[Standards] microblogging maintainer :)

Dave Cridland dave at cridland.net
Mon Dec 20 12:29:00 UTC 2010

On Mon Dec 20 13:10:19 2010, Thomas Baquet wrote:
> - As written in the draft, "The client MAY add a “reply”  
> attribute which gives the parent existing collection node. The  
> default value of this attribute is the node where the item is  
> published." - Actually I was a sick when i've written this draft  
> (perhaps that's why I couldn't express me correctly --), but I  
> would meant that by /default/ this is a child node of the node  
> where the item is published. But, that can be a nice feature if the  
> user is available to choose the node where the item will be  
> published;

The trouble with this model is that you're redefining how pubsub  
works quite radically.

Nodes, currently can be one of two types:

Collection Nodes, which contain other nodes (or, more strictly,  
pointers to other nodes).

Leaf Nodes, which contain items.

In both cases, they can contain as many things (nodes or items) as  
you like, although in PEP implementations this needs to be only one,  
and most (all?) current implementations restrict the number of items  
to a minimum.

As such, I'd expect implementations of this to support many items per  
node, and full access model control to allow (or disallow) replies on  
the node as needs be. There's really no need to a protocol extension  
to pubsub to support this.

In order to effectively unify the two models (ie, posting replies on  
your own node versus posting replies on the original poster's), I'd  
have thought you do this:

1) Replies are posted to the poster's µblog node.

2) The replier also posts a pointer (possibly duplicating the post)  
to a replies node on their own personal pubsub service.

Then, the only "risk" is that a replier actually posts a top-level  
entry into someone else's µblog - in this case, though, it's pretty  
easy to spot and the node owner can remove it as well.

Dave Cridland - mailto:dave at cridland.net - xmpp:dwd at dave.cridland.net
  - acap://acap.dave.cridland.net/byowner/user/dwd/bookmarks/
  - http://dave.cridland.net/
Infotrope Polymer - ACAP, IMAP, ESMTP, and Lemonade

More information about the Standards mailing list