termEngineering & architecture

HTTP

The application-level request-response protocol that defines shared semantics for transferring representations and operating web systems.

Download Markdown (.md)

01What it means

HTTP, the Hypertext Transfer Protocol, is a stateless application-level protocol in which clients send requests and servers return responses. It defines methods, status codes, header fields, message content, and intermediary behavior; HTTPS is HTTP protected by TLS rather than a separate set of application semantics.[1][2]

Stateless request and response protocol A client addresses a resource; a server returns a described outcome
ClientIntermediaryServer
  1. 01Client IntermediaryRequestMethod, target, headers, optional content
  2. 02Intermediary ServerForward or satisfyApply routing and cache semantics
  3. 03Server IntermediaryResponseStatus, headers, optional content
  4. 04Intermediary ClientDeliver resultMetadata guides handling and reuse

Proxies, gateways, and caches may handle the message. HTTPS protects the same HTTP semantics with TLS.

02Why it matters

Browsers, APIs, proxies, caches, gateways, and content delivery networks rely on HTTP's shared vocabulary. Correct use affects interoperability, security, latency, caching, retries, observability, and what users experience during failures.

03When to use it

Use HTTP for web pages, resource-oriented APIs, file delivery, and other interactions that benefit from its broad tooling and intermediary support. Persistent low-latency bidirectional exchange or specialized device and messaging workloads may require another protocol or an HTTP-based upgrade.

04How it works

  1. A client identifies a target resource and creates a request with a method, target, headers, and optional content.
  2. Connections and protocol versions transport the message, potentially through proxies, gateways, or caches.
  3. The server interprets the request according to method and header semantics and processes it.
  4. The server returns a response with a status code, headers, and optional content.
  5. The client and intermediaries use metadata such as cache controls, validators, and content type to handle the result.

05A practical example

Example

A browser requests a product page with GET over HTTPS. A CDN reuses a fresh cached response; when it becomes stale, the CDN validates it with the origin and can receive 304 Not Modified instead of downloading the full representation again.

06Where it breaks

HTTP semantics do not provide application authorization, trustworthy payloads, or business correctness by themselves. Stateless requests can repeat context, network calls can partially fail, and HTTP version or transport improvements do not fix inefficient application behavior.

Common mistakes

  • Treating HTTP and HTTPS as unrelated application protocols rather than HTTP with transport security.
  • Ignoring standardized method, status, cache, and header semantics in application design.
  • Assuming a successful connection means the business operation succeeded.
  • Sending sensitive information without considering URLs, logs, caches, and intermediary visibility.

07Sources

  1. HTTP Semantics (RFC 9110)RFC Editor · Internet StandardHTTP Semantics (RFC 9110)
  2. Overview of HTTPMDN Web Docs · Official technical documentationOverview of HTTP
View saved topics

Start typing to search titles, definitions, categories, and frameworks.