w_smith@wookie.enet.dec.com (Willie Smith, LTN Components Eng.) (11/16/89)
This is not an attempt to start the CLID wars and associated ratholes
that so clogged this Digest in the past few months. On a more
technical bent:
I got the Bellcore Technical References (thanks to the reader who
sent in the pointer!) for:
CLASS (SM) Feature: Calling Number Delivery TR-TSY-000031
which is the technical spec on how CND (widely discussed in
TELECOM as Calling Line ID) works, and;
SPCS/Customer Premises Equipment Data Interface TR-TSY-00030
which is the low level details of how the data gets over the
phone lines from the telco to your house.
They are about $25 each from Bellcore (201) 699-5800 [there's also a
toll-free number, but I don't seem to have it handy], and are well
worth the price, as they answer a lot of the questions that come up in
this Digest.
Anyway, I've got a few observations and questions:
1) The basic interface is a 1200 baud FSK data stream between the first and
second rings. The interface is not supposed to work in an off-
hook state, though it might be interesting to see if they really
work that way. All kinds of spoofing might be possible if the
box depends on the fact that the CO will not send data when the
phone is off-hook...
2) You need a computer to decode the data, as there's a fair amount of
complexity, including a checksum and multiple messages. The
interface is kind of general, so you could do several different
things with it. One use that seems built into the spec is
message-waiting (as used in hotels). Does anyone know any other
uses? Is it worth making a transciever (so you can talk back to
the telco) or should you only bother with a reciever if you are
going to 'roll your own'? If you have a computer, you could
build the interface (a simple FSK modem) pretty cheap! Does anyone
know if any of the commercial units have an RS-232 output?
3) CND (CLID) is available in two different flavors, "subscription" is
the usual one, where you sign up for it once, CND is used for all
calls into your house, and you are billed (I assume) monthly. The
other option is "usage-sensitive", where you still have to sign up
for the service (and I imagine pay a connection fee), but you can
turn it on and off at will by dialing *65 or *85 (on and off
respectively, numbers may vary in your area). This service appears
to be billed on a 'number of CNDs sucessfully delivered'. Why would
anyone use the "usage-sensitive" option?
4) The calling directory number, "if available and can be disclosed",
is transmitted to you. See TR-TSY-000391 for details on blocking.
I'm not that concerned about it, but if you are, spend the $25 or so
and get the info from the source. Then maybe we can have a reasoned
discussion with facts and such. Nawww.... :+)
5) Here's an interesting option I don't remember hearing discussed. The
calling party can dial a 4-digit (or longer) PIN that will be displayed
instead of the calling DN. This brings up some rather interesting
questions on spoofing or hiding your number, does anyone know if the
commonly used boxes differentiate between DN (directory number) and
PIN? I believe it's in the low-level protocol, but it would appear
to be implementation-dependent as to how it's displayed.
6) Another neat future use mentioned in passing is an interface to directory
assistance or another database to provide calling party name instead of
calling DN. Yes the low level spec will pass ASCII, do any of the
current boxes allow display of alphanumerics?
7) The requirement for a customer initiated testing number is "desirable".
This would allow the customer to dial a special number, hang up, and
get a series of test transmissions (display each digit in each
position, etc). Do any of the CND trials out there provide such a
service?
8) The various call forwarding options send the originating number to the
far end. In other words, a call from A to B that gets forwarded to
C should deliver A's number to C.
9) Automatic Callback is the only instance I found where the CND shows a
'private' number. This section is a but hard to describe, but
basically: [See TR-TSY-000215 for more information]
If I call someone with an unlisted number (I already know
their number!) then fire up AC, when the connection
is made and I get a special ring signal, that
number shows up on _my_ box (so I know who I'm
calling back).
If I didn't put the number in in the first place and it's
a 'private' number, then I'll get a privacy
indication.
10) In a Centrex, you may only get the extension number on your box.
Anyway, they seem to have thought this all out pretty carefully, and
it's a very well written set of documents, so if you have more
questions, you can try Emailing me, but the definitive answers are
directly available. Kind of pricey if you get all the associated
documents, but that's life.
I called New England Telephone to try to figure out when this feature
would be available in my area, but the customer service rep didn't
know what I was talking about. In fact, references to Bellcore didn't
seem to ring any bells with her either (sorry :+). She went away for
a while doing some research on the question and came back with the
response that they had no idea when that might be available on (508)
369-xxxx, but if I call back when the strike is over and they get back
on their feet, they might have a better idea....
This has nothing to do with my employer!
Willie Smith
w_smith@wookie.enet.dec.com
w_smith%wookie.enet.dec.com@decwrl.dec.com
{Usenet!Backbone}!decwrl!wookie.enet.dec.com!w_smithnogeea@asuvax.EAS.ASU.EDU (Allen Nogee) (11/17/89)
In article <telecom-v09i0515m01@vector.dallas.tx.us>, w_smith@wookie.enet. dec.com (Willie Smith, LTN Components Eng.) writes: > 1) The basic interface is a 1200 baud FSK data stream between the first and > second rings. The interface is not supposed to work in an off- > hook state, though it might be interesting to see if they really > work that way. All kinds of spoofing might be possible if the > box depends on the fact that the CO will not send data when the > phone is off-hook... The hardware I designed for our switch (GTE) does work on or off hook. So do the boxes I've tested. Whether the software will allow it is another question. I guess it all depends if you mind hearing FSKT tones in your ear if you pick up your phone too fast. > 6) Another neat future use mentioned in passing is an interface to directory > assistance or another database to provide calling party name instead of > calling DN. Yes the low level spec will pass ASCII, do any of the > current boxes allow display of alphanumerics? The future for ASCII is there but the boxes I've tested couldn't do it. I fact, The AT&T box I tested couldn't even display more than 7 digits. The only message type supported at present is 4, digits only.
deej@bellcore.bellcore.com (David Lewis) (11/18/89)
In article <telecom-v09i0515m01@vector.dallas.tx.us>, w_smith@wookie.enet.dec. com (Willie Smith, LTN Components Eng.) writes: > I got the Bellcore Technical References (thanks to the reader who > sent in the pointer!) for: > CLASS (SM) Feature: Calling Number Delivery TR-TSY-000031 > which is the technical spec on how CND (widely discussed in > TELECOM as Calling Line ID) works, and; > SPCS/Customer Premises Equipment Data Interface TR-TSY-00030 > which is the low level details of how the data gets over the > phone lines from the telco to your house. One thing to keep in mind is that Bellcore TRs are essentially written to the industry to provide requirements for manufacturers; the manufacturers are free to implement products in exact compliance to the Bellcore TRs or not, as they see fit. In other words, the TRs specify how Bellcore says the product/service/system *should* work, which may or may not be the same as the way it *does* work. > They are about $25 each from Bellcore (201) 699-5800 [there's also a > toll-free number, but I don't seem to have it handy] 1-800-521-CORE (2673) > and are well > worth the price, as they answer a lot of the questions that come up in > this Digest. Gosh, it's nice to know someone appreciates us... > 3) CND (CLID) is available in two different flavors, "subscription" is > the usual one, where you sign up for it once, CND is used for all > calls into your house, and you are billed (I assume) monthly. The > other option is "usage-sensitive", where you still have to sign up > for the service (and I imagine pay a connection fee), but you can > turn it on and off at will by dialing *65 or *85 (on and off > respectively, numbers may vary in your area). This service appears > to be billed on a 'number of CNDs sucessfully delivered'. Why would > anyone use the "usage-sensitive" option? This is one place where the above comment comes in. Bellcore wrote the requirements to permit either subscription or usage-sensitive billing. Equipment vendors may or may not have built their products to permit either subscription or usage-sensitive billing; I don't honestly know. The local telco then has the option to sell the service as either subscription or usage-sensitive; I don't know if any telcos have offered it as usage-sensitive. The point of the requirements is to define the technical capabilities which should be available, not the choice of how to sell them -- which is a business decision, which Bellcore emphatically does *not* recommend anything about. (In fact, we have standard disclaimer notices which say, essentially, "we have nothing to do with decisions about pricing, sales, or anything else that could draw out the nasty C-word, collusion...") > 5) Here's an interesting option I don't remember hearing discussed. The > calling party can dial a 4-digit (or longer) PIN that will be displayed > instead of the calling DN. Again -- this is in the TR, but may not be implemented currently or offered currently. Calling Number Delivery Blocking is also in the TR, but I don't know if any telco is offering it. > 6) Another neat future use mentioned in passing is an interface to directory > assistance or another database to provide calling party name instead of > calling DN. Ditto above. > 7) The requirement for a customer initiated testing number is "desirable". > This would allow the customer to dial a special number, hang up, and > get a series of test transmissions (display each digit in each > position, etc). Do any of the CND trials out there provide such a > service? Things which are "desirable" are even less likely to have been implemented than things which are "required"... > Anyway, they seem to have thought this all out pretty carefully, and > it's a very well written set of documents, so if you have more > questions, you can try Emailing me, but the definitive answers are > directly available. Kind of pricey if you get all the associated > documents, but that's life. Gee thanks... And just think -- every dollar you pay Bellcore for documents is one dollar less that has to come out of various people's phone bills to pay my salary!! > I called New England Telephone to try to figure out when this feature > would be available in my area, but the customer service rep didn't > know what I was talking about. In fact, references to Bellcore didn't > seem to ring any bells with her either (sorry :+). Hey, don't sweat it -- some telco people don't even know (or particularly care) which holding company they're owned by... > This has nothing to do with my employer! It does, however, have everything to do with mine... David G Lewis ...!bellcore!nvuxr!deej (@ Bellcore Navesink Research & Engineering Center) "If this is paradise, I wish I had a lawnmower."
scott@gatech.edu (Scott Barman) (11/21/89)
In article <telecom-v09i0519m02@vector.dallas.tx.us> nvuxr!deej@bellcore. bellcore.com (David Lewis) writes: >One thing to keep in mind is that Bellcore TRs are essentially written >to the industry to provide requirements for manufacturers; the >manufacturers are free to implement products in exact compliance to >the Bellcore TRs or not, as they see fit. In other words, the TRs >specify how Bellcore says the product/service/system *should* work, >which may or may not be the same as the way it *does* work. Caveat on the above (since we've been through these problems here): If you are with a company wanting to do business with the RBOCs, then you should really consider implementing your product/etc. to as close to exact compliance as possible. The question will come up asking if your product is not only Bellcore compliant but even certification of that compliance may be required. If you are not Bellcore compliant, then you will need to demonstrate a clear path for that compliance. I know that we lost a sale to (among a few other things) to a slight lack of Bellcore compliance. Now if you are dealing with some of the smaller carriers, this is not as much of an issue as long as you are willing to comply to their requirements and come close to the Bellcore standards. Sometimes, if you can come up with something unique and first (like a 3/1 cross-connect switch), you can get away with not being exactly Bellcore compliant (however, we still have to show a plan for future compliance). Based on things I've heard, I would suggest that if you are designing a product, the physical standards are very important--especially for the RBOCs. Standards like rack sizes and depth are very important because of the floor plans of many of the central offices already in existance. DISCLAIMER: I am not in charge of compliance here, nor do I speak for the company and I am NOT an expert on Bellcore compliance. What I present here are just impressions while working on a product to be sold to the telecom industry. scott barman {gatech, emory}!nastar!scott
schulman@umd5.umd.edu (Marty Schulman) (11/27/89)
Regarding Incoming Caller Identification (or whatever you call the service that transmits the number of the originating telephone between the 1st and 2nd rings): 1. Could somebody offer a brief list of telephones which will *not* be reported correctly (i.e. will people using PBX's have their number displayed properly? What about people calling long distance through various long distance companies? Are there other exceptions within the local dialing area? And do I get just the last seven digits, or area code too?) 2. With Call Waiting, my phone conversations are interrupted with two brief beeps when somebody's trying to call me. Suppose the telephone company wanted to replace those smooth-sounding beeps with FSK data bursts containing the number of the person calling. Would it entail an extensive modification to their equipment? Given the amplitude levels required by the decoding equipment, would the new "beeps" be too loud or "rude"? And if I was in the middle of a sentence when the FSK tones arrived, would the decoder be able to read them? Just curious. Thanks for any and all responses.