Re: [Ecrit] FW: I-D Action:draft-rosen-ecrit-premature-disconnect-rqmts-02.txt

[Ted Hardie <hardie at qualcomm.com>]

The document says:

The PSTN has a feature
   available, "Called Party Hold" (CPH) which is used in some
   jurisdictions to meet this requirement.


It was my understanding that this feature was no longer universally
available in the PSTN, not that it was universally available and
not used in all jurisdictions.  Have I misunderstood this?

As the document notes, it is not available in wireless networks,
which are the majority of initiating networks for emergency
calls in at least some jurisdictions.  I believe that this makes
this statement:

   Some jurisdictions are
   desirous of maintaining their current PSAP call disconnect control
   capability, while other jurisdictions would like to regain access to
   those capabilities.

only partly a true representation of the current situation.

The document says:

   When PD-1 is enforced, the call taker must be able to cause
      alerting to the caller which has attempted to prematurely
      disconnect from the emergency call.
     Rationale:  The caller believes they have disconnected.  The
      ability to alert is needed to encourage the caller to reconnect.

Previous parts of the document have described "ringback"
as distinct from call back.  Is this "ringback" in requirements
form?

I understand that the folks working on this feel strongly about
this, but this document is neither complete nor persuasive.  As
it stands, it is missing all discussion of the negotiation phase of
this, which has been pointed out repeatedly as a necessary item.
Without the requirements there describing what the negotiation
of this function must do, the security considerations are paper
thin and the rest of it gives no real sense of the overarching system.

Can a phone say "no" in this negotiation, so that the PSAP is
told it is not being surrendered call control?  Without answers
to questions like this, the document can't be evaluated well.

I also believe that to be truly persuasive the document would
have to cover the UI issues and the multi-line issues which have
been raised repeatedly.  As it is, I read it to describe a system
that could just as easily be handled entirely by UI and client
side mechanisms, with no network requirements at all for
abandonment. 

I do not support this document becoming a working group item
as it currently stands. 
			regards,
				Ted Hardie

_______________________________________________
Ecrit mailing list
Ecrit at ietf.org
https://www.ietf.org/mailman/listinfo/ecrit