On 04-Apr-12 12:51, Adam H. Kerman wrote:
Stephen Sprunk wrote:
On 04-Apr-12 03:14, Adam H. Kerman wrote:
Stephen Sprunk wrote:
On 03-Apr-12 14:49, Adam H. Kerman wrote:
Stephen Sprunk wrote:
PBX trunks aren't numbered.

If outbound trunks aren't numbered, how does ANI work?

For a trunk, both called and calling number are explicitly signaled at
the start of each call in either direction.

So they are numbered.

There is not a 1:1 correspondence between trunks and numbers, as there
is with POTS lines. That is what makes them trunks!

I thought it was the bit that the PBX selects it for the outbound call,
possibly on a least cost routing basis.

Huh? I don't see the relevance of that comment.

A PBX may have one trunk or multiple trunks. Multiple trunks may be to
the same carrier or to multiple carriers for redundancy or to take
advantage of cost differences. Multiple trunks to the same carrier will
usually be arranged in a trunk group, with all trunks within a group
being equal.

The customer can use _any_ of their numbers on _any_ of those trunks.

If I have a block of 1000 directory numbers, all of them are routed to
the entire trunk group, so no trunk can be said to have any particular
number. Same if I only have one (high-volume) number: it is routed to
the entire trunk group, so all trunks have the "same" number, which also
means they don't have unique numbers.

Now you're moving the goal posts.

No, I'm attempting to explain to you my original point that trunks are
not numbered, which you still don't seem to get.

I asked a VERY specific question about outbound trunks. If you don't
know the answer, don't post a followup.

I apparently didn't understand your question. Could you rephrase it?

Of course, trunks still have _circuit_ numbers for tracking and billing
purposes, but those are not dialable _directory_ numbers, which is what
we were discussing. For POTS lines, the directory number _is_ the
circuit number.

Why would you go off on a tangent about how outbound trunks aren't
dialable?

No ****!

There is no such thing as an "outbound trunk". Trunks are bidirectional.

(Certain esoteric analog MF trunks are unidirectional, eg. DID and CAMA,
but they aren't used much anymore--and they are unnumbered as well.
Today, "trunk" normally refers to E&M, T1 or PRI circuits.)

ANI is about billing.

No, it is not.

ANI provides an Inward WATS (aka toll-free) customer with the caller's
number so they can do intelligent things with it, like connect them to
the nearest store location. ANI is _not_ guaranteed; the customer gets
billed the appropriate amount even when it's not available. And it's
irrelevant to billing today anyway since customers pay a flat rate per
minute (~$0.01/min) or even a flat monthly rate regardless of the
calling numbers.


The outbound trunk has to have a number, else the call can't be billed.
As far as I know, each outbound trunk has its own number allowing
specific calls to be logged to the specific trunk.

Um, no.

Billing for outbound calls is _not_ based on the calling number (CNIS);
there is one bill is for all calls on the entire trunk group, with the
rate for each call determined by the called number (DNIS). If a calling
number (CNIS) is provided by the customer (it's optional), the carrier
will put it on the bill for the customer's convenience, eg. so they can
assign the cost of each call to the correct department. The carrier
doesn't care about those numbers itself, which is why they allow CNIS of
001-001-0001 if that's what customers send them.

S

--
Stephen Sprunk "God does not play dice." --Albert Einstein
CCIE #3723 "God is an inveterate gambler, and He throws the
K5SSS dice at every possible opportunity." --Stephen Hawking

Stephen Sprunk wrote:
On 04-Apr-12 12:51, Adam H. Kerman wrote:
Stephen Sprunk wrote:
On 04-Apr-12 03:14, Adam H. Kerman wrote:
Stephen Sprunk wrote:
On 03-Apr-12 14:49, Adam H. Kerman wrote:
Stephen Sprunk wrote:

PBX trunks aren't numbered.

If outbound trunks aren't numbered, how does ANI work?

For a trunk, both called and calling number are explicitly signaled at
the start of each call in either direction.

So they are numbered.

There is not a 1:1 correspondence between trunks and numbers, as there
is with POTS lines. That is what makes them trunks!

I thought it was the bit that the PBX selects it for the outbound call,
possibly on a least cost routing basis.

Huh? I don't see the relevance of that comment.

A PBX may have one trunk or multiple trunks. Multiple trunks may be to
the same carrier or to multiple carriers for redundancy or to take
advantage of cost differences. Multiple trunks to the same carrier will
usually be arranged in a trunk group, with all trunks within a group
being equal.

The customer can use _any_ of their numbers on _any_ of those trunks.

The subscriber CANNOT set the number of the trunk that's sent in ANI.
If he could, phone companies would have a hell of a time billing.

If I have a block of 1000 directory numbers, all of them are routed to
the entire trunk group, so no trunk can be said to have any particular
number. Same if I only have one (high-volume) number: it is routed to
the entire trunk group, so all trunks have the "same" number, which also
means they don't have unique numbers.

Now you're moving the goal posts.

No, I'm attempting to explain to you my original point that trunks are
not numbered, which you still don't seem to get.

I didn't ask you about inbound trunks, Stephen. What I asked is quoted
above.

ANI is about billing.

No, it is not.

Yes it is.

ANI provides an Inward WATS (aka toll-free) customer with the caller's
number so they can do intelligent things with it, like connect them to
the nearest store location.

That works only if it's passed along PRI-ISDN or some similar digital
line. Analog? No real-time ANI, but it was logged on the bill.

The outbound trunk has to have a number, else the call can't be billed.
As far as I know, each outbound trunk has its own number allowing
specific calls to be logged to the specific trunk.

Um, no.

Billing for outbound calls is _not_ based on the calling number (CNIS);
there is one bill is for all calls on the entire trunk group, with the
rate for each call determined by the called number (DNIS).

One bill? No ****. You don't believe the carrier logs which trunk was
used regardless of whether it's reported to the subscriber on the bill?
You're wrong.

CNIS isn't the trunk number. For gawd's sake, will you knock off
these tangents?

On 04-Apr-12 13:56, Adam H. Kerman wrote:
Stephen Sprunk wrote:
On 04-Apr-12 12:51, Adam H. Kerman wrote:
Stephen Sprunk wrote:
On 04-Apr-12 03:14, Adam H. Kerman wrote:
Stephen Sprunk wrote:
On 03-Apr-12 14:49, Adam H. Kerman wrote:
Stephen Sprunk wrote:
PBX trunks aren't numbered.

If outbound trunks aren't numbered, how does ANI work?

For a trunk, both called and calling number are explicitly signaled at
the start of each call in either direction.

So they are numbered.

There is not a 1:1 correspondence between trunks and numbers, as there
is with POTS lines. That is what makes them trunks!

I thought it was the bit that the PBX selects it for the outbound call,
possibly on a least cost routing basis.

Huh? I don't see the relevance of that comment.

A PBX may have one trunk or multiple trunks. Multiple trunks may be to
the same carrier or to multiple carriers for redundancy or to take
advantage of cost differences. Multiple trunks to the same carrier will
usually be arranged in a trunk group, with all trunks within a group
being equal.

The customer can use _any_ of their numbers on _any_ of those trunks.

The subscriber CANNOT set the number of the trunk that's sent in ANI.

Yes, Adam, they can--and do. I've been in the telecom industry for
nearly two decades, and we can set ANI/CNIS to whatever we want. The
carriers don't care.

If he could, phone companies would have a hell of a time billing.

Why? They bill WATS calls on a flat rate per minute; the calling number
is now only provided to the customer as a convenience.

They _used to_ care, so they configured their switches to restrict the
ANI/CNIS that customers set on a given trunk to numbers that were routed
_to_ that trunk, but that went away years ago.

If I have a block of 1000 directory numbers, all of them are routed to
the entire trunk group, so no trunk can be said to have any particular
number. Same if I only have one (high-volume) number: it is routed to
the entire trunk group, so all trunks have the "same" number, which also
means they don't have unique numbers.

Now you're moving the goal posts.

No, I'm attempting to explain to you my original point that trunks are
not numbered, which you still don't seem to get.

I didn't ask you about inbound trunks, Stephen. What I asked is quoted
above.

Trunks are bidirectional, Adam.

ANI is about billing.

No, it is not.

Yes it is.

No, it is not.

ANI provides an Inward WATS (aka toll-free) customer with the caller's
number so they can do intelligent things with it, like connect them to
the nearest store location.

That works only if it's passed along PRI-ISDN or some similar digital
line.

Trunks are now almost exclusively PRI or VoIP, both of which include
calling number.

Analog? No real-time ANI,

Analog trunks are almost unheard of today.

but it was logged on the bill.

Some customers like to see that information for statistical purposes,
but most don't care because they're now paying a flat rate per minute.

The outbound trunk has to have a number, else the call can't be billed.
As far as I know, each outbound trunk has its own number allowing
specific calls to be logged to the specific trunk.

Um, no.

Billing for outbound calls is _not_ based on the calling number (CNIS);
there is one bill is for all calls on the entire trunk group, with the
rate for each call determined by the called number (DNIS).

One bill? No ****. You don't believe the carrier logs which trunk was
used regardless of whether it's reported to the subscriber on the bill?
You're wrong.

The carrier doesn't _care_ which trunk was used. Why would they?

CNIS isn't the trunk number. For gawd's sake, will you knock off
these tangents?

These are not tangents, Adam. It is an explanation of complex things
that you simply don't understand and therefore mistakenly attribute to
other, unrelated things.

S

--
Stephen Sprunk "God does not play dice." --Albert Einstein
CCIE #3723 "God is an inveterate gambler, and He throws the
K5SSS dice at every possible opportunity." --Stephen Hawking

Stephen Sprunk wrote:
On 04-Apr-12 13:56, Adam H. Kerman wrote:

The subscriber CANNOT set the number of the trunk that's sent in ANI.

Yes, Adam, they can--and do. I've been in the telecom industry for
nearly two decades, and we can set ANI/CNIS to whatever we want. The
carriers don't care.

Is CNIS what shows up as Caller ID?

If he could, phone companies would have a hell of a time billing.

Why? They bill WATS calls on a flat rate per minute; the calling number
is now only provided to the customer as a convenience.

The call has to be billed to somebody.

ANI provides an Inward WATS (aka toll-free) customer with the caller's
number so they can do intelligent things with it, like connect them to
the nearest store location.

That works only if it's passed along PRI-ISDN or some similar digital
line.

Trunks are now almost exclusively PRI or VoIP, both of which include
calling number.

Analog? No real-time ANI,

Analog trunks are almost unheard of today.

Right. So don't make it seem like ANI was exclusively for billing purposes
on inward WATS.

The outbound trunk has to have a number, else the call can't be billed.
As far as I know, each outbound trunk has its own number allowing
specific calls to be logged to the specific trunk.

Um, no.

Billing for outbound calls is _not_ based on the calling number (CNIS);
there is one bill is for all calls on the entire trunk group, with the
rate for each call determined by the called number (DNIS).

One bill? No ****. You don't believe the carrier logs which trunk was
used regardless of whether it's reported to the subscriber on the bill?
You're wrong.

The carrier doesn't _care_ which trunk was used. Why would they?

Carriers log everything.

CNIS isn't the trunk number. For gawd's sake, will you knock off
these tangents?

These are not tangents, Adam. It is an explanation of complex things
that you simply don't understand and therefore mistakenly attribute to
other, unrelated things.

Statement: CNIS is not the trunk number. I notice you're not
disputing that. You're still arguing, though.