On Thu, Jul 19, 2012 at 4:19 AM, Lance Stout <lancest...@gmail.com> wrote:
> > I think it's worth including the id on every RTT edit, > > rather than just the first - it makes the state machine easier for the > > receiving clients and doesn't hurt the sending client. > > +1 on this. Even though the use of the seq value and error detection can > be used to ensure that the RTT update is still within the context of > editing a previous message, keeping that context explicit is a good thing > (and was what my hypothetical version of a XEP-0308 flow was intended to > accomplish). > I like it, but I don't want to make it required in situations where last message is not being edited. Many XMPP libraries are hardcoded to generate an 'id' only at the moment <body/> is generated. Over time, this can change. Therefore, I prefer that 'id' be allowed to be omitted; the omission of 'id' represents the editing of the most recent message. However, it's practical to require 'id' for all <rtt/> elements during last-message editing, to simplify the state machine for the receiving client. With that, I'm strongly in favor of Mark's proposal. It should be a small > enough section to fit in fine with XEP-0301, and shouldn't slow down its > progress to draft state, which is a consideration given the broader > legislative situations. > The only problem is that XEP-0308 would be the ONLY "Experimental" standard I am referencing, and I'm trying to avoid referencing any experimental standards at the moment...
