Re: [Sipping-emergency] Emergency Scenarios Document

[Tom Taylor <[e-mail]>]

About time I finished this and got the document submitted.  My remarks prefaced with 
[PTT].  I had help from Jim McEachern.

Henning Schulzrinne wrote:

>>       centresin the PSTN, the scenarios involving stationary or 
>> nomadic SIP 
> 
> 
> spelling
[PTT] Fixed
> 
>>       phones are described and then alternative strategies for 
>> supporting       the information flows for the scenarios needed to 
>> meet these       requirements are examined.  The analysis presented 
>> here is a       necessary step in defining the SIP specific 
>> requirements for invoking       emergency assistance and understanding 
>> the level of functionality       that can be provided by existing SIP 
>> devices and network elements and       that which requires further 
>> development. 
> 
> 
>>       in the European Market in general, or 000 in Australia.  (Globally 
> 
> 
> I thought it was European Community?
[PTT] Fixed.
> 
>>       there are about 60 different emergency services numbers in use 
>> [4].)  
> 
> 
>>       This document uses the term SIP phone to denote a device which 
>> plugs       into a wired network.  This can be a either a physical 
>> phone device       or a PC running a software based SIP client (often 
>> referred to as a       SIP softphone).  These devices can be 
>> stationary or nomadic. 
> 
> 
> Why a wired network? I think the issues for a PDA or laptop or even a 
> phone (via an Ethernet-to-802.11 bridge) that connects via an 802.11 
> hotspot are fairly similar.

[PTT] I've rewritten this section to focus on the distinction between stationary, 
nomadic, and mobile, with little mention of access technology.  This should also 
answer your next comment.
> 
>>
>>       Scenarios involving cellular phones using SIP signalling are not 
>>       necessarily considered, because the current expectation is that 
>> they       use the emergency service solutions devised for cellular 
>> phones in       general. However, it is anticipated that many aspects 
>> of a general       SIP signaling solution could apply to those 
>> scenarios.  The       requirements for Wi-Fi enabled devices (e.g. a 
>> SIP softphone with Wi-
>>       Fi access, a Wi-Fi phone etc.) using SIP signalling are 
>> considered       from the perspective that these devices can be 
>> represented as nomadic       SIP users and thus addressed by several 
>> of the scenarios described in       this document. 
> 
> 
> I would not make the distinction up front.
> 
>>       "legacy" is used to refer to "Enhanced 911" functionality.) 
> 
> 
> You mean 'basic 911'?

[PTT]  I mean E911.  A number of the requirements come up only in that context.

> 
>>       the difference in medium and the fact that a different number 
>> may be       dialed to initiate the call.  
> 
> 
> I think all single-number systems support TDD on the same number; that's 
> at least true for 911, as far as I know.

[PTT] Qualified to say that it may be different in some countries (Australia, for sure).
> 
>>       (1) The caller requires an easily-remembered, 
>> location-independent           means of identifying an emergency 
>> call.  Moreover, it must be           possible to indicate whether 
>> this is a voice call or one to be           directed to a text ECC. 
> 
> 
> I don't think so. I think the 911 destination listens for the TDD modem 
> tone and then switches to a TDD. This is not by number.
> 
>>
>>       (5) The ECC operator can stay in touch with the caller, even 
>> though           the telephone set goes on hook for an intervening 
>> period.  An           example of when this is useful is if the caller, 
>> feeling unwell,           accidentally drops the telephone back onto 
>> its cradle, but then           is able to retrieve it. 
> 
> 
> As discussed previously, this is not a universal feature. This is more 
> of an urban legend at this point :-)

[PTT] Qualified to say it depends on the country.  It's required in Canada.
> 
>>
>>       (6) Emergency calls are given priority handling in call 
>> processing           and high quality media connections.  (The quality 
>> of media           connections is outside the scope of this document.) 
> 
> 
> Again, I don't think this is generally true, at least in the US.

[PTT] Qualified to say it depends on the country.  It's done in the UK.
> 
> 
>>
>>       (7) The caller can be held accountable for the call, to dissuade 
>>           callers from misuse of the system. 
> 
> 
> If you ignore pay phones :-)
> 
>>
>>    3.3   Legacy System Operation
>>       In the legacy network, the public telephone system operator is 
>>       responsible for maintaining the mapping database.   
> 
> 
> This is imprecise - at least in the US, there is no 'the' telephone 
> system operator. It is generally the ILEC, but I think it would be 
> unhelpful to imply that we live in pre-1984/pre-1996 land.

[PTT] Changed to operators (plural).
> 
>>       the location.  Alternatively, various locations can be 
>> identified       within the business and assigned a pseudo directory 
>> number (DN), 
> 
> 
> Even for phones with DID, extensions are usually not visible to the 
> outside. (I've tried it - all of our extensions show up as 7000. We have 
> DID in 7000 to 7199.)

[PTT] Right.  It requires additional arrangements.  When it is done, blocks of 
numbers are allocated to specific trunks.  For each outward call, the telco checks 
the calling line identification against the trunk it is arriving on and accepts if 
if that check passes.  My text is actually correct, but I've expanded upon the DID 
case a bit.
> 
>>       A special arrangement at the switch serving the ECC allows the 
>>       emergency operator to hold open the voice path back to the 
>> caller as       long as necessary, even if the caller goes on-hook.  
>> It should be       noted that this functionality is not supported for 
>> ISDN. Thus, the       operator can also ring the caller's line in this 
>> situation or others       where it might be necessary.  
> 
> 
> I would qualify this as "In analog switches, ....". Most switches today 
> are effectively ISDN (or equivalent), so this feature is likely only 
> available to a tiny fraction of end systems.

[PTT] You're getting confused between digital switches and analogue vs. ISDN access. 
  In the US and Canada, most lines are analogue, even though practically all the 
switches are digital.  In Europe, ISDN access is the dominant mode.
> 
>>    4.1   A General Look At The Problem 
> 
> 
> at the
> 
>>       signalling passes through other intelligent elements -- SIP 
>> proxies       and a gateway -- before reaching the telephone network.  
>> The SIP 
> 
> 
> A SIP proxy isn't mandatory.

[PTT] Agreed, and I've always had that in mind.  I added "perhaps" after the proxies.
> 
> [More in next message.]
> 
> 
> 
> _______________________________________________
> Sipping-emergency mailing list
> [e-mail]
> https://www1.ietf.org/mailman/listinfo/sipping-emergency
> 

_______________________________________________
Sipping-emergency mailing list
[e-mail]
https://www1.ietf.org/mailman/listinfo/sipping-emergency