Re: [core] [T2TRG] CoRE Applications side-meeting @IETF105

[Klaus Hartke <[e-mail]>]

(trimming the cc list a bit...)

Carsten Bormann wrote:
> Here are a couple of comments I haven’t made yet to the mailing lists:
>
> (1) A subscriber that is waiting for the first Data should observe the info object? An example might help.

>From the point of view of a subscriber, the idea is that setting up a
topic is a two-step process: First, the topic configuration has to be
set up, and then the topic data resource has to be created by
publishing the first item. While only the first step has been
performed, the topic doesn't exist to a subscriber. Topic creators can
expedite the creation of the data resource by publishing the first
item themselves.

> (2) So the data link had to be there from the outset (with a 404) so the publisher can PUT to it? But then it's not a 404, but a 405.

A link can reference a resource that doesn't exist (yet) [1]. I think
it's simpler to say that the topic data resource doesn't exist until
the first item has been published, rather than that the resource
exists but doesn't support GET. But I don't have a strong opinion on
this...

Klaus

[1] " A resource can map to the empty set, which allows references to
be made to a concept before any realization of that concept exists"
<https://www.ics.uci.edu/~fielding/pubs/dissertation/rest_arch_style.htm>