Re: [CfC] Erratum for ActivityPub on the definition of `inbox` the property

Can we apply the erratum as written, and deal with "all" in a separate 
issue?

And yes, please, open an issue on the ActivityPub issues repo, please!

Evan

On 2025-06-01 6:58 a.m., Melvin Carvalho wrote:
>
>
> pá 30. 5. 2025 v 20:15 odesílatel Evan Prodromou <evan@prodromou.name> 
> napsal:
>
>     Great question.
>
>     The original text is:
>
>         /A reference/to an [ActivityStreams] OrderedCollection
>         comprised of _all_ the messages received by the actor; see 5.2
>         Inbox.
>
>     Note the underlined all. What we're changing here is "/A
>     reference/to an [ActivityStreams] OrderedCollection"
>     to "An OrderedCollection", to be more parallel with the outbox
>     property.
>
>     I do see your point; some implementations don't even make this
>     collection readable, so even though POSTing to the collection is
>     adding an activity to it (POST-to-create pattern), it doesn't
>     "stay there".
>
>     It's a little orthogonal to this erratum, though. Would you mind
>     adding a new issue to the repo?
>
>
> Thanks Evan, that makes sense.
>
> Just to add a bit of historical context, when I originally suggested 
> the "Semantic Inbox", the intention was quite similar to an email 
> inbox, only better, since it leveraged HTTP to support richer, 
> structured data messages. Semantic Inbox then evolved into what became 
> Solid's inbox pattern, and from there, it influenced the inbox concept 
> used today in ActivityPub.
>
> I'm genuinely delighted to see this idea getting traction and being 
> used so effectively! You're right that implementations vary, email 
> inboxes, for example, tend to be private by default in the email 
> world, and people often delete items or even strive for "inbox zero." 
> An inbox can be a temporary staging area rather than permanent storage 
> (think physical offices), and that's a perfectly valid use-case too.
>
> I completely agree we're looking at both errata and architecture 
> topics here, althought they're slightly intertwined. Perhaps it'd be 
> helpful if we could reflect briefly on how inboxes are practically 
> used across implementations today, and then use those insights to 
> inform our errata and clarify the definition?
>
> Happy to open an issue to dive deeper into this if that helps.
>
>
>     Evan
>
>     On 2025-05-28 2:21 a.m., Melvin Carvalho wrote:
>>
>>
>>     pá 23. 5. 2025 v 18:33 odesílatel Evan Prodromou
>>     <evan@prodromou.name> napsal:
>>
>>         Issue #289 of the ActivityPub GitHub repository notes the
>>         inexact and asymmetrical language used for defining the
>>         `inbox` and `outbox` collections:
>>
>>         https://github.com/w3c/activitypub/issues/289
>>
>>         To resolve this, there is a proposed erratum for Section 4.1
>>         to bring the definition of the `inbox` property closer to
>>         that of the `outbox` property:
>>
>>           * In section 4.1 "Actor objects", the definition of "inbox"
>>             should read, "An OrderedCollection comprised of all the
>>             messages received by the actor; see 5.2 Inbox."
>>
>>         We usually handle approval of errata in our synchronous
>>         meetings, which takes a lot of time and focus away from other
>>         topics that require more immediate presence and conversation.
>>         In speaking with the chairs and others in the issue triage
>>         meeting, we think that handling this task through the CFC
>>         decision-making process will be more efficient.
>>
>>         So, I am seeking consensus to add this erratum to our errata
>>         page for ActivityPub. Please reply either to the GitHub issue
>>         or here on the mailing within 14 days of this message.
>>
>>
>>     Thanks Evan, small point to consider: does “all” messages mean
>>     every received message should be archived in the inbox? I’m not
>>     sure all implementations do that.
>>
>>         Evan
>>
>>

Received on Monday, 2 June 2025 13:37:53 UTC