Abstract: The present disclosure is directed techniques for bootstrap management. A method includes: upon an initial launch of an application on a client device, fetching, from a server and using a native component of the application, content for loading a web component of the application on the client device; determining whether a bootstrap management mode is enabled on the client device; and responsive to the bootstrap management mode is enabled and in response to the web component being launched: receiving, at the native component and from the web component, a manifest and a request for bootstrapping resources; caching, by the native component, the manifest from the web component; fetching, from the server and using the native component, the bootstrapping resources requested by the web component; caching, by the native component, the fetched bootstrapping resources in the memory; and providing, by the native component, the fetched bootstrapping resources to the web component.

Abstract: Disclosed are methods, apparatus, systems, and computer-readable storage media for selectively interacting with a server to build a local dictation database for speech recognition at a device. In some implementations, a computing device receives an audio sample. The computing device may determine that the received audio sample does not match any of one or more existing audio samples stored in the local dictation database of the computing device. The received audio sample may be transmitted to a remote server for detection of one or more words indicated by the received audio sample. The computing device may receive data identifying the one or more words, and update the local dictation database to store the received audio sample in association with the one or more words.

Abstract: Disclosed are methods, apparatus, systems, and computer-readable storage media for selectively interacting with a server to build a local dictation database for speech recognition at a device. In some implementations, a computing device receives an audio sample. The computing device may determine that the received audio sample does not match any of one or more existing audio samples stored in the local dictation database of the computing device. The received audio sample may be transmitted to a remote server for detection of one or more words indicated by the received audio sample. The computing device may receive data identifying the one or more words, and update the local dictation database to store the received audio sample in association with the one or more words.

Abstract: A scheduler integrates worker time and location into customer appointment schedules by obtaining the status of workers' meetings with customers and forwarding updated schedule messages to waiting customers if the worker is late for an appointment due to traffic or unexpected delays at the previous customer meeting. The waiting customers receiving the updated schedules can respond by accepting the updated schedule, adjusting the meeting time or cancelling the appointment. The scheduler eliminates the need for customers to have to wait for long periods of time for service workers to arrive and allows them to adjust the meeting appointment based upon their schedules.