Re: Inventory of Activity types etc actually implemented in the wild

Hm,
well, the spec. Conformance Section says “The Entirety” :)

No guarantee, basically continuous trying …
Only a few Activities are supported, whyever. Not sure about Actor types.
This is about the used Object types:

Currently in mastodon anything seems to be converted to Note but you can send some other
types. Not sure about “Collection”s …
whyever.

Bonfire and misskey have rather generic models and are open.
misskey has some extensions like `quote`, `reaction`, `votes`, `talk`, `isCat`
https://misskey-hub.net/ns.html <https://misskey-hub.net/ns.html>

Then it is e.g. 
• Event in mobilizon alongside schema:address 
https://docs.joinmobilizon.org/contribute/activity_pub/ <https://docs.joinmobilizon.org/contribute/activity_pub/>
• Video in peertube 
• Image in pixelfed.
• `Book` and `Edition` extensions (wikidata integrated) in inventaire.
https://wiki.inventaire.io/wiki/What_is_the_Fediverse <https://wiki.inventaire.io/wiki/What_is_the_Fediverse> ff.
• `Review`, `Comment` and `Quotation` and `Shelf` and `List` in bookwyrm.
https://docs.joinbookwyrm.com/activitypub.html <https://docs.joinbookwyrm.com/activitypub.html> 
The purpose is always the whole app scope, whyever.

Then
• `Place` in the “Fetaverse” [Lomma]: 
https://immers-space.github.io/immers-client/Activities.html#.APPlace <https://immers-space.github.io/immers-client/Activities.html#.APPlace>
Not sure about Lemmy [404 for first link in]
https://join-lemmy.org/docs/en/federation/overview.html <https://join-lemmy.org/docs/en/federation/overview.html>

• Podcasts:
Alongside `Audio` Funkwhale has 
`Artist`, `Album`, `Track`, `Library`
https://docs.funkwhale.audio/federation/index.html#supported-objects <https://docs.funkwhale.audio/federation/index.html#supported-objects>
Other podcast people started their thing in 
https://github.com/Podcastindex-org/podcast-namespace/blob/main/proposal-docs/social/social.md <https://github.com/Podcastindex-org/podcast-namespace/blob/main/proposal-docs/social/social.md>
There is much more smaller specialised software which we do not support 
cause looking everywhere anytime is much unneeded work.

About us:
redaktor will support all core-types of course.
Currently we are working on the UI for Place and Video, other drafts are ready in 
https://github.com/redaktor/widgets-preview <https://github.com/redaktor/widgets-preview>
[btw, Place comes together with 400.000 well known Places ready in ActivityVocabulary]

We will support Client-to-Server for any specialized client 
(assignments e.g. for german Transport or Public Broadcaster sector).
Our extensions are simply said wikidata and IPTC and EBU vocabs.

Our assumption is just the view of the user getting any data in an inbox and wants to have 
all data for em to be visible. This deserves nice designs for all the core types and then just 
the assumption that most things can have multiple items and/or languages. 
A simple details/summary list can handle the unknown.
The Fediverse will just never work if types are ignored: 
I'd send an Article to a software only accepting Event and do not even know what happens.
Normally, any LD software should at least accept input independent from “type”.
For me as journalist it is “stunning” to “separate” complete apps by “type” without any relations in the data.
A `Video` shows multiple `Image` and they show `Place` which all might host `Event` organised by `Group` 
which is mentioned in `Article` published in `Collection` etc. pp.


Inventory:
About the “inventory”, it is WIP, sorry. 
We tried to document extensions with 
https://github.com/redaktor/vocab <https://github.com/redaktor/vocab> 
If you click “Community Namespaces” in
https://redaktor.github.io/vocab/w3id.org/class/redaktor/scheme.en.html <https://redaktor.github.io/vocab/w3id.org/class/redaktor/scheme.en.html>

cheers, Sebastian