[Standards] -198, was Re: Namespaces, specifications, and protocol life cycles

Dave Cridland dave at cridland.net
Tue Sep 9 12:18:36 CDT 2008


On Tue Sep  9 17:54:34 2008, Kevin Smith wrote:
> On Tue, Sep 9, 2008 at 5:49 PM, Dirk Meyer <dmeyer at tzi.de> wrote:
> > you wrote about XEP-0198 and that
> > is more complicated. IMHO it is a MUST HAVE extension and I will
> > implement it as soon as I can, but I have no server to test it
> > with. And the same is true for server developer: they have no
> > client.
> 
> At least in the 198 case, server implementations can test against
> themselves, while clients require the server, so it's possible to
> bootstrap things.

And I was tempted to do just that, because I'm quite keen on 198 for  
various reasons.

Only I don't really want to implement anything that will definitely  
become incompatible, and the message using a :tmp: namespace is that  
not only will anything I write at least need a namespace change, but  
it'll also be changed on the assumption that nobody's implementing,  
too.

In other words, if, say, someone decides that 198's <r/> and <a/>  
elements are entirely semantically equivalent [they aren't, but could  
be made so easily] and decides to remove the one I happen to be  
sending, the namespace won't change and I'll end up with an  
incompatible implementation.

Dave.
-- 
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