Simon Willison's Weblog: standards

Open Responses

2026-01-15T23:56:56+00:00

This is the standardization effort I've most wanted in the world of LLMs: a vendor-neutral specification for the JSON API that clients can use to talk to hosted LLMs.

Open Responses aims to provide exactly that as a documented standard, derived from OpenAI's Responses API.

I was hoping for one based on their older Chat Completions API since so many other products have cloned the already, but basing it on Responses does make sense since that API was designed with the feature of more recent models - such as reasoning traces - baked into the design.

What's certainly notable is the list of launch partners. OpenRouter alone means we can expect to be able to use this protocol with almost every existing model, and Hugging Face, LM Studio, vLLM, Ollama and Vercel cover a huge portion of the common tools used to serve models.

For protocols like this I really want to see a comprehensive, language-independent conformance test site. Open Responses has a subset of that - the official repository includes src/lib/compliance-tests.ts which can be used to exercise a server implementation, and is available as a React app on the official site that can be pointed at any implementation served via CORS.

What's missing is the equivalent for clients. I plan to spin up my own client library for this in Python and I'd really like to be able to run that against a conformance suite designed to check that my client correctly handles all of the details.

Via VB

Tags: json, standards, ai, openai, generative-ai, llms, openrouter, conformance-suites

Agentic AI Foundation

2025-12-09T22:24:48+00:00

Agentic AI Foundation

Announced today as a new foundation under the parent umbrella of the Linux Foundation (see also the OpenJS Foundation, Cloud Native Computing Foundation, OpenSSF and many more).

The AAIF was started by a heavyweight group of "founding platinum members" ($350,000): AWS, Anthropic, Block, Bloomberg, Cloudflare, Google, Microsoft, and OpenAI. The stated goal is to provide "a neutral, open foundation to ensure agentic AI evolves transparently and collaboratively".

Anthropic have donated Model Context Protocol to the new foundation, OpenAI donated AGENTS.md, Block donated goose (their open source, extensible AI agent).

Personally the project I'd like to see most from an initiative like this one is a clear, community-managed specification for the OpenAI Chat Completions JSON API - or a close equivalent. There are dozens of slightly incompatible implementations of that not-quite-specification floating around already, it would be great to have a written spec accompanied by a compliance test suite.

Tags: open-source, standards, ai, openai, llms, anthropic, ai-agents, model-context-protocol

model.yaml

2025-06-21T17:15:21+00:00

model.yaml

From their GitHub repo it looks like this effort quietly launched a couple of months ago, driven by the LM Studio team. Their goal is to specify an "open standard for defining crossplatform, composable AI models".

A model can be defined using a YAML file that looks like this:

model: mistralai/mistral-small-3.2
base:
  - key: lmstudio-community/mistral-small-3.2-24b-instruct-2506-gguf
    sources:
      - type: huggingface
        user: lmstudio-community
        repo: Mistral-Small-3.2-24B-Instruct-2506-GGUF
metadataOverrides:
  domain: llm
  architectures:
    - mistral
  compatibilityTypes:
    - gguf
  paramsStrings:
    - 24B
  minMemoryUsageBytes: 14300000000
  contextLengths:
    - 4096
  vision: true

This should be enough information for an LLM serving engine - such as LM Studio - to understand where to get the model weights (here that's lmstudio-community/Mistral-Small-3.2-24B-Instruct-2506-GGUF on Hugging Face, but it leaves space for alternative providers) plus various other configuration options and important metadata about the capabilities of the model.

I like this concept a lot. I've actually been considering something similar for my LLM tool - my idea was to use Markdown with a YAML frontmatter block - but now that there's an early-stage standard for it I may well build on top of this work instead.

I couldn't find any evidence that anyone outside of LM Studio is using this yet, so it's effectively a one-vendor standard for the moment. All of the models in their Model Catalog are defined using model.yaml.

Tags: standards, yaml, ai, generative-ai, llms, llm, lm-studio

tc39/proposal-regex-escaping

2025-02-18T21:53:56+00:00

tc39/proposal-regex-escaping

I just heard from Kris Kowal that this proposal for ECMAScript has been approved for ECMA TC-39:

Almost 20 years later, @simon’s RegExp.escape idea comes to fruition. This reached “Stage 4” at ECMA TC-39 just now, which formalizes that multiple browsers have shipped the feature and it’s in the next revision of the JavaScript specification.

I'll be honest, I had completely forgotten about my 2006 blog entry Escaping regular expression characters in JavaScript where I proposed that JavaScript should have an equivalent of the Python re.escape() function.

It turns out my post was referenced in this 15 year old thread on the esdiscuss mailing list, which evolved over time into a proposal which turned into implementations in Safari, Firefox and soon Chrome - here's the commit landing it in v8 on February 12th 2025.

One of the best things about having a long-running blog is that sometimes posts you forgot about over a decade ago turn out to have a life of their own.

Tags: blogging, ecmascript, javascript, regular-expressions, standards

RFC 7807: Problem Details for HTTP APIs

2022-11-01T03:15:05+00:00

RFC 7807: Problem Details for HTTP APIs

This RFC has been brewing for quite a while, and is currently in last call (ends 2022-11-03). I’m designing the JSON error messages for Datasette at the moment so this could not be more relevant for me.

Via Nicolas Fränkel

Tags: errors, http, json, mark-nottingham, rfc, standards

Announcing the Consortium for Python Data API Standards

2020-08-19T05:48:11+00:00

Announcing the Consortium for Python Data API Standards

Interesting effort to unify the fragmented DataFrame API ecosystem, where increasing numbers of libraries offer APIs inspired by Pandas that imitate each other but aren’t 100% compatible. The announcement includes some very clever code to support the effort: custom tooling to compare the existing APIs, and an ingenious GitHub Actions setup to run traces (via sys.settrace), derive type signatures and commit those generated signatures back to a repository.

Via @ralfgommers

Tags: python, standards, data-science, github-actions

Quoting Mark Pilgrim

2009-11-03T07:20:26+00:00

HTML has always been a conversation between browser makers, authors, standards wonks, and other people who just showed up and liked to talk about angle brackets. Most of the successful versions of HTML have been “retro-specs,” catching up to the world while simultaneously trying to nudge it in the right direction. Anyone who tells you that HTML should be kept “pure” (presumably by ignoring browser makers, or ignoring authors, or both) is simply misinformed. HTML has never been pure, and all attempts to purify it have been spectacular failures, matched only by the attempts to replace it.

— Mark Pilgrim

Tags: html, html5, mark-pilgrim, standards

There is no WebKit on Mobile

2009-10-07T12:23:26+00:00

There is no WebKit on Mobile

PPK ran 27 tests against 19 different WebKit-on-mobile implementations and found enormous disparities between the levels of support in currently available mobile phones.

Tags: mobile, ppk, standards, testing, webkit

CSS 3: Progress!

2009-08-22T11:52:17+00:00

CSS 3: Progress!

Alex Russell on the new exciting stuff going in to CSS 3 based on real-world implementations in the modern set of browsers. Of particular interest is the new Flexible Box specification, which specifies new layout primitives hbox and vbox (as seen in XUL) and is already supported by both WebKit and Gecko.

Tags: alex-russell, browsers, css, css3, flexiblebox, gecko, hbox, standards, vbox, webkit

On HTML 5 Drag and Drop

2009-08-17T12:31:12+00:00

On HTML 5 Drag and Drop

Francisco Tolmasky investigated HTML 5 drag and drop, which allows web apps to implement drag and drop between windows and between the browser and the desktop. He found a number of problems with the spec and proposes detailed solutions.

Tags: 280slides, draganddrop, francisco-tolmasky, html5, javascript, standards

10 Cool Things We'll Be Able To Do Once IE6 Is Dead

2009-04-15T14:17:14+00:00

10 Cool Things We'll Be Able To Do Once IE6 Is Dead

Highlights include child and attribute selectors, 24bit PNGs and max-width and min-width. Simple pleasures, but I can hardly wait.

Tags: brothercake, browsers, css, ie6, maxwidth, minwidth, pngs, selectors, standards

Counting the ways that rev="canonical" hurts the Web

2009-04-14T14:11:58+00:00

Counting the ways that rev="canonical" hurts the Web

Mark Nottingham complains about misapplied trust (a page can falsely claim to be the canonical URL for another page), the easy confusion between rev and rel and the lack of discussion with relevant communities.

Tags: mark-nottingham, revcanonical, standards, urls

Making the HTML5 time element safe for historians

2009-04-06T14:01:37+00:00

Making the HTML5 time element safe for historians

PPK presents a detailed history of dates and calendars and points out that the HTML5 time element is ill prepared to faithfully represent the kind of dates historians are interested in.

Tags: calendars, dates, datetime, historians, history, html5, ppk, standards, time

Quoting Ian Hickson

2008-12-19T13:58:00+00:00

Someone asked for onbeforeunload, so I started fixing it. Then I found that there was some rot in the drywall. So I took down the drywall. Then I found a rat infestation. So I killed all the rats. Then I found that the reason for the rot was a slow leak in the plumbing. So I tried fixing the plumbing, but it turned out the whole building used lead pipes. So I had to redo all the plumbing. But then I found that the town's water system wasn't quite compatible with modern plumbing techniques, and I had to dig up the entire town. And that's basically it.

— Ian Hickson

Tags: html5, ian-hickson, javascript, onbeforeunload, standards, whatwg

OAuth in Minneapolis

2008-11-20T18:55:21+00:00

OAuth in Minneapolis

OAuth looks like it’s on track for an IETF Working Group.

Tags: ietf, mark-nottingham, oauth, standards

Embedding custom non-visible data in HTML 5

2008-04-19T22:58:23+00:00

Embedding custom non-visible data in HTML 5

“Every HTML element may have any number of attributes starting with the string ’data-’ specified, with any value.”—this will be incredibly useful for unobtrusive JavaScript where there’s no sensible place to store configuration data as HTML content. It will also mean Dojo has an approved method for adding custom attributes to declaratively instantiate Dojo widgets.

Tags: customattributes, dojo, html, html5, javascript, standards, unobtrusive-javascript

HTML 5 vs. Yadis

2008-04-19T16:35:37+00:00

HTML 5 vs. Yadis

The draft HTML5 spec currently disallows values for http-equiv and link rel which aren’t listed in the spec—meaning both methods of specifying a link to an OpenID server are invalid for HTML5. This should probably be fixed...

Tags: html5, openid, standards, yadis

Quoting Benoît Jacob

2008-04-02T20:30:07+00:00

The ISO are now calling a "standard" the Microsoft Office format [...] What is interesting is that TeX, LaTeX, OGG/Vorbis, OGG/Theora, Perl, Python, PHP, Ruby, OCaml, are not standardized by any organization. [...] This shows that standardization organizations are no longer relevant in the software field. What really matters is free full documentation, free full implementation source code, and of course the absence of any patent risk. [...] In other words, what matters is evidence that any independent third-party can create and distribute a fully-conforming implementation.

— Benoît Jacob

Tags: benoit-jacob, iso, standards

CSS Compatibility and Internet Explorer

2008-04-02T20:05:28+00:00

CSS Compatibility and Internet Explorer

Official Microsoft guide to which CSS properties are supported by which versions of IE. This is the kind of documentation browser vendors should be providing as a matter of course.

Via Ian Muir

Tags: css, documentation, internet-explorer, microsoft, standards

Quoting Mark Pilgrim

2008-03-10T14:01:28+00:00

For the record, my site is valid HTML 5, except the parts that aren't. My therapist says I shouldn't rely so much on external validation.

— Mark Pilgrim

Tags: html5, mark-pilgrim, standards, validation

JavaScript in Internet Explorer 8

2008-03-06T23:59:11+00:00

JavaScript in Internet Explorer 8

John Resig’s analysis. News to me: IE 8 doesn’t support the W3C event model—I had assumed that would be a priority.

Tags: ie8, javascript, john-resig, standards

Internet Explorer 8 Readiness Toolkit

2008-03-05T18:28:20+00:00

Internet Explorer 8 Readiness Toolkit

The new built-in development tools look similar enough to Firebug to make me very happy. Also of interest: Selectors API (for fast getElementsBySelector), CSS 2.1 support, support for XHTML style namespaces in HTML, an interesting Web Slices feature based on the hAtom microformat and 6 connections per host (up from 2) which should make Comet easier.

Tags: comet, css, hatom, ie8, javascript, microformats, namespaces, selectors, standards

Acid3 is out

2008-03-05T00:34:20+00:00

Acid3 is out

The third Acid test, again compiled by Ian Hickson. This one viciously tests DOM Scripting standards compliance and currently exposes flaws in every browser.

Tags: acid3, ian-hickson, javascript, standards, web-standards-project

Principles and Legality

2008-03-04T19:45:55+00:00

Principles and Legality

Eric Meyer notes that language about legality in Microsoft’s recent IE announcement suggests that Opera’s much criticised EU threat may have helped positively influence the result.

Tags: eric-meyer, ie8, law, microsoft, opera, standards

Quoting Yoz Grahame

2007-12-26T17:58:32+00:00

Boxing Day toy discovery: Mega Bloks not compatible with Duplo! See, Alex Russell? THIS IS WHAT HAPPENS WHEN YOU INNOVATE AHEAD OF STANDARDS

— Yoz Grahame

Tags: alex-russell, duplo, lego, megabloks, standards, twitter, web-standards, yozgrahame

The future of web standards

2007-12-17T13:16:43+00:00

The future of web standards

Nice analysis from James Bennett, who suggests that successful open source projects (Linux, Python, Perl etc) could be used as the model for a more effective standards process, and points out that Ian Hickson is something of a BDFL for the WHAT-WG.

Tags: bdfl, ian-hickson, james-bennett, linux, open-source, perl, python, standards, w3c, web-standards, whatwg

Quoting Alex Russell

2007-12-16T23:33:44+00:00

To get a better future, not only do we need a return to "the browser wars", we need to applaud and use the hell out of "non-standard" features until such time as there's a standard to cover equivalent functionality. Non-standard features are the future, and suggesting that they are somehow "bad" is to work against your own self-interest.

— Alex Russell

Tags: alex-russell, browserwars, css, standards, w3c

Quoting James Snell

2007-11-18T00:15:22+00:00

I think it is well established that HTTP Authentication needs a major kick in the ass and OpenID and OAuth may get us most of the way there. However, until I see RFC#s attached to both I'm hardly going to consider them to be complete. I propose the creation of an IETF WG on Identity and Authentication. The WG would be chartered to produce two RFCs covering each of the two areas. OpenID and OAuth could be used to seed the WG effort.

— James Snell

Tags: http, ietf, james-snell, oauth, openid, rfc, standardisation, standards

The Story Behind ES4

2007-11-02T06:15:57+00:00

The Story Behind ES4

If you’re scratching your head at the recent eruption of acrimony surrounding ECMAScript 4 (the next standardised version of JavaScript) Neil Mix has a relatively easy to follow catch-up post.

Tags: ecmascript, es4, javascript, neil-mix, standards

Quoting Jeremy Allison

2007-09-15T10:40:18+00:00

My own favorites were Cuba voting "yes" to the fast-tracking of OOXML, even though Microsoft is prohibited by the US Government from selling any software on the island that might even be able to read and write the new format, and Azerbaijan's "yes" vote, even though OOXML as defined isn't able to express a Web URL address in Azeri, their official language.

— Jeremy Allison

Tags: azerbaijan, cuba, iso, jeremy-allison, microsoft, odf, ooxml, standards