Services / Migration
I move documentation out of wikis and CMS-style tools such as Confluence, Notion and Quip, and into AsciiDoc or Markdown in Git. I handle both the mass conversion and the critical editing and auditing that go with it, so you end up with a smaller, current set of docs rather than the same old problems in a new format.
A migration doesn't mean every page gets the same care. Older articles that nobody reads may only need bulk conversion and archiving, which is fast and cheap. Customer-facing pages, such as release notes, need careful conversion and checking, because people rely on them being accurate. Sorting content this way during the audit is what keeps a large migration manageable, and puts the effort where it matters.
At Kaptio, an audit found a daunting collection of out-of-date articles, previously stored in a paid CMS. Using Pandoc to convert a huge number of these articles in bulk, I moved them to our Notion archives, saving us the CMS bill. This process also impressed upon my colleagues the benefits of docs-as-code, as well as showing the advantages of mass conversion and this form of archiving.
The release notes were where I really shined. I set about my own self-driven project with the aim of building a new public-facing site. This site would naturally require an Antora build achieved through a docs-as-code CI/CD pipeline. I converted the release notes to AsciiDoc with Pandoc and updated the links as I went, so the new site was current and usable for customers.
With docs-as-code, documentation can be versioned to specific releases, so your product and documentation update together. The documentation system also fits into your existing engineering and update processes.
Related: Antora and Docusaurus sites · Docs-as-code CI/CD pipelines