Scott Lawrence wrote:
>
> On Fri, 2008-09-26 at 09:43 -0400, Paul Mossman wrote:
> > Hi all,
> > 
> > In the existing 2007-02-28 sipXrls proposal the Polycom's 
> BLF button 
> > is configured with a contact that includes the retrieve 
> code, so that 
> > pressing the button results in dialing of the unpark code for the 
> > orbit.  The drawback with this is that because the contact 
> is not the 
> > Park Orbit's extension, the BLF button cannot then be used when 
> > parking a call.  i.e. "Trnsfr" softkey --> "Blind" softkey --> Park 
> > Orbit BLF button.
> > 
> > A main goal is that the Park Orbit extension should never 
> need to be 
> > composed on the keypad.  We can do this today with two 
> buttons.  The 
> > Park Orbit is BLF monitored with one button, which is only pressed 
> > while invoking a Transfer to park a call.  Parked calls are then 
> > retrieved by pressing another button which is a basic speed dial 
> > containing the system Call Park retrieve code followed by the Park 
> > Orbit's extension.
> > 
> > But two buttons per Park Orbit is confusing.  (The boss wants to be 
> > able to press the button that is lit to retrieve the call.)  It is 
> > also a waste of a button.  The challenge with converging 
> this into a 
> > single button is that in different situations the phone 
> would need to 
> > use a different contact for the Park Orbit button: either the 
> > extension, or the retrieve code followed by the extension.
> > 
> > 
> > This proposal would achieve single button retrieve, while still 
> > allowing the button to be used when parking the call.  It 
> required two 
> > changes, both to sipXecs only:
> > 
> > 1. sipXpark would intentionally fudge its Dialog Event NOTIFYs to 
> > report the dialog state of parked calls as "early" instead of 
> > "confirmed".  This would have the following two desirable effects:
> > 
> >   
> > a) BLF monitoring of an occupied park orbit would show "ringing"
> > indication.  While the parked calls are not technically ringing, I 
> > think the user experience is actually desirable.  (It is a 
> queued call 
> > waiting to be attended to.)  And since the park orbit 
> doesn't actually 
> > ring for any significant period of time, I think we can 
> commandeer the 
> > "ringing" indication without confusing the user.
> > 
> >   
> > b) Since parked calls show up with "ringing" BLF 
> indication, they can 
> > be un-parked by pressing the BLF button to invoke the Polycom's new 
> > Call Pickup functionality.  (XCF-2716 added the required
> > configuration...)
> > 
> >   
> > 2.  For b) to happen I think the Call Pickup redirector 
> will need to 
> > be a little smart.  It will have to know that the Call 
> Pickup INVITE 
> > with Replaces for park orbits should not have the 
> "early-only" flag, 
> > since the target set is actually in a "confirmed" dialog.  
> But, this 
> > redirector already does call un-park anyway, so that shouldn't be a 
> > big change.
> > 
> >   
> > There is no change to sipXconfig required.  To configure this you 
> > simply create a User Speed Dial for the Park Orbit, and select 
> > "Subscribe to Presence" in order to enable BLF monitoring.
> > 
> > This would also effectively remove the need to have distinct Call 
> > Pickup and Park Retrieve codes.
> >   
> > I've also mimicked a "multiple call" park orbit with a plain user 
> > extension.  The Polycom sets are even smart enough to handle this 
> > well.  i.e. When you hit the "ringing" BLF button during a Transfer 
> > the set knows to dial the contact, instead of attempting a Call 
> > Pickup.
> > 
> >  
> > 
> > Does anyone see problems, refinements, or anything that 
> I've missed?  Thanks.
> 
> Sounds like the right user experience.
> 
> If it can be implemented in a way that doesn't break other 
> phones, this sounds like a good approach.

Arjun pointed out that some phones in confirmed dialogs may not obey a
REFER with Replaces, as item 2 above depends on.  

I can eliminate this dependency and reduce the amount of work with an
alternate approach: 

2. For b) to happen the Call Pickup redirector will need a slight change
to its handling of Call Pickups executed on Park Orbits.  When a Call
Pickup is executed the redirector must determine if the target is a Park
Orbit.  If so, then the redirector will proceed with an Un-Park, instead
of a Call Pickup.  (Including using the existing FIFO un-park order
functionality.)

In this way we are simply mapping the *78 that the Polycom dials into
the *4 we want for the existing Un-Park sequence to occur.  

This should work seamlessly (in theory) on any phones that can already
un-park and be un-parked.

Thanks.

-Paul

_______________________________________________
sipx-dev mailing list
[email protected]
List Archive: http://list.sipfoundry.org/archive/sipx-dev
Unsubscribe: http://list.sipfoundry.org/mailman/listinfo/sipx-dev

Reply via email to