The blog and the release history

The blog writes about this product: how a feature works, why a boundary was drawn where it was, and how to get something useful out of a tool that already ships. It is sorted into four categories, and every article is a real page in version control rather than a post in a content system.

The changelog is a dated record of what shipped. It is not rewritten when wording standards change, because a historical record that is edited to match a current style rule is no longer a record. The blog explains; the changelog documents.

• The blog publishes an RSS feed, and each category publishes its own.
• Everything is a plain page, so a reader that follows links will see the whole article rather than a teaser.
• There is no newsletter signup, because there is no mailing list.