Hi, writing from the implementer side. I run Writizzy, a hosted blog platform using standard.site. But I'm also a reader/user on standard-reader.app 😃
Context #
There is currently no way for a reader to offer subject-based browsing of publications, because nothing in the publication record says what a publication is about.
Standard Reader's Discover page does present topics, which is great, but I guess they have to read the content and clusterize the content themselves. Some topics are duplicated and it's not perfect (most listed publications are in english, I don't know if it's related). That's what you get when subject has to be inferred from article text rather than declared: it mostly works but it could be better.
Proposal #
An optional categories field on site.standard.publication:
- Optional. A publication without one stays perfectly valid.
- Not enforced. Declared with
knownValues, notenum, so the listed values are guidance and any other string remains legal. Consumers must tolerate values they don't recognise. - Low cardinality. An array with a small maximum. My instinct is 3, with 1 being the common case, but I'd rather hear arguments than pick a number. A single string is also defensible and I don't feel strongly.
I suggest free text with know values because it will be very hard to find a common vocabulary for every platform. Let's say someone creates a platform for tech writers only. Categories could be quite granular in tech and not adapted to a generalist platform. However, having a common vocabulary could help reader's apps
Offered only so the discussion has something concrete to argue with. This is the taxonomy we use on our own platform, so it is inevitably shaped by the kind of blogs we host, and I'd expect it to be wrong in several places:
tech, ai, startup, finance, science, health, politics,
culture, art, music, sports, food, travel, education,
productivity, marketing, design, gaming, parenting,
environment, other
I can contribute if ever this proposal gets some approvals.