2D and 3D images displayed by a computing system. The method begins by traversing branches of a first tree hierarchy of visual objects to leaf objects. Next the method detects whether the next unprocessed leaf object is a visual 2D object with 2D geometry or a model 3D object with 3D geometry. Lastly the method calls a 2D process to detect a hit on the 2D geometry of a visual object if a visual 2D object is detected and calls the 3D process to detect a hit on the 3D geometry of a visual object if a model 3D object is detected.

Translating two-dimensional user input on three-dimensional scene

Growing community of inventors

Seattle, WA, United States of America

Greg D Schechter