Jay, I think the spec should say something like : "The publisher MUST inform the hub about new content." without which, obviously the protocol won't work. However the "how" should be left outside the protocol, or may be something like "the hub MAY implement method X to get notifications". I think one of the "design" principles of the protocol was that subscribers should not care which hub a feed is using as their subscription should work with *any* hub. On the other end, since the publisher designates a *specific hub*, the hub and the publisher can agree on the method they want to use without affecting anyone else. In other words : there should be no coupling between subscribers and hubs, no coupling between subscribers and publishers, but since there is coupling (thru the discovery) between the hub and the publisher no matter what, we should probably not spec it. Does it make sense?
Julien On Tue, Nov 22, 2011 at 11:06 PM, Jay Rossiter <[email protected]> wrote: > > I disagree with this. Defining the 'publish'/ping mechanism is > important to the specification. Just because there can be alternate > methods for obtaining this behavior, such as a self-contained > publisher/hub, doesn't mean that its purpose is not useful. The > publisher notifying the hub that there are updates is a very > significant core feature, and leaving that undefined would seem to be > a critical mistake, to me. > > Section 7.1 is labeled MUST, however if the hub does not provide > public services it's rather moot, and that exception is already > covered by "If the notification is not acceptable for some reason, the > hub MUST return an appropriate HTTP error response code (4xx and > 5xx)." "For some reason" is obviously undefined, and can be anything, > including the fact that the hub just doesn't support that kind of > request. Certainly, if implemented, aspects of those sections are > required. i.e. you can't support 'publish' and not support hub.url. > > On Nov 20, 7:59 am, Julien Genestoux <[email protected]> > wrote: > > All, > > > > I believe that the Publisher -> Hub part of the protocol should be left > out > > of the "core" spec. > > The motivation behind it is that many hubs out there (Wordpress, > > Superfeedr's... etc) have different ways of handling this without > affacting > > much of how the general "idea" of the protocol works. > > I believe the Section 7.1 should be set as optional, as well as section > 7.2. > > Something like this would work well, according to me : > > Once a publisher designates a hub he wants to use, he then SHOULD use any > > of the supported method by this hub to notify new content. 7.1 and 7.2 > > should certainly be put as an annex/example of how this can be done. > > > > What do you think? > > > > Julien >
