Re: [Ecrit] R: country specific emergency URNs

[Henning Schulzrinne <[e-mail]>]

For cross-national MVNOs, routing to a service that does not exist in the
country you are visiting is not helpful (or even possible). Thus, if you
define a country-specific service that exists only in, say, Italy, and then
visit Austria, the call will have to fail. In the SOS model, you get a
close approximation, which seems much more user friendly.

As I have stated repeatedly, and you seem to continue to ignore, is that
the label is an internal protocol label where the only requirement is that
there is a unique 1-1 mapping within one country from number to label. This
requirement exist even if you create country-specific labels unless your
"label" is sos:fr.147, i.e., you directly encode the number. This is
clearly a bad idea.

I disagree that this requires a "consolidation of emergency service
requirements". It requires a listing of services, which I have again asked
for again and again.

As I have also tried to explain several times, we only have to make sure
two things are true:

(1) The description is precise enough so that it is clear in each country
whether it matches an existing service. If a newly-discovered service does
not seem to match an existing one, a new label is created. The description
does not have to be a legal one or cover every aspect of operation (which
changes over time anyway). In all the cases we have discussed, this seems
relatively straightforward.

(2) Within each country, all existing numbers have to map to unique URNs.

It is very difficult to make progress if you (and your colleagues) insist
that something is not possible, but then make it impossible to make
progress by not providing the necessary information. It would be most
helpful if you could engage constructively rather than insisting on a
solution that violates the spirit and design principles of the architecture.

I continue to look forward to starting the list of services we should
discuss.

Henning


On Tue, Jul 11, 2017 at 6:00 AM, Procopio Roberto <
[e-mail]> wrote:

> Dear All,
>
>
>
> I share the points raised by Vasil. In some of the emails below I red that
> a way forward could be to identify a reasonable mapping between an
> emergency service and the related sos urns. I don’t think that this is a
> good way forward. The reason is that this way forward requires a
> consolidation of emergency service requirements among different countries
> and this requires a deep understanding of the emergency services definition
> in each country.
>
>
>
> Simply looking at the emergency services collected in the table shared by
> Vasil, it can be understood that services having the same name can have
> different requirements in different countries. This mistmatch generates
> problems in the emergency service operation as showed in the example raised
> by Vasil (e.g. roaming, borders). If we look at future possible deployments
> of mobile networks, we can also find other scenarios (such as
> cross-national MVNO) where is crucial to identify exactly one specific
> emergency service in one country.
>
>
>
> Therefore in my opinion is very important to guarantee the flexibility in
> the naming definition of the sos urns. As Vasil said, this flexibility was
> one of the reasons behind the proposal of the country specific emergency
> URN.
>
>
>
> BR
>
> Roberto
>
>
>