[Standards] NEW: XEP-0387 (XMPP Compliance Suites 2017)

Sam Whited sam at samwhited.com
Fri Feb 10 17:13:05 UTC 2017


On Fri, Feb 10, 2017 at 5:58 AM, Georg Lukas <georg at op-co.de> wrote:
> | - Removal of 'User Avatars' from IM-Core, making it IM-Advanced only.
>> | - avatars require a graphical display to make sense (this is slightly
> |   mitigated by footnote 20)

So don't implement them in your TUI client. I added a footnote to
clarify this, but even without it as I've said before, if you can't
support it for technical reasons, then obviously you can't support it.
We're not going to throw you in XMPP jail for missing a spec when it
obviously can't be done in a sensible manner.

> | - avatars consume network bandwidth
> | - avatars are more of a cosmetic element than a functional one (not having
> |   them will not cause your messages to get lost in transmission)

To create a user experience that's both interoperable and meets
peoples expectations of what chat should look like, we need
functional as well as cosmetic elements.


> | - there are interop problems with file formats (some clients insist on
> |   using webp :-P )

As people have also said before: if a client isn't standards
compliant, there's nothing we can do about that. I don't see how
that has any bearing on this document.

> | - there are interop problems with MUC

It would be lovely if someone fixed that; I could possibly be
convinced to switch to the other avatars spec if people think this is
a problem here. With MIX on the horizon, us pushing for PEP everywhere
else, and "temp" in the name of the alternative, I just haven't felt
the need to change it.

> | I'd rather see 0198 and XEP-0184: Message Delivery Receipts as core
> | features, before addressing avatars.

I also don't see how these two things have anything to do with
Avatars. However, if you'd like to discuss moving 0198 to core or
adding 0184 as a separate discussion, that's fine.
I'm not really sure where 0198 makes the most sense; I only put it in
IM because I assumed there would be other use cases where it didn't
make much sense (maybe IoT or something).

—Sam


More information about the Standards mailing list