This is an automated email from the ASF dual-hosted git repository. davsclaus pushed a commit to branch blog/website-redesign-mission in repository https://gitbox.apache.org/repos/asf/camel-website.git
commit 0b9e94d9659ca7ad481f5232e51a3de8f2f02e57 Author: Claus Ibsen <[email protected]> AuthorDate: Thu Sep 17 07:32:24 2026 +0200 Blog: a new front page for Camel, written for humans and for AI Announces the website redesign with the front page and its mission as the centre: what Camel is, three ways to start, one route in three syntaxes, measured numbers, and the AI-ready card. Takes the rest of the redesign from #1761 in short form, then the documentation work in progress: build-checked code listings, the content restructure, and Markdown mirrors and llms.txt for AI agents, with the website pages as the next step. Co-Authored-By: Claude Fable 5.1 <[email protected]> Claude-Session: https://claude.ai/code/session_01Bp3538HRBPMQkb5ta9xRaj --- .../09/camel-website-new-front-page/featured.jpg | Bin 0 -> 105769 bytes .../front-page-after.jpg | Bin 0 -> 119531 bytes .../front-page-before.jpg | Bin 0 -> 128361 bytes .../2026/09/camel-website-new-front-page/index.md | 58 +++++++++++++++++++++ 4 files changed, 58 insertions(+) diff --git a/content/blog/2026/09/camel-website-new-front-page/featured.jpg b/content/blog/2026/09/camel-website-new-front-page/featured.jpg new file mode 100644 index 00000000..04f4e7c1 Binary files /dev/null and b/content/blog/2026/09/camel-website-new-front-page/featured.jpg differ diff --git a/content/blog/2026/09/camel-website-new-front-page/front-page-after.jpg b/content/blog/2026/09/camel-website-new-front-page/front-page-after.jpg new file mode 100644 index 00000000..fe618fa7 Binary files /dev/null and b/content/blog/2026/09/camel-website-new-front-page/front-page-after.jpg differ diff --git a/content/blog/2026/09/camel-website-new-front-page/front-page-before.jpg b/content/blog/2026/09/camel-website-new-front-page/front-page-before.jpg new file mode 100644 index 00000000..167fb789 Binary files /dev/null and b/content/blog/2026/09/camel-website-new-front-page/front-page-before.jpg differ diff --git a/content/blog/2026/09/camel-website-new-front-page/index.md b/content/blog/2026/09/camel-website-new-front-page/index.md new file mode 100644 index 00000000..65ae6965 --- /dev/null +++ b/content/blog/2026/09/camel-website-new-front-page/index.md @@ -0,0 +1,58 @@ +--- +title: "A new front page for Camel, written for humans and for AI" +date: 2026-09-17 +draft: false +authors: [davsclaus, ammachado] +categories: ["Community"] +keywords: ["apache camel", "website", "redesign", "front page", "documentation", "llms.txt", "ai agents"] +preview: "camel.apache.org has a new front page and a new job: tell a person, or an AI agent, what Camel is and get them to a running route in a minute. What changed, and what comes next." +--- + +The Camel website has needed a modern front page for years. The old one said that Camel "is an Open Source integration framework that empowers you to quickly and easily integrate various systems", showed a picture of gears, and left the rest to you. True, but it could describe any integration product, and it did not tell a first-time visitor how to start. + +Since that page was written, the visitor has changed. A growing share of the readers of camel.apache.org are not people but AI coding assistants and crawlers, reading the site on a developer's behalf. When someone asks their assistant how to get messages from Kafka into a database, the answer is shaped by what the assistant has read about Camel. The front page is the most authoritative page there is about Camel, so it has to answer that question for a person and for a model alike. + +That is the mission of the new front page, and it is the most important change in the redesign. + +## The front page + + + + + +Read it from the top, the way a visitor does: + +- **One sentence that says what Camel is.** Connect anything, route everything: 350+ connectors, written in Java, YAML or XML, deployed on Spring Boot, Quarkus or standalone with the Camel CLI. In production since 2007, Apache License 2.0. +- **Three ways to start, with the commands.** The Camel CLI comes first: install it, `camel init hello.yaml`, `camel run hello.yaml`. No project, no IDE, no Java compilation. Next to it, a Spring Boot project from Spring Initializr and a Quarkus extension, for the two runtimes most Camel applications run on. +- **The numbers.** 83,000+ commits, 1,500+ contributors, 350+ components, 300+ releases. Each figure was measured against the repository on a stated date, and the source data file records how to re-measure it. The page makes no claim it cannot back. +- **One real route, in the three syntaxes.** Messages from Kafka into a database, as YAML, Java and XML, with the `camel run` command that matches the tab you pick. A reader sees in ten seconds what a Camel route looks like, and a model sees the same route three times with the syntax named. +- **Why developers choose Camel**, in six short cards: connectors, three DSLs, AI-ready integration with the Camel MCP server and `camel trace` and `camel test` to verify AI-generated routes, the Enterprise Integration Patterns, a standard Maven or Gradle build, and open source. The section ends with a link to [why teams trust Camel in production](/trust/). +- **The rest of the ecosystem** moved below: Camel Core, Kamelets, Camel K, Karavan, the Kafka Connector and Karaf. The front page shows what most users use first, and everything else is one click away. +- **Latest from the blog**, and a closing invitation: Camel is your project, here is how to get involved. + +The words on the page were chosen with both readers in mind. "Integration framework", "350+ connectors", "Java, YAML or XML", "Spring Boot", "Quarkus", "Enterprise Integration Patterns", "MCP server" and "Apache License 2.0" are the terms a person scans for and the terms a model associates with Camel. There is nothing on the page a visitor cannot run or check. + +## The rest of the site + +The front page sets the tone, and the other pages follow it: + +- **Docs and website share one look.** The documentation pages now have the same header and footer as the rest of the site, so moving between a blog post, the user manual and a component page no longer feels like changing websites. Marketing pages scale with the viewport, and documentation pages, which carry a sidebar and a table of contents, get all the width they need. +- **Why Camel** is the first link in the top bar. The trust page, with the project's security track record, release cadence and adoption, used to hide in the footer; now it is one click from anywhere. +- **Search** runs on DocSearch v5, themed to match the site. The result curation carried over: core documentation ranks above component pages, and no single page can fill the list. +- **The blog** shows ten posts per page instead of three, with a compact card per post, and opens with a "Start here" rail of eight hand-picked posts for a newcomer. + +URLs, RSS feeds and the documentation content are unchanged. Every link that worked before works now. + +## The documentation is next + +The design is done. The content is where the work is now, and it has already started. + +**Code listings that are checked, not trusted.** The Camel documentation has thousands of examples, and for years the only check was a reader trying one. We now run them through the same validation that the Camel CLI and the MCP server give an AI assistant, at build time, so a broken example fails the build. The first pass over the 290 YAML examples on the Enterprise Integration Pattern pages found 71 wrong ones. The next pass over the 1,820 YAML route examples on the component pages fou [...] + +**A restructure of the content.** The user manual has grown for nineteen years, and its structure reflects that history. The next step is a restructure so that a newcomer, or an agent, lands on the right page without having to know how Camel evolved. + +**Every page readable by machines.** Every documentation page is available as Markdown: replace `.html` with `.md` in the URL, and the page comes back as plain Markdown without navigation, which is what an assistant wants to read. [llms.txt](/llms.txt) describes Camel for AI agents in one file and points at versioned offline documentation bundles, with the catalog metadata and the YAML DSL schema included, for agents that work without internet access. The website pages themselves, this b [...] + +## Tell us what you think + +The site is built in the open in [apache/camel-website](https://github.com/apache/camel-website). If something looks wrong, a link points at the wrong place, or a page does not read well to you or to your assistant, please [open an issue](https://github.com/apache/camel-website/issues). The front page is the first page a visitor and a model read about Camel, and we want it to be right.
