Feedblock

Feedblock is a term used in information technology to describe a modular processing unit within a data or content feed. A feedblock represents a discrete portion of a feed, encapsulating metadata about the block and a collection of items or records. Feedblocks are designed to be composed into pipelines that ingest, transform, filter, and deliver content to consumers.

A feedblock typically contains a header with identifiers such as block_id, source, timestamp, and a payload comprising items. Each item carries its own metadata and content fields. Rules associated with a feedblock may include filters, enrichment steps, deduplication, rate limiting, and error handling. The block-based approach supports parallel processing, backpressure management, and easier fault isolation, because failures are localized to a single block.

In deployment, feedblocks can be created on a server as part of a streaming pipeline, or assembled on the client side for offline processing. They can be stateless or stateful, depending on whether the block needs to remember previous items. Real-time feedblocks prioritize low latency, while batch feedblocks optimize throughput.

Common use cases include aggregation of news or social feeds, API data wrapping, content moderation workflows, and modular content delivery systems where different blocks can be rearranged or replaced without affecting the entire pipeline.

See also: feed, RSS, data pipeline, stream processing, content delivery network, backpressure.