[Standards] XEP-0392: Consistent Color Generation - Issues using JID

Jonas Wielicki jonas at wielicki.name
Mon Nov 20 08:19:29 UTC 2017


On Montag, 20. November 2017 09:09:48 CET Maxime Buquet wrote:
> Hi there,
> 
> A few remarks regarding 0392 after having it implemented in poezio,
> thanks to Jonas.
> 
> Using JIDs as the source for the hash brings at least an issue to me in
> MUCs. You'll see the same person in different colors depending on what room
> you're looking at.

To be clear, the algorithm uses the bare JID of the occupant instead of the 
occupants nickname if it is known to the client (requires non-anon rooms or at 
least moderator privilegues, IIRC).

> Another point I feel about but I don't know how to defend is that people
> not using MSN would see their different nicks colored the same.

Indeed.

This is a trade-off:

(a) prefer nickname: consistent coloring everywhere, iff the same people use 
the same nicknames. The downside is that entities using different nicknames 
while the JID is known get colored differently.

(b) prefer JID: inconsistent coloring between rooms where the JID of the 
participants is known and where it is not known. Note that this inconsistency 
is consistent when showing avatars (usually, those are only available iff the 
bare JID is visible, too). Has the other advantage that the same entity gets 
the same color if the JID is known.

So (b) has a much nicer UX if we simply ignore (semi-)anonymous rooms. It has 
a slightly worse UX if the set of people overlaps between (semi-)anonymous and 
non-anonymous rooms.

I have no strong opinion one way or the other, but we need a configuration-
switch-free consistent behaviour here. Any arguments welcome.

kind regards,
Jonas
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 833 bytes
Desc: This is a digitally signed message part.
URL: <http://mail.jabber.org/pipermail/standards/attachments/20171120/13c8b369/attachment.sig>


More information about the Standards mailing list