Dynamic Titles are not read

Titles are not scraped properly when generating the url and collection title if they are dynamically generated. Example https://observablehq.com/@tomlarkworthy/untitled

This came up for me when I was trying to setup a master template notebook for forking and reuse multiple times.

Yup, the title is only updated if the first cell in a notebook is updated (search in your dev tools for “extract_title”). However, I get why they did it this way. Shifting control over title updates to the iframe (where the runtime lives) could cause all kinds of issues and invite abuse.

I guess they could pass on a token that would ensure that only the first reported update after an editor update gets registered.