asterisk users - Sep 2008 - Seemingly easy question: NPA/NXX

Question: 
How does the local Telco know you're done dialing a seven digit number? 
Easy you may say:  If your dial string begins with 1, the parser expects
11 digits total, otherwise seven, 011 is international.

The reason suspect it's more complex is that: 
1) International numbers can vary widely in length and 
2) Our local analog Telco will route a ten digit NANP numbers with no
leading 1 and with no terminator--seemingly instantly

Obviously this could be done with 'timeouts'--implicitly
'sending'
after a delay.  But it works so well I suspect there's more logic in
there.   For example I have dozens of ATA's provisioned with timeouts,
and I find it difficult or impossible to replicate the Telco dialing
experience (Either the delay is too long, or you have frequent 'reorder'
tones because it 'sent' before you were finished).

Therefore I assume that there is something more 'fancy' going on.  Can
someone validate, debunk or clarify this?

Theory 1
Is it all done with timeouts, but they're CONDITIONAL timeouts.
i.e. give a LONG timeout if the number:
-did not start with a 1 and is still shorter than 7 digits, 
-started with a 1 and is still shorter than 11 digits
-started with a 011 and is shorter than the theoretical international
minimum lenght

Theory 2
As you know, a few years ago the 2nd digit of the NPA was always 1 or 0.
 Therefore the switch could easily determine(without the leading 1) if
your first three digits were an NPA or just an NXX (exchange).  They
were nationally unambiguous.   Now that's no longer true.  STILL, it 
could be possibleto consider all known valid NPA's and exchanges so they
can determine via context what you're trying to do, and thereby optimize
the dialing experience?  

Can anyone speak to this?  I would very much appreciate any knowledgable
input.

-Karl

International numbers are variable length, so the timeout applies for those.

North American "National" numbers are a fixed length.

Generally, the phone company will collect 7, 10 or 11 digits for North  
American numbers.

For example, I live in Minneapolis, MN.

My number is 612-xxx-xxxx.

I have free calling to 612, 651, 952, 763 and a few numbers in 507 and 320.

If I dial 1, the phone company will collect 10 more digits.  (The call  
may or may not go through if I dial 1+ a 10-digit local number  
depending on the carrier.  MN regulations prohibit charging for local  
calls dialed as toll)

If I dial 612, 651, 952, 763, 507 or 320, the phone company will wait  
for the remaining 7 digts as there are no numbers within area code 612  
that start with those digits.  Anything else will only be collected as  
7 digts and assumed to be 612.

Because of that, I can't dial a california number (for example),  
without dialing it as 1+.

I wouldn't call it "fancy", the phone company just knows what is a
valid local number for you.

Making a digit map in an ATA isn't that hard, you just need to think  
about what you want it to do.  If you want to permit 10 digit dialing  
without the 1+ for long distance *and* support 7 digit local dialing,  
you'll need a timeout.

There are also the N11 numbers, which of course should stop collecting  
after the second "1".

--Shane


Quoting Karl Fife <asterisk-users at kfife.mailworks.org>:
> Question:
> How does the local Telco know you're done dialing a seven digit number?
> Easy you may say:  If your dial string begins with 1, the parser expects
> 11 digits total, otherwise seven, 011 is international.
>
> The reason suspect it's more complex is that:
> 1) International numbers can vary widely in length and
> 2) Our local analog Telco will route a ten digit NANP numbers with no
> leading 1 and with no terminator--seemingly instantly
>
> Obviously this could be done with 'timeouts'--implicitly
'sending'
> after a delay.  But it works so well I suspect there's more logic in
> there.   For example I have dozens of ATA's provisioned with timeouts,
> and I find it difficult or impossible to replicate the Telco dialing
> experience (Either the delay is too long, or you have frequent
'reorder'
> tones because it 'sent' before you were finished).
>
> Therefore I assume that there is something more 'fancy' going on. 
Can
> someone validate, debunk or clarify this?
>
> Theory 1
> Is it all done with timeouts, but they're CONDITIONAL timeouts.
> i.e. give a LONG timeout if the number:
> -did not start with a 1 and is still shorter than 7 digits,
> -started with a 1 and is still shorter than 11 digits
> -started with a 011 and is shorter than the theoretical international
> minimum lenght
>
> Theory 2
> As you know, a few years ago the 2nd digit of the NPA was always 1 or 0.
>  Therefore the switch could easily determine(without the leading 1) if
> your first three digits were an NPA or just an NXX (exchange).  They
> were nationally unambiguous.   Now that's no longer true.  STILL, it
> could be possibleto consider all known valid NPA's and exchanges so
they
> can determine via context what you're trying to do, and thereby
optimize
> the dialing experience?
>
> Can anyone speak to this?  I would very much appreciate any knowledgable
> input.
>
> -Karl
>
> _______________________________________________
> -- Bandwidth and Colocation Provided by http://www.api-digital.com --
>
> AstriCon 2008 - September 22 - 25 Phoenix, Arizona
> Register Now: http://www.astricon.net
>
> asterisk-users mailing list
> To UNSUBSCRIBE or update options visit:
>    http://lists.digium.com/mailman/listinfo/asterisk-users
>


----------------------------------------------------------------

I don't see where it is difficult to figure out.

  First of all, system keeps looking up on the table as user dial each
number.

  When number starts with 1, expect USA. When number doesn't start with
either 1 nor 0, expect USA too.

  When number starts with 011, and as country code and city code is
identified, expect as many numbers as determined by country+city code
(once you know country and city code, you know how many local digits to
expect)

CS
>>>Question:
How does the local Telco know you're done dialing a seven digit number?
Easy you may say:  If your dial string begins with 1, the parser expects
11 digits total, otherwise seven, 011 is international.

The reason suspect it's more complex is that:
1) International numbers can vary widely in length and
2) Our local analog Telco will route a ten digit NANP numbers with no
leading 1 and with no terminator--seemingly instantly

Obviously this could be done with 'timeouts'--implicitly
'sending'
after a delay.  But it works so well I suspect there's more logic in
there.   For example I have dozens of ATA's provisioned with timeouts,
and I find it difficult or impossible to replicate the Telco dialing
experience (Either the delay is too long, or you have frequent 'reorder'
tones because it 'sent' before you were finished).

Therefore I assume that there is something more 'fancy' going on.  Can
someone validate, debunk or clarify this?

Theory 1
Is it all done with timeouts, but they're CONDITIONAL timeouts.
i.e. give a LONG timeout if the number:
-did not start with a 1 and is still shorter than 7 digits, -started with
a 1 and is still shorter than 11 digits -started with a 011 and is shorter
than the theoretical international minimum lenght

Theory 2
As you know, a few years ago the 2nd digit of the NPA was always 1 or 0.
 Therefore the switch could easily determine(without the leading 1) if
your first three digits were an NPA or just an NXX (exchange).  They
were nationally unambiguous.   Now that's no longer true.  STILL, it
could be possibleto consider all known valid NPA's and exchanges so they
can determine via context what you're trying to do, and thereby optimize
the dialing experience?

Can anyone speak to this?  I would very much appreciate any knowledgable
input.

-Karl

>  When number starts with 011, and as country code and city code is
> identified, expect as many numbers as determined by country+city code
> (once you know country and city code, you know how many local digits to
> expect)
... except in some countries, the phone numbers vary in length in the
same city. Say in Hamburg, Germany, your number can be as short as 5
digits or as long as 10. You really have no way of knowing.

Luki

At 09:29 AM 9/22/2008, you wrote:
>... except in some countries, the phone numbers vary in length in the
>same city. Say in Hamburg, Germany, your number can be as short as 5
>digits or as long as 10. You really have no way of knowing.

The unanswered part of that, is this? Can 5 digit number, say, 12345, 
be the beginning part of a 10 digit number, say, 1234567890?

If so than you must use time outs, if not than a dial plan can handle it.

Ira

Ira schrieb:
> At 09:29 AM 9/22/2008, you wrote:
>>... except in some countries, the phone numbers vary in length in the
>>same city. Say in Hamburg, Germany, your number can be as short as 5
>>digits or as long as 10. You really have no way of knowing.
> 
> 
> The unanswered part of that, is this? Can 5 digit number, say, 12345, 
> be the beginning part of a 10 digit number, say, 1234567890?
> 
> If so than you must use time outs, if not than a dial plan can handle it.
I think so.
http://www.bundesnetzagentur.de/enid/7d0a4195f3524c5b04b7870e1f60635d,0/Ortsnetze/Struktur_und_Ausgestaltung_des_Nummernbereichs_fuer_Ortsnetzrufnummern_1gq.html
http://www.bundesnetzagentur.de/enid/7d0a4195f3524c5b04b7870e1f60635d,0/Ortsnetze/Verzeichnisse_1gg.html

But obviously this has nothing to do with the NANPA etc.

   Philipp Kempgen

-- 
http://www.das-asterisk-buch.de  -  http://www.the-asterisk-book.com
Amooma GmbH - Bachstr. 126 - 56566 Neuwied  ->  http://www.amooma.de
Gesch?ftsf?hrer: Stefan Wintermeyer, Handelsregister: Neuwied B14998
--

Philipp Kempgen schrieb:
> Ira schrieb:
>> At 09:29 AM 9/22/2008, you wrote:
>>>... except in some countries, the phone numbers vary in length in
the
>>>same city. Say in Hamburg, Germany, your number can be as short as 5
>>>digits or as long as 10. You really have no way of knowing.
>> 
>> 
>> The unanswered part of that, is this? Can 5 digit number, say, 12345, 
>> be the beginning part of a 10 digit number, say, 1234567890?
>> 
>> If so than you must use time outs, if not than a dial plan can handle
it.
> 
> I think so.
>
http://www.bundesnetzagentur.de/enid/7d0a4195f3524c5b04b7870e1f60635d,0/Ortsnetze/Struktur_und_Ausgestaltung_des_Nummernbereichs_fuer_Ortsnetzrufnummern_1gq.html
>
http://www.bundesnetzagentur.de/enid/7d0a4195f3524c5b04b7870e1f60635d,0/Ortsnetze/Verzeichnisse_1gg.html
Strange session IDs.

http://www.bundesnetzagentur.de/enid/Ortsnetze/Struktur_und_Ausgestaltung_des_Nummernbereichs_fuer_Ortsnetzrufnummern_1gq.html
http://www.bundesnetzagentur.de/enid/Ortsnetze/Verzeichnisse_1gg.html


   Philipp Kempgen

-- 
http://www.das-asterisk-buch.de  -  http://www.the-asterisk-book.com
Amooma GmbH - Bachstr. 126 - 56566 Neuwied  ->  http://www.amooma.de
Gesch?ftsf?hrer: Stefan Wintermeyer, Handelsregister: Neuwied B14998
--

Philipp Kempgen schrieb:
> Philipp Kempgen schrieb:
>> Ira schrieb:
>>> At 09:29 AM 9/22/2008, you wrote:
>>>>... except in some countries, the phone numbers vary in length
in the
>>>>same city. Say in Hamburg, Germany, your number can be as short
as 5
>>>>digits or as long as 10. You really have no way of knowing.
>>> 
>>> 
>>> The unanswered part of that, is this? Can 5 digit number, say,
12345,
>>> be the beginning part of a 10 digit number, say, 1234567890?
>>> 
>>> If so than you must use time outs, if not than a dial plan can
handle it.
>
http://www.bundesnetzagentur.de/enid/Ortsnetze/Struktur_und_Ausgestaltung_des_Nummernbereichs_fuer_Ortsnetzrufnummern_1gq.html
> http://www.bundesnetzagentur.de/enid/Ortsnetze/Verzeichnisse_1gg.html
One of the documents says that if your number is 123456 (for
example) you are allowed to use 123456x, 123456xx etc.

OTOH it also says that the carriers are not obligated to make
sure the additional digits make it to your line.


   Philipp Kempgen

-- 
http://www.das-asterisk-buch.de  -  http://www.the-asterisk-book.com
Amooma GmbH - Bachstr. 126 - 56566 Neuwied  ->  http://www.amooma.de
Gesch?ftsf?hrer: Stefan Wintermeyer, Handelsregister: Neuwied B14998
--

----- "Karl Fife" <asterisk-users at kfife.mailworks.org>
wrote:
> Theory 1
> Is it all done with timeouts, but they're CONDITIONAL timeouts.
> i.e. give a LONG timeout if the number:
> -did not start with a 1 and is still shorter than 7 digits, 
> -started with a 1 and is still shorter than 11 digits
> -started with a 011 and is shorter than the theoretical international
> minimum lenght
> 
> Theory 2
> As you know, a few years ago the 2nd digit of the NPA was always 1 or
> 0.
>  Therefore the switch could easily determine(without the leading 1)
> if
> your first three digits were an NPA or just an NXX (exchange).  They
> were nationally unambiguous.   Now that's no longer true.  STILL, it 
> could be possible to consider all known valid NPA's and exchanges so
> they
> can determine via context what you're trying to do, and thereby
> optimize
> the dialing experience?  
> 
> Can anyone speak to this?  I would very much appreciate any
> knowledgable input.
Well, my input is knowledgeable, though not authoritative.

Yes, each NANP switch actually does have a routing table loaded locally
(they call them "translations") that tells it where to route calls for
each and every valid NPA-NXX in the NANP, and this could be used to
authenticate the first 3/6 digits of 7/10/11 digit dialled numbers for
intra-NANP calls, and in fact, I would bet that you're correct that
that's
how they accomplish it.

I have never actually seen live switch code on this, but I think I could
locate some people who have -- but yes, you'll play hell duplicating it
exactly on something with as small a brain as an ATA.

Cheers,
-- jra
-- 
Jay R. Ashworth                   Baylink                      jra at
baylink.com
Designer                     The Things I Think                       RFC 2100
Ashworth & Associates     http://baylink.pitas.com                    
'87 e24
St Petersburg FL USA      http://photo.imageinc.us             +1 727 647 1274

             Those who cast the vote decide nothing.
             Those who count the vote decide everything.
               -- (Josef Stalin)