Abstract: Described herein is a system and method for coded streaming data to facilitate recovery from failed or slow processor(s). A batch of processing stream data can be partitioned into a plurality of data chunks. Parity chunk(s) for the plurality of data chunks. The plurality of data chunks and the parity chunk(s) can be provided to processors for processing. Processed data of at least some (e.g., one or more) of the plurality of data chunks, and, processed data of parity chunk(s) are received. When it is determined that processed data for a pre-defined quantity of data chunks has not been received by a pre-defined period of time, the processed data for particular data chunk(s) of particular processor(s) from which processed data has not been received are determined based, at least in part, upon the received processed parity chunk(s) and the received processed data chunk(s).

Abstract: A computing device including memory storing a quantum computing device model. The quantum computing device model may include a plurality of quantum computing device components having a respective plurality of actual boundaries. The computing device may further include a processor configured to generate a first discretized model of the quantum computing device model. The first discretized model may indicate a respective estimated boundary for each quantum computing device component. The processor may be further configured to solve a first differential equation discretized with the first discretized model. The processor may be further configured to generate a second discretized model of a focus region of the quantum computing device model. In the second discretized model, the focus region may have the estimated boundary computed for the focus region in the first discretized model. The processor may be further configured to solve a second differential equation discretized with the second discretized model.

Abstract: Email has become commonplace as a means to communicate among parties non-synchronously by exchanging email messages. Some communications, however, may be more effective in real-time or in near real-time. For instance, service providers may offer a chat feature to accommodate real-time or near real-time help or service communications with customers. However, in this case, the user is likely required to enter credentials and/or context information to transition into a different communication type with the service provider. The present systems and methods provide a continuous conversation experience for the user whereby a message recipient of one message type (e.g., email, text, etc.) may continue a conversation in a different message type (e.g., a chat session, instant message, etc.) without manually providing message context for communicating via the second message type.

Abstract: A method for distributing a shareable video experience among plural video-consumer devices includes receiving a data stream defining the shareable video experience and generating first and second shared video experiences. The first experience is configured for video-consumer devices capable of real-time interaction with the shareable video experience; it includes a first video stream conveying visual content of the shareable video experience and a first set of interface controls, the first set of interface controls being useable by video-consumer devices capable of real-time interaction to modify the shareable video experience. The second experience is configured for video-consumer devices incapable of real-time interaction; it includes a second video stream conveying visual content of the shareable video experience and a second set of interface controls that are useable by video-consumer devices incapable of real-time interaction to present supplementary data in connection to the shareable video experience.

Abstract: In some examples, dynamic responsive form creation may include obtaining, in a live environment and for a form, an indication of a modification to a field of the form. A modification to a schema may be determined in the live environment and based on the indication of the modification to the field of the form. A transformed schema may be generated based on transformation, in the live environment, of the modification to the schema. The transformed schema may be in a format that is usable to modify the form. An input associated with the field may be obtained in the live environment and based on a modified form generated from the transformed schema. A determination may be made, in the live environment, as to whether the input is valid by analyzing a validation associated with the field.

Abstract: Described herein is a system and method of journaling of a streaming anchor resource. An input node can store a value of a property associated with the streaming data in a persistent indexed data structure. The input node can generate an anchor that describes a particular point in time in a data stream. The anchor can include an index into the persistent indexed data structure of the stored value of the property associated with the streaming data. The generated anchor and streaming data can be provided to the downstream node. During recovery of a downstream node, the input node can utilize a received anchor to retrieve a value of a property associated with the streaming data from the persistent indexed data structure, and, provide a batch of data based upon the received anchor and the retrieved property value.

Abstract: Presenting and manipulating task items includes creating a task item within a productivity application in association with a content item created with the productivity application and assigning the task item to one or more users. When the task item has been assigned, the task item is displayed to each user when each users opens the productivity application, or when the productivity application is already open to the user. User activity is then monitored in relation to the content item that is associated with the task item such that metadata for the user activity is captured. Based on the captured metadata, the users are alerted to the user activity that has occurred on the content item.

Abstract: Examples are disclosed that relate to providing transition-related assistance during a presentation. One example provides a method comprising, during a rehearsal of a presentation, receiving content of the presentation. Based on the content received, a transition within the presentation is determined via a machine learning model. During a performance of the presentation, the transition is automatically enacted.

Abstract: A media session is established between a first device (e.g., endpoint) and a second device (e.g., endpoint) via a communication network based on connectivity checks performed by the devices. A set of candidate pairs, including network addresses available to the first device and the second device, is generated. Among the candidate pairs of the candidate pair set, a respective priority is initially assigned for connectivity checks. If a connectivity check modification condition is determined to be met for candidate pairs used for connectivity checks, the priority ordering of the candidate pairs is modified to produce a second priority ordering. This second priority ordering assigns a higher priority for a candidate pair which meets the connectivity check modification condition than a candidate pair that does not meet the condition. Based on the modified ordering, connectivity checks are performed, and the media session is configured and established using a candidate pair determined to be valid.

Abstract: Innovations in use of chroma quantization parameter (“QP”) offsets when determining a control parameter for deblock filtering. For example, as part of encoding, an encoder sets a picture-level chroma QP offset and slice-level chroma QP offset for encoding of a slice of a picture. The encoder also performs deblock filtering of at least part of the slice, where derivation of a control parameter considers only the picture-level chroma QP offset. The encoder outputs at least part of a bitstream including the encoded content. As part of decoding, a corresponding decoder sets a picture-level chroma QP offset and a slice-level chroma QP offset for decoding of a slice of a picture, but derivation of a control parameter for deblock filtering considers only the picture-level chroma QP offset.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for maintaining, by a policy enforcement system in a first compute node, a plurality of policies and data associating a plurality of user credentials with the plurality of policies. A request is received from a compute process for data from a file system in the first compute node. The request includes user credentials. The request for data is sent to the file system, and the data is received from the file system. Based on the received user credentials and the data associating the plurality of user credentials with the plurality of policies, one or more policies that correspond to the received user credentials is selected from the plurality of policies. The policy enforcement system filters the data from the file system based on the one or more policies, and sends the filtered data to the compute process.

Abstract: In some implementations, an insect trap includes a plurality of individual cells, with each of the cells being configured to trap very few, and in some instances a single flying insect. Each cell defines a cavity and may be provided with an independently operable door to selectively close the cavity. One or more sensors may sense the presence of an insect to be trapped.

Abstract: An optical filter for a camera is switchable between a reflection state and a transmission state. The optical filter includes a first plurality of liquid crystals configured to dynamically form cholesteric phase structures in the reflection state that block right-handed circularly polarized light in a spectral light sub-band and transmit light outside of the spectral light sub-band. The first plurality of liquid crystals dynamically forms a nematic phase arrangement in the transmission state that transmits light in the spectral light sub-band. The optical filter further includes a second plurality of liquid crystals configured to dynamically form cholesteric phase structures in the reflection state that block left-handed circularly polarized light in the spectral light sub-band and transmit light outside of the spectral light sub-band. The second plurality of liquid crystals dynamically form a nematic phase arrangement in the transmission state that transmits light in the spectral light sub-band.

Abstract: Techniques of implementing software filtered non-volatile memory in a computing device are disclosed herein. In one embodiment, a method includes detecting an entry being written to a guest admin submission queue (gASQ) by a memory driver of a virtual machine hosted on the computing device. Upon detecting the entry written to the gASQ by the memory driver, the command in the entry is analyzed to determine whether the command is allowed based on a list of allowed or disallowed commands. In response to determining that the command in the entry is not allowed, without sending the command to the non-volatile memory, generating an execution result of the command in response to the entry being written to the gASQ by the memory driver. As such, potentially harmful commands from the memory driver are prevented from being executed by the non-volatile memory.

Abstract: Computing environments are scalably generated and provisioned based on metadata associated with separately defined containers of groups of sandboxes. Each container of a group of sandboxes is its own digital construct and is wrapped with metadata that defines the nature and aspects of the grouping, including enumerating individual members of a team, factors or attributes of assignments, and the like. The metadata, therefore, identifies the sandboxes to generate and, within those sandboxes, the computing environments to generate and how such computing environments are to be provisioned. In variable team settings quantities and identities of individuals can be unknown and can dynamically change. Administrators provide team definitions without rosters, or with rosters that comprise only previously registered individuals.

Abstract: Systems and methods are provided for controlling the position of an interactive movable menu in a mixed-reality environment. In some instances, a mixed-reality display device presents a mixed-reality environment to a user. The mixed-reality device then detects a first gesture associated with a user controller while presenting the mixed-reality environment and, in response to the first gesture, triggers a display of an interactive movable menu within the mixed-reality environment as a tethered hologram that is dynamically moved within the mixed-reality environment relative to and corresponding with movement of the user controller within the mixed-reality environment. Then, in response to a second detected gesture, the mixed-reality device selectively locks a display of the interactive movable menu at a fixed position that is not tethered to the user controller.

Abstract: The disclosed technology is generally directed to IoT technology. In one example of the technology, the following actions are performed for each module of a plurality of modules on a first edge device. An identification message that includes information associated with identification of the module is received. The validity of the module is then verified. After the module is verified, based at least in part on the identification message, an IoT support service is selected from a plurality of IoT support services. The module is then caused to be registered with the selected IoT support service. The plurality of modules are compositable together into an application for the first edge device. The modules of the plurality of modules are capable of being used interoperably with other modules without altering the other modules.

Abstract: An Artificial Intelligence (AI) driven content generation system is configured to interact with and inject user inputs into the decision-making process to navigate the user through the boundless creative space offered by content generation algorithms powered by AI. A content generation application may generate an initial collection of artifacts (e.g., multimedia) for exposure to a user, which may be generated based on user input, context awareness, or randomly generated. Responsive to user selection of one of the initial artifacts, a subsequent collection of artifacts is generated based on characteristics of the selected artifact. The generative process is recursive in that the generative application and the user continue to interact until the user decides to use one of the generated artifacts. The generative process enables the user to navigate the virtually infinite creative space of artifacts which may be generated by AI algorithms.

Abstract: Correcting data in a dataset. A set of data tokens from a tabular data store are grouped into a plurality of different clusters based on similarity of tokens. A reference cluster is selected from among the plurality of different clusters such that the plurality of clusters includes a reference cluster and one or more other clusters. One or more tokens in the one or more other clusters are transformed. The effect on the reference cluster of adding the transformed tokens to the reference cluster is determined. Using this information, a correction for a token in the dataset is identified. The data store is updated to correct the token using the identified correction.

Abstract: Techniques for application launching in a multi-display device are described. In one or more implementations, an apparatus such as a mobile device includes multiple interconnected display devices. According to one or more implementations, techniques described herein enable application launching behavior to be determined based on context information. For instance, based on a determined context condition of a multi-display client device, an application launch behavior is determined and used to launch an application on the client device.

Abstract: Data services are often provided with consistency guarantees of either strong consistency models, comprising uniform wall-clock consistency, or eventual consistency models, where temporary logical inconsistency is guaranteed to be resolved only after full data propagation. However, the performance characteristics of contemporary services often require an intermediate consistency model, where some aspects of the service have specific consistency expectations and other aspects of the service are flexible, such as bounded staleness (e.g., a maximum delay in reaching consistency); session consistency (e.g., individual sessions remain logically consistent, but ordering may vary across sessions); and prefix consistency (e.g., each view during a session is logically consistent, but ordering may vary between session views).

Abstract: A low-pressure hinge includes a plurality of segments with at least one upper frictional element and at least one lower frictional element where the at least one upper frictional element and the at least one lower frictional element are positioned on opposing sides of a segment of the plurality of segments. The hinge further includes a compression element that compresses the at least one upper frictional element and at least one lower frictional element against the segment with a compression force having a pressure less than 5,000 pounds per square inch.

Abstract: A server computing device, including memory storing a knowledge graph. The server computing device may further include a processor configured to receive a natural language input and generate a tokenized utterance based on the natural language input. Based on the tokenized utterance, the processor may select a predefined intention indicating a target ontology entity type of the natural language input. The processor may identify at least one input ontology entity token included in the tokenized utterance and may identify at least one relation between the predefined intention and the input ontology entity token. Based on the predefined intention, the at least one input ontology entity token, and the relation, the processor may generate a structured query. Based on the structured query and the knowledge graph, the processor may output an output ontology entity token having the target ontology entity type.

Abstract: Provisioning a cloud based high performance computing cluster. The method includes from a cloud based provisioning service deployed in a cloud based computing provider, providing a user interface. The method further includes receiving user input at the cloud based provisioning service from the user interface. The method further includes from the user input, determining a configuration for a cloud based user service for the user. Using the cloud based provisioning service, the method further includes provisioning the cloud based user service for the user, in the cloud, based on the determined configuration.

Abstract: On a display configured to provide a photorepresentative view from a user's vantage point of a physical environment in which the user is located, a method is provided comprising receiving, from the user, an input selecting a theme for use in augmenting the photorepresentative view. The method further includes obtaining, optically and in real time, environment information of the physical environment and generating a spatial model of the physical environment based on the environment information. The method further includes identifying, via analysis of the spatial model, one or more features within the spatial model that each corresponds to one or more physical features in the physical environment. The method further includes based on such analysis, displaying, on the display, an augmentation of an identified feature, the augmentation being associated with the theme.

Abstract: A waveguide display includes multiple diffractive optical elements (DOEs) that are configured to in-couple image light, provide expanded exit pupil in two directions, and out-couple the image light to a user. An in-coupling DOE is configured to split the full field of view (FOV) of the image light into left and right portions. The left and right FOV portions are respectively propagated laterally in left and right directions in intermediate DOEs which comprise upper and lower portions. The intermediate DOEs provide for exit pupil expansion in a horizontal direction while coupling light to an out-coupling DOE. The out-coupling DOE provides for exit pupil expansion in a vertical direction and out-couples image light with expanded exit pupil for the full FOV. The intermediate DOE portions are configured to steer image light back towards the center of the waveguide to avoid dark areas or stripes in portions of the out-coupling DOE.

Abstract: The use of user-specific data to process a biometric print, such that use of the biometric print is revoked by invalidating the user-specific data. The processed print is generated by performing one-way processing of the biometric print using the user-specific data. The processed print, not the biometric print, is then provided to the authentication system for later authentication of the user. During matching, the user later provides a current biometric, resulting in generation of a current biometric print. For each of multiple users, the user-specific is obtained for that user, and at least one processed print is generated for each user based on the current biometric print. The current processed prints are used by the authentication system to match against each of the enrolled processed prints. If a match is found, the user is identified as being the user associated with the matching enrolled print.

Abstract: This document describes techniques for application reporting in an application-selectable user interface. These techniques permit a user to view reports for applications in a user interface through which these applications may be selected. By so doing, a user may quickly and easily determine which applications to select based on their respective reports and then select them or their content through the user interface.

Abstract: A proxy-based thin-client web browsing framework enables cooperative web browsing of multiple devices. The multiple devices may include devices that are not intended for web browsing and have limited or no web browsers and/or user input capabilities. The proxy-based thin client web browsing framework employs a virtual browser at a proxy server to perform all browser-engine logics, and retrieve, render and encode web pages on behalf of the multiple devices. The multiple devices therefore only need to have limited decoding and display capabilities to perform web browsing. The proxy-based thin client web browsing framework further includes a touch controller as a remote controller for a device that has no or limited user texting or manipulating capabilities.

Abstract: Systems, methods, and software are disclosed herein for triaging test results of a failed code validation. Test results are received indicative of a failed code validation performed in a client device. Failure data is generated from the test results indicating one or more failure modes which occurred during the failed code validation. A follow-up action associated with each of the one or more failure modes is then determined by comparing each of the one or more failure modes to previously identified failure modes. Instructions are communicated to the client device based on the determined follow-up action associated with each of the one or more failure modes.

Abstract: Parallel tasks are created, and the tasks include a first task and a second task. Each task resolves a future. At least one of three possible continuations for each of the tasks is supplied. The three continuations include a success continuation, a cancellation continuation, and a failure continuation. A value is returned as the future of the first task upon a success continuation for the first task. The value from the first task is used in the second task to compute a second future. The cancellation continuation is supplied if the task is cancelled and the failure continuation is supplied if the task does not return a value and the task is not cancelled.

Abstract: An eye movement tracking device that can be mounted to standard eyeglasses as disclosed. The device comprises an illumination source, a time-of-flight (TOF) camera and a processor. The source transmits energy within a frequency band from a location proximate to an eye of a person, such that at least a first portion of the transmitted energy is reflected off a lens of eyewear worn by the person to subsequently reflect off the eye, and such that at least a second portion of the transmitted energy is transmitted through the lens to reflect off objects in the person's environment. The TOF camera detects reflections of at least the first portion of the transmitted energy, and distinguishes them from other energy detected by the TOF camera in said frequency band, based on TOF principles. The processor uses the detected reflections of the first portion of the transmitted energy to determine eye position.

Abstract: A service may be provided that reads sensors, and that communicates information based on the sensor readings to applications. In one example, an operating system provides a sensor interface that allows programs that run on a machine to read the values of sensors (such as an accelerometer, light meter, etc.). A service may use the interface to read the value of sensors, and may receive subscriptions to sensor values from other programs. The service may then generate messages that contain the sensor value, and may provide these messages to programs that have subscribed to the messages. The messages may contain raw sensor data. Or, the messages may contain information that is derived from the sensor data and/or from other data.

Abstract: Databases are often provided according to various organizational models (e.g., document-oriented storage, key/value stores, and relational database), and are accessed through various access models (e.g., SQL, XPath, and schemaless queries). As data is shared across sources and applications, the dependency of a data service upon a particular organizational and/or access models may become confining. Instead, data services may store data in a base representation format, such as an atom-record-sequence model. New data received in a native item format may be converted into the base representation format for storage, and converted into a requested format to fulfill data requests. Queries may be translated from a native query format into a base query format that is applicable to the base representation format of the data set, e.g., via translation into an query intermediate language (such as JavaScript) and compilation into opcodes that are executed by a virtual machine within the database engine.

Abstract: Incident information that describes a timing of an incident in a service, is received from a service. User activity information is received from a client system in a tenant. An impact of an incident, on a tenant, is identified by combining the incident information with the user activity information.

Abstract: The management of applications in a computing environment often involves a significant number of computing operations, such as acquiring the application, deploying data objects in various locations, and configuring the application with respect to the deployed device. Conventional computing environments may provide inadequate support throughout a typical application life cycle, especially for a computing environment distributed across many devices. Instead, the computing environment may be represented in a deployable object hierarchy that may be distributed to various devices. An application management service may therefore be provided to install representations of the application in the object hierarchy, and to support many other aspects of the application life cycle, such as application discovery, sharing, updating, and removal.

Abstract: Computing environments can be scalably generated and provisioned based on one or more team definitions and definitions of one or more projects assigned to a particular team. Digital constructs can be generated to represent the teams and the projects assigned to such teams, and such digital constructs can comprise metadata identifying sandboxes to generate and, within those sandboxes, computing environments to generate and how such computing environments are to be provisioned. The sandboxes can be generated in parallel and the computing environments can be provisioned in parallel, or staggered based on individual user access. Assignment of consumption limitations can be individually adjusted from default values, and monitoring of such consumption limitations can be performed on an aggregated basis.

Abstract: In an embodiment, the disclosed technologies include receiving a set of digital inputs; where the set of digital inputs includes a candidate entity-member entity pair that includes candidate entity data and member entity data; where the member entity data has been extracted from a node of an online service; where an exact match has not been found between the candidate entity data and the member entity data; in response to the set of digital inputs, outputting models of the candidate entity data and the member entity data, respectively; where the models indicate weight values assigned to text in the candidate entity data and weight values assigned to text in the member entity data, respectively; calculating a similarity score using the models; in response to the similarity score matching a threshold, inputting the candidate entity-member entity pair to a classifier to produce a classification; where the classifier uses a machine learning model that has been trained using features derived from previously-analyz

Abstract: In an example, one or more member profiles and corresponding elapsed times indicating, for each of the one or more member profiles, how long the corresponding member of a social networking service took to respond to a request for confidential data with a submission of confidential data are obtained. Then a first set of one or more features are extracted from the one or more member profiles. The first set of one or more features and corresponding elapsed times are fed into a machine learning algorithm to train a confidential data response time prediction model to output a predicted time to respond to a request for confidential data for a candidate member profile. A second set of one or more features are obtained from a candidate member profile and fed to the confidential data response time prediction model, outputting the predicted time to respond to a request for confidential data.

Abstract: Examples are disclosed that relate to reading stored data. The method comprises obtaining a representation of a measurement performed on a data-storage medium, the representation being based on a previously recorded pattern of data encoded in the data-storage medium in a layout that defines a plurality of data locations. The method further comprises inputting the representation into a data decoder comprising a trained machine-learning function, and obtaining from the data decoder, for each data location of the layout, a plurality of probability values, wherein each probability value is associated with a corresponding data value and represents the probability that the corresponding data value matches the actual data value in the previously recorded pattern of data at a same location in the layout.

Abstract: Methods, systems, and devices are described herein for delivering protected data to a trusted execution environment (TrEE) associated with a potentially untrusted requestor. In one aspect, a targeting protocol head may receive a request for protected data from a potentially untrusted requestor associated with a TrEE, and an attestation statement of the TrEE. The targeting protocol head may retrieve the protected data, and obtain a targeting key of the TrEE from, for example, the request in the case of clean room provisioning, or the attestation statement. The targeting protocol head may generate targeted protected data by encrypting the protected data with the targeting key, and provide the targeted protected data to the potentially untrusted requestor, where a private targeting key of the TrEE is required to decrypt the targeted protected data.

Abstract: A solution for formatting document objects by visual suggestions is proposed. For a document including multiple editing units each of which contains multiple objects, a first editing rule is generated, the first editing rule indicating an edit performed on a first object in a document. Then, an editing suggestion is generated based on the first editing rule, the editing suggestion describing a candidate edit to be performed on a second object within a suggestion scope. The editing suggestion is then presented in a visual form to a user for selection.

Abstract: The techniques disclosed herein improve the efficiency of a system by providing intelligent agents for managing data associated with objects that are displayed within mixed-reality and virtual-reality collaboration environments. Individual agents are configured to collect, analyze, and store data associated with individual objects in a shared view. The agents can identify real-world objects and virtual objects discussed in a meeting, collect information about each object and generate recommendations for each object based on the collected information. The recommendations can suggest modifications to the objects, provide resources for obtaining or modifying the objects, and provide actionable information allowing users to reach a consensus regarding an object. The data can be shared between different communication sessions without requiring users to manually store and present a collection of content for each object.

Abstract: A hybrid configuration engine and associated method for reducing the complexity and burden of configuring rich coexistence between an on-premise solution and a cloud-based solution is described herein and illustrated in the accompanying figures. The hybrid configuration engine determines the current state of the on-premise solution and the cloud-based solution and learns the desired configuration state. After obtaining the current and desired configuration state information, the hybrid configuration engine determines and automatically performs steps to reach the desired configuration state. Finally, the hybrid configuration engine provides instructions describing the manual steps needed to reach the desired configuration state.

Abstract: Fault-tolerant error correction (EC) is desirable for performing large quantum computations. In this disclosure, example fault-tolerant EC protocols are disclosed that use flag circuits, which signal when errors resulting from ? faults have weight greater than ?. Also disclosed are general constructions for these circuits (also referred to as flag qubits) for measuring arbitrary weight stabilizers. The example flag EC protocol is applicable to stabilizer codes of arbitrary distance that satisfy a set of conditions and uses fewer qubits than other schemes, such as Shor, Steane and Knill error correction. Also disclosed are examples of infinite code families that satisfy these conditions and analyze the behaviour of distance-three and -five examples numerically. Using fewer resources than Shor EC, the example flag EC protocols can be used in low-overhead fault-tolerant EC protocols using large low density parity check quantum codes.

Abstract: A content creation application with an ink augmenting feature can communicate at least one recognized term from ink strokes input to a canvas interface of the content creation application to an ink augmenting service. The content creation application can receive at least one keyword from the ink augmenting service, where each keyword indicates a corresponding recognized term having available content identified by the ink augmenting service. The content creation application then transforms a grouping of the ink strokes corresponding to the recognized term indicated by each keyword to an actionable item in the canvas interface. Selection of the actionable item invokes the command to provide the available content.

Abstract: The disclosed technology is generally directed to device authentication in an IoT environment. For example, such technology is usable in authenticating IoT devices to an IoT Hub. In one example of the technology, data field targets are received for an IoT device. The data field targets may include at least one device identity data field target and at least one telemetry data field target. Data field entries are received from the IoT device at a first time. The data field entries may include at least one device identity data field entry and at least one telemetry data field entry. A determination is made as to whether the data field entries match the corresponding data field targets for the IoT device. The IoT device is selectively allowed to connect to the IoT hub based on the determination.

Abstract: Disclosed embodiments provide for collaborative access to a distributed data structure. Operations such as inserting, deleting, and annotating are performed by various clients participating in the collaboration. A serialization service enforcing a single order of these operations and ensures the operations and order are communicated to all of the clients participating in the collaboration. Once all clients have acknowledged particular operations, this information is also communicated to all of the clients, such that the clients may perform maintenance activities on the collaboration data structures. The result is a rolling collaboration window shared by all of the clients of the collaboration, with a bottom of the collaboration window defined by a lowest ordered operation not yet acknowledged by all clients of the collaboration, and a highest ordered operation as assigned by the serialization service.