> On 14 Nov 2017, at 21:37, Sam Whited <s...@samwhited.com> wrote: > >> On Tue, Nov 14, 2017, at 12:06, Dave Cridland wrote: >> So arguing over whether it's a Core, Advanced, or neither feature >> seems a bit pointless - except that it means XEP-0387 may reflect >> neither the current reality nor any particular desirable future. > > I think the distinction between Core/Advanced might not be fantastic and > have ideas for changing that, but I wasn't going to bring it up until we > start on next years. > > As it stands right now, I could still go either way on this and think > the only thing that really matters is "does it present an undue burden > for server developers to implement"? I'm not sure. It was a bit of a > pain that last time I implemented it (and required a major refactoring > of the way I handle connections), but I had assumed that was just my > library because I hadn't considered alternative stream negotiation > protocols at all (presumably if I'd done the websocket protocol first it > would have been the same problem). > > What do the server devs here think?
My knee-jerk reaction was that it should go. My slightly more considered reaction is that it should stay. It’s not particularly onerous to implement, and it’s widely used (and, importantly for Dave’s criterion, expected). /K _______________________________________________ Standards mailing list Info: https://mail.jabber.org/mailman/listinfo/standards Unsubscribe: standards-unsubscr...@xmpp.org _______________________________________________
