[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

RE: [JDEV] Contact Methods



On Fri, 30 Apr 1999, Thomas Charron wrote:

> > So, the tree under a user in a roster is based on their sessions.  Any
> > "transports" that user is making use of is that users business, nothing we
> > will care about.
> 

*and*

>      +---------------------------------------+
>      | Contact List                          |
>      |  + Jane Doe (Jane Doe@jabberhost.com) |
>      |  - Joe Blow (IWuvJer@aol.com)         |
>      |     `- 167289@pagers.jabberhost.com   |
>      |  - Ant C. Eater (234567@pagers.jabber>|
>      +---------------------------------------+

we're REALLY close now.  the one problem i have with this diagram is the
pager email address.  for all intents and purposes, a SMTP transort could
be used for the 167289@pagers.jabberhost.com (maybe masked as a Pager
entry?), but still vital info regarding a person's pager is given away
(straight email to that pager, etc).  if however, the number you are using
here is itself an address for the transport to understand itself, nice.

the gig is that if i have a pager, i don't want sally in san fran to know
my pager number of ID.  i just want certain ppl to have access to it
through jabber (authorize on add?) and others who i know won't abuse it to
have the actual phone number.

> 
> 	In this diagram, I have the 'Primary address' next to their names,and
> alternate addy's under..  Notice Joe Blow.  I know I can contact him via
> AOL,or his pager.  In this case, I DO care HOW ELSE I can reach him.  I may
> not care about what transports Joe Doe uses, but if he tells me he can also
> be XSexyHoeX@icq.wuvviedovie.com, I want to be able to add that as an
> 'alternate' address, and not have it as another primary address.  This
> appears to be what your saying, but the initial question (At least it's what
> I want to know) is how will the status messages to SHOW this roster look?

this is interesting.. i originally thought it would be more elegant to
have a Name directly linked to a jabber ID (or not linked to one at all if
that person didn't have a jabber account).  but now, after thinking about
a person only having one contact transport the idea of a Primary address
and then multiple Secondary addresses becomes clearer.

if Joe Doe doesn't have a jabber address, but only a ICQ number we add him
with primary address being his ICQ number@ICQ.  when he does finally get a
jabber address (cos he's heard so much about jabber ;) then the server
automagically switches the primary address for this user (or does this
happen as one of the user's prefs?).

> [snip]
> 
> 	or this:
> 
>   <roster>
>     <group name="main">jenny</group>
>     <group name="main">545212@ICQ<ext><Alternate
> number=1/>jenny</ext></group>
>     <group name="friends">user@jabber.server.com</group>
>     <group name="system">olduser</group>
>   </roster>
> 

this looks good, but the "number" section does what exactly?  if it
specifies the priority of which method of contact to use first, and the
client enforces it, perhaps there's a better way.

otherwise, if it is something the user sets when maneuvering people around
his/her roster, nice.  this would keep the order based on What You Want It
To Be.

> 
> 	Actually, while we're on the topic, is there a way in the status messages
> to give 'em a person use name.  Aka, perhaps I want my brother in law,
> XNateReadX@aol.com simply as 'Nate' in my contact list.  How is that done
> via the current status system?  Kind of like above, how is 'Joe Doe' passed
> as a name?(As you can tell, the status box in the current Win32 client
> bites..  Working on it this weekend..)
> 

could there be an emergence of a <entry> tag with extensions being
Pager="" ICQ="" etc etc:

 <group>Main</group>
 <entry j="Joe Doe@jabberhost.com" pager="JoeDoe.pager@jabberhost.com">\
 Joe Doe</entry>
 </group>

or is this inefficient/not needed/too late in the game?  i'm sure the
current spec will function wonderfully if so.

regards,
Chase Phillips
--
  shepard at ameth.org ][ Only you will know
 http://www.ameth.org/ ][ if I proofread this