> Markdown version of [/videos/1534-introducing-json-structure-a-better-schema?t=281](https://www.wearedevelopers.com/videos/1534-introducing-json-structure-a-better-schema?t=281). Every page supports `.md` or `Accept: text/markdown`. Links point to the HTML versions so they work for humans too. Agent guide: [/agents.md](https://www.wearedevelopers.com/agents.md).

---

# Introducing JSON Structure - A Better Schema

Tired of JSON Schema's conditional traps and overwhelming complexity? Discover JSON Structure, a newly proposed IETF standard designed strictly for type-safe pipelines and predictable code generation.

- **Speakers:** [Clemens Vasters](https://www.wearedevelopers.com/@clemens-vasters)
- **Event:** World Congress 2025
- **Published:** August 20, 2025
- **Duration:** 20:29
- **URL:** https://www.wearedevelopers.com/videos/1534-introducing-json-structure-a-better-schema

## Summary

Created out of developer frustration with JSON Schema's overwhelming complexity and validation-first approach, JSON Structure is a newly proposed IETF standard designed strictly as a type definition language. Built by the team behind Azure messaging services and CloudEvents, it directly addresses the friction of creating type-safe pipelines and generating predictable code across polyglot environments by moving away from conditional traps like allOf or anyOf.

JSON Structure core introduces explicit naming and true primitive types, natively supporting formats like decimals, dates, and 64-bit integers. Recognizing standard JSON's 53-bit number limit, it enforces string encoding for large integers to prevent data loss. It also replaces clumsy object workarounds with native compound types, including maps and highly efficient tuples. By stripping field names off the wire, tuples function smoothly as flat arrays, making them ideal for high-volume time-series streaming. Additionally, discriminated unions are cleanly formalized via the choice keyword, eliminating the fragmented inheritance hacks commonly required by OpenAPI generators.

To prevent schema spaghetti across large architectures, the standard strictly separates local references from modular external loading. The explicit import system and formal namespaces map directly into programming paradigms like Java classes or C# structures. Through separate companion specifications, developers can optionally layer on complex validation logic, internationalized serialization targets, and critical metadata like scientific units or currency symbols. By embedding definitive semantic units into the schema, JSON Structure ensures that AI models and downstream analysts understand precisely what values represent, significantly elevating event-driven data quality.

**Keywords:** json structure standard, type definition schemas, json schema alternatives, type-safe messaging pipelines, cross-language code generation, time-series data streaming, schema module resolution, IETF draft specification, discriminated union types, namespace mapping, tuple data encoding, scientific unit metadata, AI data constraints, API schema standardization

## Chapters

1. **Standardizing messaging metadata for artificial intelligence and cloud** (00:05) — A dedicated team at Azure is improving metadata standards to enable artificial intelligence safely in messaging pipelines.
1. **Understanding the complexity and traps of JSON schema** (01:45) — The inherent complexities of conditional validation make JSON schema difficult to use as a pure type definition tool.
1. **Submitting JSON structure as an internet engineering draft** (02:48) — Refactoring JSON schema into six separate specifications creates a dedicated type definition language to solve API dependency issues.
1. **Distinguishing type definition constructs from data validation routines** (04:41) — New primitive types like integers and precise dates enforce proper JSON serialization beyond standard numeric limits.
1. **Adding maps and tuples for efficient telemetry streaming** (06:45) — Native compound types optimize repetitive time series data by removing redundant field names entirely from the wire.
1. **Establishing formal discriminated unions to simplify type selection** (08:11) — Native choice structures replace conditional workarounds by formally defining discrete data type intersections and component hierarchies.
1. **Enforcing internal definition references and explicit code namespaces** (09:18) — Restricting object boundaries prevents unmanageable code while aligning cleanly with object-oriented programming constructs like classes.
1. **Separating object imports securely from local definition references** (11:39) — Implementing explicit cross-file inclusions limits chaotic file graphs by demanding strict resolution against optionally imported modules.
1. **Replacing ambiguous compositions with explicit object extension properties** (12:56) — Abstract components and clear extension identifiers prevent arbitrary structure permutations from confusing language code generators.
1. **Incorporating internationalization conventions via alternative field naming arrays** (15:01) — Extensible naming parameters allow developers to map localized strings and custom document encodings directly within the schema.
1. **Embedding scientific units and currencies into schema models** (16:15) — Native standard notations for metrics ensure downstream interpreters and automated systems perceive measurement values accurately.
1. **Restoring complex condition validation through separate optional specifications** (17:50) — Isolating conditionals and numeric limits into optional addons preserves base definitions while enabling full legacy functionality.
1. **Exploring the draft schema documentation via AI tools** (19:00) — Open collaboration and automated specification indexers invite developer feedback for iterating on the internet standard draft.

## Related Moments

- [Adopting JSON Structure as a strict data definition language](https://www.wearedevelopers.com/videos/1616-bringing-clarity-to-event-streams-enabling-analytics-and-ai-through-rich-metadata) (from "Bringing Clarity to Event Streams: Enabling Analytics and AI Through Rich Metadata")
- [Refactoring JSON Schema into a formal data definition language](https://www.wearedevelopers.com/videos/100219-introducing-json-structure) (from "Introducing JSON Structure")
- [Creating a standardized JSON schema for framework-agnostic discoverability](https://www.wearedevelopers.com/videos/1012-micro-frontends-discovery) (from "Micro-Frontends Discovery")
- [Structuring model payloads precisely with custom JSON schemas](https://www.wearedevelopers.com/videos/1269-exploring-google-gemini-and-generative-ai) (from "Exploring Google Gemini and Generative AI")
- [Leveraging the JSON API standard for data structure](https://www.wearedevelopers.com/videos/33-api-design-getting-started) (from "API Design - Getting Started")
- [Enforcing structured responses using native JSON schema configurations](https://www.wearedevelopers.com/videos/100157-ai-in-your-browser-exploring-chrome-s-built-in-llm) (from "AI in Your Browser: Exploring Chrome's Built-In LLM")

## Related Articles

- [Douglas Crockford - JSON and Beyond](https://www.wearedevelopers.com/magazine/481-douglas-crockford-json-and-beyond)
- [Dev Digest 136 - No JS(on) of mine](https://www.wearedevelopers.com/magazine/482-dev-digest-136-no-js-on-of-mine)
- [The Overflow: 5 Tools for Developers Working with JSON](https://www.wearedevelopers.com/magazine/706-the-overflow-5-tools-for-developers-working-with-json)
- [Tips and Tricks: Working with JSON](https://www.wearedevelopers.com/magazine/493-tips-and-tricks-working-with-json)

## Related Jobs

- [Agile Full Stack Engineer](https://www.wearedevelopers.com/jobs/48325-agile-full-stack-engineer) at **synava GmbH**
- [AI Software Engineer (Germany)](https://www.wearedevelopers.com/jobs/48317-ai-software-engineer-germany) at **Sunhat**
- [Senior Software Engineer](https://www.wearedevelopers.com/jobs/ext/15942-senior-software-engineer) at **GitHub**
- [Staff Software Engineer](https://www.wearedevelopers.com/jobs/ext/1425755-staff-software-engineer) at **GitHub**
- [Security Architect - AI](https://www.wearedevelopers.com/jobs/ext/1581899-security-architect-ai) at **ZEISS Group**
- [Lead Software Engineer - Data Engineering](https://www.wearedevelopers.com/jobs/ext/2000968-lead-software-engineer-data-engineering) at **Dynatrace**