Letting a phone ring forever?
We have a customer who has asked us to let his main line ring "forever" or at least five minutes. As a standard practice we forcefully hang up on an unanswered call at 90 seconds if the customer has turned off voicemail and auto-attendants. It seems wrong to let a line ring forever or even for minutes at a time, and I recall something in the back of my head about a traditional industry-standard limit. Your opinions requested. -- Carlos Alvarez TelEvolve 602-889-3003
On 07/26/2011 02:09 PM, Carlos Alvarez wrote:
We have a customer who has asked us to let his main line ring "forever" or at least five minutes. As a standard practice we forcefully hang up on an unanswered call at 90 seconds if the customer has turned off voicemail and auto-attendants. It seems wrong to let a line ring forever or even for minutes at a time, and I recall something in the back of my head about a traditional industry-standard limit.
Moreover, most calling UACs will not allow a dialog to stay in early state for more than ~3 minutes. Neither will many termination providers; early media (if present) consumes bearer resources without the corresponding per-minute billing. -- Alex Balashov - Principal Evariste Systems LLC 260 Peachtree Street NW Suite 2200 Atlanta, GA 30303 Tel: +1-678-954-0670 Fax: +1-404-961-1892 Web: http://www.evaristesys.com/
Carlos, what is the benefit of the phone ringing for 5 mins? maybe a call queue would fix the overall setup? the most I usually see is 25-30 seconds. -Michael On Jul 26, 2011, at 12:09 PM, Carlos Alvarez wrote:
We have a customer who has asked us to let his main line ring "forever" or at least five minutes. As a standard practice we forcefully hang up on an unanswered call at 90 seconds if the customer has turned off voicemail and auto-attendants. It seems wrong to let a line ring forever or even for minutes at a time, and I recall something in the back of my head about a traditional industry-standard limit.
Your opinions requested.
-- Carlos Alvarez TelEvolve 602-889-3003
_______________________________________________ VoiceOps mailing list VoiceOps at voiceops.org https://puck.nether.net/mailman/listinfo/voiceops
This is kind of a strange customer who makes odd requests and usually refuses to discuss them with us. We are very hands-on about solving business problems, but he won't tell us what he's trying to do or why. He has resisted all suggestions for a better solution, and made us remove the one we had implemented before (answer, tell the caller to please hold, then ring all his phones again). Last week I set a notification to alert me if anyone hung on for the full 90 seconds he's set to now, and predictably, nobody has. I wanted to sanity check myself here before I tell him this violates industry standards and we won't do it. michael sterl wrote:
Carlos, what is the benefit of the phone ringing for 5 mins? maybe a call queue would fix the overall setup? the most I usually see is 25-30 seconds.
-Michael On Jul 26, 2011, at 12:09 PM, Carlos Alvarez wrote:
We have a customer who has asked us to let his main line ring "forever" or at least five minutes. As a standard practice we forcefully hang up on an unanswered call at 90 seconds if the customer has turned off voicemail and auto-attendants. It seems wrong to let a line ring forever or even for minutes at a time, and I recall something in the back of my head about a traditional industry-standard limit.
Your opinions requested.
-- Carlos Alvarez TelEvolve 602-889-3003
_______________________________________________ VoiceOps mailing list VoiceOps at voiceops.org https://puck.nether.net/mailman/listinfo/voiceops
-- Carlos Alvarez TelEvolve 602-889-3003
On 7/26/11 11:09 AM, Carlos Alvarez wrote:
We have a customer who has asked us to let his main line ring "forever" or at least five minutes. As a standard practice we forcefully hang up on an unanswered call at 90 seconds if the customer has turned off voicemail and auto-attendants. It seems wrong to let a line ring forever or even for minutes at a time, and I recall something in the back of my head about a traditional industry-standard limit.
Interfacing to conventional telephony devices gets kind of sticky in these cases. If the line hasn't been seized and returned answer supervision then there is no way to signal the originating device to tear down the call from the end switch. The originating switch can time out and tear down the call without supervision of course. And that is likely to be the issue with honoring your customer's request. The main obstacle to letting it ring forever is that even if *you* set a ridiculously long timeout, the rest of the world calling his main line isn't going to be inclined to follow suit. Virtually all carriers on the originating side will treat the call as abandoned after some time that is likely going to be less than 300 seconds. Cellular carriers especially are going to be aggressive about tearing down calls that are consuming airtime and not generating revenue. Likewise long distance and international carriers. Back in the good old days of various colored boxes, it wasn't uncommon to have calls that weren't officially answered (OK, technically they were *very* briefly answered) that would last for an hour or more. I don't personally know how any of this was done, of course. ;-) -- Jay Hennigan - CCIE #7880 - Network Engineering - jay at impulse.net Impulse Internet Service - http://www.impulse.net/ Your local telephone and internet company - 805 884-6323 - WB6RDV
+1 on this answer. Very few carriers (from the origination side) will allow a call to ring forever, so the best answer to this is to test a few of the relevant providers (Major cell carriers, local iLEC landline, etc.) and see when they time out, and then let that be the limiting factor. This will be easy to show the customer, of course these types of customers are rarely influenced by facts. Not to mention there isn't a sane human in the world that will stay on the line for 5 minutes when the line is just ringing and ringing. Way before that time they'll hang up and try your customer's competitor, who probably has an AA that answers immediately. -Scott -----Original Message----- From: voiceops-bounces at voiceops.org [mailto:voiceops-bounces at voiceops.org] On Behalf Of Jay Hennigan Sent: Tuesday, July 26, 2011 2:34 PM To: voiceops at voiceops.org Subject: Re: [VoiceOps] Letting a phone ring forever? On 7/26/11 11:09 AM, Carlos Alvarez wrote:
We have a customer who has asked us to let his main line ring "forever" or at least five minutes. As a standard practice we forcefully hang up on an unanswered call at 90 seconds if the customer has turned off voicemail and auto-attendants. It seems wrong to let a line ring forever or even for minutes at a time, and I recall something in the back of my head about a traditional industry-standard limit.
Interfacing to conventional telephony devices gets kind of sticky in these cases. If the line hasn't been seized and returned answer supervision then there is no way to signal the originating device to tear down the call from the end switch. The originating switch can time out and tear down the call without supervision of course. And that is likely to be the issue with honoring your customer's request. The main obstacle to letting it ring forever is that even if *you* set a ridiculously long timeout, the rest of the world calling his main line isn't going to be inclined to follow suit. Virtually all carriers on the originating side will treat the call as abandoned after some time that is likely going to be less than 300 seconds. Cellular carriers especially are going to be aggressive about tearing down calls that are consuming airtime and not generating revenue. Likewise long distance and international carriers. Back in the good old days of various colored boxes, it wasn't uncommon to have calls that weren't officially answered (OK, technically they were *very* briefly answered) that would last for an hour or more. I don't personally know how any of this was done, of course. ;-) -- Jay Hennigan - CCIE #7880 - Network Engineering - jay at impulse.net Impulse Internet Service - http://www.impulse.net/ Your local telephone and internet company - 805 884-6323 - WB6RDV _______________________________________________ VoiceOps mailing list VoiceOps at voiceops.org https://puck.nether.net/mailman/listinfo/voiceops
Scott Berkman wrote:
Not to mention there isn't a sane human in the world that will stay on the line for 5 minutes when the line is just ringing and ringing. Way before that time they'll hang up and try your customer's competitor, who probably has an AA that answers immediately.
This of course is the most important thing; the human part. When we had it configured with a "please hold while we find someone to assist you," people would hang on for 2-3 minutes with a very low abandon rate. The customer refuses to tell us why he wanted it changed to just ring. echo customername >> crazylist.txt I've answered with the facts about industry standards, other carriers, and that nobody has yet hung on for even 90 seconds, and closed the support ticket. -- Carlos Alvarez TelEvolve 602-889-3003
On Jul 26, 2011, at 2:57 PM, Carlos Alvarez wrote:
Scott Berkman wrote:
Not to mention there isn't a sane human in the world that will stay on the line for 5 minutes when the line is just ringing and ringing. Way before that time they'll hang up and try your customer's competitor, who probably has an AA that answers immediately.
This of course is the most important thing; the human part.
Scott, you're thinking about a very narrow range of telephony users. Carlos, your premise is valid, but it's too easy just to call the customer crazy and dismiss this type of request. There are good reasons you might want to sound an alarm at a distant location, and get positive acknowledgment when that alarm has been answered. -> Report of a fire in some division of a factory; contact the fire chief -> Place an urgent order that must take precedence over all others. E.g., gas up the helicopter immediately because we have a human organ to deliver -> Contact a security guard to escort an important visitor around the secure area. On some traditional telephone switches, such as the DMS100 or 1AESS, you can in fact make calls that ring forever, at least if both legs of the call are on the same switch. And, in fact, some businesses have business processes depending on this feature. And I suspect you could accomplish the same within most VoIP platforms by either (a) adjusting timers, or (b) sending periodic SIP provisional responses, such as the SIP 182 response, to keep the early-dialog active. Or (c) make creative use of a Call Center feature to keep the calling user in a queue, but continue to send new call attempts to the same destination Call Center Agent until that user answers. But this all suggests another question: is there a better way to get the job done than just using a traditional phone call? mark at ecg.co | +1-229-316-0013 | http://ecg.co/lindsey
We have a different customer who needs to call and signal remote mountain sites over a microwave-based IP network. We let him ring his ATAs up there a LONG time and will do any crazy configs he needs for the purpose. The customer being discussed today is receiving calls from his customers for service on nights and weekends, nothing more. Unless he's willing to explain a reason, I'm not inclined to entertain oddities, particularly when I can prove that nobody is staying on the ringing line for even 90 seconds. Mark R Lindsey wrote:
On Jul 26, 2011, at 2:57 PM, Carlos Alvarez wrote:
Scott Berkman wrote:
Not to mention there isn't a sane human in the world that will stay on the line for 5 minutes when the line is just ringing and ringing. Way before that time they'll hang up and try your customer's competitor, who probably has an AA that answers immediately. This of course is the most important thing; the human part.
Scott, you're thinking about a very narrow range of telephony users.
Carlos, your premise is valid, but it's too easy just to call the customer crazy and dismiss this type of request. There are good reasons you might want to sound an alarm at a distant location, and get positive acknowledgment when that alarm has been answered.
-> Report of a fire in some division of a factory; contact the fire chief
-> Place an urgent order that must take precedence over all others. E.g., gas up the helicopter immediately because we have a human organ to deliver
-> Contact a security guard to escort an important visitor around the secure area.
On some traditional telephone switches, such as the DMS100 or 1AESS, you can in fact make calls that ring forever, at least if both legs of the call are on the same switch. And, in fact, some businesses have business processes depending on this feature.
And I suspect you could accomplish the same within most VoIP platforms by either (a) adjusting timers, or (b) sending periodic SIP provisional responses, such as the SIP 182 response, to keep the early-dialog active. Or (c) make creative use of a Call Center feature to keep the calling user in a queue, but continue to send new call attempts to the same destination Call Center Agent until that user answers.
But this all suggests another question: is there a better way to get the job done than just using a traditional phone call?
mark at ecg.co | +1-229-316-0013 | http://ecg.co/lindsey
-- Carlos Alvarez TelEvolve 602-889-3003
What about a looping call-forward no answer? Lee
-----Original Message----- From: voiceops-bounces at voiceops.org [mailto:voiceops- bounces at voiceops.org] On Behalf Of Carlos Alvarez Sent: Tuesday, July 26, 2011 2:18 PM Cc: voiceops at voiceops.org Subject: Re: [VoiceOps] Letting a phone ring forever?
We have a different customer who needs to call and signal remote mountain sites over a microwave-based IP network. We let him ring his ATAs up there a LONG time and will do any crazy configs he needs for the purpose.
The customer being discussed today is receiving calls from his customers for service on nights and weekends, nothing more. Unless he's willing to explain a reason, I'm not inclined to entertain oddities, particularly when I can prove that nobody is staying on the ringing line for even 90 seconds.
Mark R Lindsey wrote:
On Jul 26, 2011, at 2:57 PM, Carlos Alvarez wrote:
Scott Berkman wrote:
Not to mention there isn't a sane human in the world that will stay on the line for 5 minutes when the line is just ringing and ringing. Way before that time they'll hang up and try your customer's competitor, who probably has an AA that answers immediately. This of course is the most important thing; the human part.
Scott, you're thinking about a very narrow range of telephony users.
Carlos, your premise is valid, but it's too easy just to call the customer crazy and dismiss this type of request. There are good reasons you might want to sound an alarm at a distant location, and get positive acknowledgment when that alarm has been answered.
-> Report of a fire in some division of a factory; contact the fire -> chief
-> Place an urgent order that must take precedence over all others. -> E.g., gas up the helicopter immediately because we have a human -> organ to deliver
-> Contact a security guard to escort an important visitor around the secure area.
On some traditional telephone switches, such as the DMS100 or 1AESS, you can in fact make calls that ring forever, at least if both legs of the call are on the same switch. And, in fact, some businesses have business processes depending on this feature.
And I suspect you could accomplish the same within most VoIP platforms
by either (a) adjusting timers, or (b) sending periodic SIP provisional responses, such as the SIP 182 response, to keep the early-dialog active. Or (c) make creative use of a Call Center feature to keep the calling user in a queue, but continue to send new call attempts to the same destination Call Center Agent until that user answers.
But this all suggests another question: is there a better way to get the job
done than just using a traditional phone call?
mark at ecg.co | +1-229-316-0013 | http://ecg.co/lindsey
-- Carlos Alvarez TelEvolve 602-889-3003
_______________________________________________ VoiceOps mailing list VoiceOps at voiceops.org https://puck.nether.net/mailman/listinfo/voiceops
Of course, you're free to decide you don't want to offer that particular service. But AT&T made a great business out of selling custom and bizarre services at custom pricing. For example, AT&T had video calling available as a feature, in 1964. They also sold faxing to major news services long before fax machines were readily available; IIRC it was over $10,000/month. The real problem, imho, happens when a salesguy offers a custom-and-bizarre service -- but without accounting for the true cost of the custom engineering. On Jul 26, 2011, at 3:17 PM, Carlos Alvarez wrote:
We have a different customer who needs to call and signal remote mountain sites over a microwave-based IP network. We let him ring his ATAs up there a LONG time and will do any crazy configs he needs for the purpose.
The customer being discussed today is receiving calls from his customers for service on nights and weekends, nothing more. Unless he's willing to explain a reason, I'm not inclined to entertain oddities, particularly when I can prove that nobody is staying on the ringing line for even 90 seconds.
Mark R Lindsey wrote:
On Jul 26, 2011, at 2:57 PM, Carlos Alvarez wrote:
Scott Berkman wrote:
Not to mention there isn't a sane human in the world that will stay on the line for 5 minutes when the line is just ringing and ringing. Way before that time they'll hang up and try your customer's competitor, who probably has an AA that answers immediately. This of course is the most important thing; the human part.
Scott, you're thinking about a very narrow range of telephony users.
Carlos, your premise is valid, but it's too easy just to call the customer crazy and dismiss this type of request. There are good reasons you might want to sound an alarm at a distant location, and get positive acknowledgment when that alarm has been answered.
-> Report of a fire in some division of a factory; contact the fire chief
-> Place an urgent order that must take precedence over all others. E.g., gas up the helicopter immediately because we have a human organ to deliver
-> Contact a security guard to escort an important visitor around the secure area.
On some traditional telephone switches, such as the DMS100 or 1AESS, you can in fact make calls that ring forever, at least if both legs of the call are on the same switch. And, in fact, some businesses have business processes depending on this feature.
And I suspect you could accomplish the same within most VoIP platforms by either (a) adjusting timers, or (b) sending periodic SIP provisional responses, such as the SIP 182 response, to keep the early-dialog active. Or (c) make creative use of a Call Center feature to keep the calling user in a queue, but continue to send new call attempts to the same destination Call Center Agent until that user answers.
But this all suggests another question: is there a better way to get the job done than just using a traditional phone call?
mark at ecg.co | +1-229-316-0013 | http://ecg.co/lindsey
-- Carlos Alvarez TelEvolve 602-889-3003
_______________________________________________ VoiceOps mailing list VoiceOps at voiceops.org https://puck.nether.net/mailman/listinfo/voiceops
On 07/26/2011 03:13 PM, Mark R Lindsey wrote:
And I suspect you could accomplish the same within most VoIP platforms by either (a) adjusting timers, or (b) sending periodic SIP provisional responses, such as the SIP 182 response
We've found that periodically resending 1xx provisional responses doesn't influence how long most UACs are willing to entertain a dialog in early / progress state. -- Alex Balashov - Principal Evariste Systems LLC 260 Peachtree Street NW Suite 2200 Atlanta, GA 30303 Tel: +1-678-954-0670 Fax: +1-404-961-1892 Web: http://www.evaristesys.com/
Also, some SIP stacks do not handle multiple 18X responses well (or choose/are configured not to at all). -----Original Message----- From: voiceops-bounces at voiceops.org [mailto:voiceops-bounces at voiceops.org] On Behalf Of Alex Balashov Sent: Tuesday, July 26, 2011 4:00 PM To: voiceops at voiceops.org Subject: Re: [VoiceOps] Letting a phone ring forever? On 07/26/2011 03:13 PM, Mark R Lindsey wrote:
And I suspect you could accomplish the same within most VoIP platforms by either (a) adjusting timers, or (b) sending periodic SIP provisional responses, such as the SIP 182 response
We've found that periodically resending 1xx provisional responses doesn't influence how long most UACs are willing to entertain a dialog in early / progress state. -- Alex Balashov - Principal Evariste Systems LLC 260 Peachtree Street NW Suite 2200 Atlanta, GA 30303 Tel: +1-678-954-0670 Fax: +1-404-961-1892 Web: http://www.evaristesys.com/ _______________________________________________ VoiceOps mailing list VoiceOps at voiceops.org https://puck.nether.net/mailman/listinfo/voiceops
On 7/26/11 11:57 AM, Carlos Alvarez wrote:
This of course is the most important thing; the human part. When we had it configured with a "please hold while we find someone to assist you," people would hang on for 2-3 minutes with a very low abandon rate.
Hence the difference between Erlang A and Erlang B. -- Jay Hennigan - CCIE #7880 - Network Engineering - jay at impulse.net Impulse Internet Service - http://www.impulse.net/ Your local telephone and internet company - 805 884-6323 - WB6RDV
I think the default timer for ISUP calls between ACM and ANS is somewhere around 1.5 -> 3 minutes usually.. You won't find anywhere that allows much more.. ISUP T9 timer I think, but that may not be useful for ANSI ISUP.. On Tue, Jul 26, 2011 at 1:33 PM, Jay Hennigan <jay at west.net> wrote:
On 7/26/11 11:09 AM, Carlos Alvarez wrote:
We have a customer who has asked us to let his main line ring "forever" or at least five minutes. As a standard practice we forcefully hang up on an unanswered call at 90 seconds if the customer has turned off voicemail and auto-attendants. It seems wrong to let a line ring forever or even for minutes at a time, and I recall something in the back of my head about a traditional industry-standard limit.
Interfacing to conventional telephony devices gets kind of sticky in these cases. If the line hasn't been seized and returned answer supervision then there is no way to signal the originating device to tear down the call from the end switch. The originating switch can time out and tear down the call without supervision of course. And that is likely to be the issue with honoring your customer's request.
The main obstacle to letting it ring forever is that even if *you* set a ridiculously long timeout, the rest of the world calling his main line isn't going to be inclined to follow suit. Virtually all carriers on the originating side will treat the call as abandoned after some time that is likely going to be less than 300 seconds. Cellular carriers especially are going to be aggressive about tearing down calls that are consuming airtime and not generating revenue. Likewise long distance and international carriers.
Back in the good old days of various colored boxes, it wasn't uncommon to have calls that weren't officially answered (OK, technically they were *very* briefly answered) that would last for an hour or more. I don't personally know how any of this was done, of course. ;-)
-- Jay Hennigan - CCIE #7880 - Network Engineering - jay at impulse.net Impulse Internet Service - http://www.impulse.net/ Your local telephone and internet company - 805 884-6323 - WB6RDV _______________________________________________ VoiceOps mailing list VoiceOps at voiceops.org https://puck.nether.net/mailman/listinfo/voiceops
On 7/26/2011 2:09 PM, Carlos Alvarez wrote:
We have a customer who has asked us to let his main line ring "forever" [...] Your opinions requested.
Is he in control of the ringtone ? I'm wondering if he's trying to play a message to callers without having to pay for minutes. -- Jeremy Kister http://jeremy.kister.net./
On 27-Jul-11 05:11, Jeremy Kister wrote:
On 7/26/2011 2:09 PM, Carlos Alvarez wrote:
We have a customer who has asked us to let his main line ring "forever" [...] Your opinions requested.
Is he in control of the ringtone ?
I'm wondering if he's trying to play a message to callers without having to pay for minutes.
I've had numerous customers ask for longer ringback timeouts on the calling side; it turns out that the above is a /very/ common practice for toll-free numbers. The menus and hold music are all "ringback" messages, and the call isn't actually "answered" until a human agent gets on the line. As many of us are unfortunately aware from our own experience calling customer (dis)service lines, that can easily exceed five minutes. 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
Hi, of course these music, ivr, messages are early media/session progress which is also limited to by the SIP-timer values of the involved SIP Elements. It depends on the setup, but this is definately possible in certain configurations. I have seen progress messages limited to 2-3 Minutes.. (german mobile carriers), so it would not be possible to play longer announcements/music to these parties... at least without connect :) BR Max M. Am 27.07.2011 15:28, schrieb Stephen Sprunk:
On 27-Jul-11 05:11, Jeremy Kister wrote:
On 7/26/2011 2:09 PM, Carlos Alvarez wrote:
We have a customer who has asked us to let his main line ring "forever" [...] Your opinions requested.
Is he in control of the ringtone ?
I'm wondering if he's trying to play a message to callers without having to pay for minutes.
I've had numerous customers ask for longer ringback timeouts on the calling side; it turns out that the above is a /very/ common practice for toll-free numbers. The menus and hold music are all "ringback" messages, and the call isn't actually "answered" until a human agent gets on the line. As many of us are unfortunately aware from our own experience calling customer (dis)service lines, that can easily exceed five minutes.
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
_______________________________________________ VoiceOps mailing list VoiceOps at voiceops.org https://puck.nether.net/mailman/listinfo/voiceops
On 7/27/11 6:28 AM, Stephen Sprunk wrote:
I've had numerous customers ask for longer ringback timeouts on the calling side; it turns out that the above is a /very/ common practice for toll-free numbers. The menus and hold music are all "ringback" messages, and the call isn't actually "answered" until a human agent gets on the line. As many of us are unfortunately aware from our own experience calling customer (dis)service lines, that can easily exceed five minutes.
The menus? Really? I have had issues with this and as far as I can tell, the menus should not be considered ringing. We have had reports of DTMF failures navigating the IVR of toll-free numbers from time to time. American Airlines is one specific example. What we have found is that there really isn't a DTMF issue, but that the forward audio path is being disabled until answer somewhere enroute. This is to me the correct behavior and was implemented at least a decade ago as a fraud prevention measure to prevent an endpoint from simply not providing answer supervision and conversing.
From what I recall, the forward audio path is *supposed* to be blocked until the call is answered. Reverse audio is enabled for ringback tones, intercept recordings, and the like.
For a toll-free end user to deliberately not provide answer supervision (and thus start billing) and expect to process DTMF into a menu or for that matter just carry on a conversation seems fraudulent. Of course American Airlines and the like is kind of an 800-pound gorilla and trying to get them or their carrier to alter this behavior of a free ride navigating the menus would be an uphill battle. So I could theoretically have a toll-free number that would have a menu, "Press 1 to hear an audiobook recording of 'War and Peace'; press 2 to hear the 1812 overture..." and run it for free? Doesn't seem kosher. -- Jay Hennigan - CCIE #7880 - Network Engineering - jay at impulse.net Impulse Internet Service - http://www.impulse.net/ Your local telephone and internet company - 805 884-6323 - WB6RDV
I forget the GR that specifies this, but unless a human answers, they don't have to actually send an ISUP ANS back..It's quite common.. The biggest one I can remember is US Airways.. On Wed, Jul 27, 2011 at 3:05 PM, Jay Hennigan <jay at west.net> wrote:
On 7/27/11 6:28 AM, Stephen Sprunk wrote:
I've had numerous customers ask for longer ringback timeouts on the calling side; it turns out that the above is a /very/ common practice for toll-free numbers. The menus and hold music are all "ringback" messages, and the call isn't actually "answered" until a human agent gets on the line. As many of us are unfortunately aware from our own experience calling customer (dis)service lines, that can easily exceed five minutes.
The menus? Really?
I have had issues with this and as far as I can tell, the menus should not be considered ringing.
We have had reports of DTMF failures navigating the IVR of toll-free numbers from time to time. American Airlines is one specific example.
What we have found is that there really isn't a DTMF issue, but that the forward audio path is being disabled until answer somewhere enroute. This is to me the correct behavior and was implemented at least a decade ago as a fraud prevention measure to prevent an endpoint from simply not providing answer supervision and conversing.
From what I recall, the forward audio path is *supposed* to be blocked until the call is answered. Reverse audio is enabled for ringback tones, intercept recordings, and the like.
For a toll-free end user to deliberately not provide answer supervision (and thus start billing) and expect to process DTMF into a menu or for that matter just carry on a conversation seems fraudulent.
Of course American Airlines and the like is kind of an 800-pound gorilla and trying to get them or their carrier to alter this behavior of a free ride navigating the menus would be an uphill battle.
So I could theoretically have a toll-free number that would have a menu, "Press 1 to hear an audiobook recording of 'War and Peace'; press 2 to hear the 1812 overture..." and run it for free? Doesn't seem kosher.
-- Jay Hennigan - CCIE #7880 - Network Engineering - jay at impulse.net Impulse Internet Service - http://www.impulse.net/ Your local telephone and internet company - 805 884-6323 - WB6RDV _______________________________________________ VoiceOps mailing list VoiceOps at voiceops.org https://puck.nether.net/mailman/listinfo/voiceops
On 7/27/11 1:16 PM, Tim Jackson wrote:
I forget the GR that specifies this, but unless a human answers, they don't have to actually send an ISUP ANS back..It's quite common.. The biggest one I can remember is US Airways..
OK, that alone is a huge abuse target "Call this number for a recorded traffic report (and advertisement)". Even assuming that allowing reverse non-human audio is allowed and common, what about the forward media? Is it specifically required that the caller be allowed to pass audio to the destination before ISUP ANS? Is is specifically prohibited? -- Jay Hennigan - CCIE #7880 - Network Engineering - jay at impulse.net Impulse Internet Service - http://www.impulse.net/ Your local telephone and internet company - 805 884-6323 - WB6RDV
US Airways does this, too. I agree with everything you've said, including the 800-pound gorilla comment. As such, we've been using US Airways and AT&T's Business number (800-222-3000) as test numbers when doing interop because of this very issue. - Darren Schreiber -- On 7/27/11 1:05 PM, "Jay Hennigan" <jay at west.net> wrote:
On 7/27/11 6:28 AM, Stephen Sprunk wrote:
I've had numerous customers ask for longer ringback timeouts on the calling side; it turns out that the above is a /very/ common practice for toll-free numbers. The menus and hold music are all "ringback" messages, and the call isn't actually "answered" until a human agent gets on the line. As many of us are unfortunately aware from our own experience calling customer (dis)service lines, that can easily exceed five minutes.
The menus? Really?
I have had issues with this and as far as I can tell, the menus should not be considered ringing.
We have had reports of DTMF failures navigating the IVR of toll-free numbers from time to time. American Airlines is one specific example.
What we have found is that there really isn't a DTMF issue, but that the forward audio path is being disabled until answer somewhere enroute. This is to me the correct behavior and was implemented at least a decade ago as a fraud prevention measure to prevent an endpoint from simply not providing answer supervision and conversing.
From what I recall, the forward audio path is *supposed* to be blocked until the call is answered. Reverse audio is enabled for ringback tones, intercept recordings, and the like.
For a toll-free end user to deliberately not provide answer supervision (and thus start billing) and expect to process DTMF into a menu or for that matter just carry on a conversation seems fraudulent.
Of course American Airlines and the like is kind of an 800-pound gorilla and trying to get them or their carrier to alter this behavior of a free ride navigating the menus would be an uphill battle.
So I could theoretically have a toll-free number that would have a menu, "Press 1 to hear an audiobook recording of 'War and Peace'; press 2 to hear the 1812 overture..." and run it for free? Doesn't seem kosher.
-- Jay Hennigan - CCIE #7880 - Network Engineering - jay at impulse.net Impulse Internet Service - http://www.impulse.net/ Your local telephone and internet company - 805 884-6323 - WB6RDV _______________________________________________ VoiceOps mailing list VoiceOps at voiceops.org https://puck.nether.net/mailman/listinfo/voiceops
I think I might have brought up an old thread that American Idol does the same thing they do not pick up the call they just play there IVR. We had an issue a while back with metaswitch that they where not passing dtmf on calls that the ss7 had not signaled that it had picked up. Carlos Alcantar Race Communications / Race Team Member 101 Haskins Way, So. San Francisco, CA. 94080 Phone: +1 415 376 3314 Fax: +1 650 246 8901 / carlos *at* race.com / www.race.com <http://www.race.com/> On 7/27/11 1:26 PM, "Darren Schreiber" <d at d-man.org> wrote:
US Airways does this, too.
I agree with everything you've said, including the 800-pound gorilla comment. As such, we've been using US Airways and AT&T's Business number (800-222-3000) as test numbers when doing interop because of this very issue.
- Darren Schreiber
--
On 7/27/11 1:05 PM, "Jay Hennigan" <jay at west.net> wrote:
On 7/27/11 6:28 AM, Stephen Sprunk wrote:
I've had numerous customers ask for longer ringback timeouts on the calling side; it turns out that the above is a /very/ common practice for toll-free numbers. The menus and hold music are all "ringback" messages, and the call isn't actually "answered" until a human agent gets on the line. As many of us are unfortunately aware from our own experience calling customer (dis)service lines, that can easily exceed five minutes.
The menus? Really?
I have had issues with this and as far as I can tell, the menus should not be considered ringing.
We have had reports of DTMF failures navigating the IVR of toll-free numbers from time to time. American Airlines is one specific example.
What we have found is that there really isn't a DTMF issue, but that the forward audio path is being disabled until answer somewhere enroute. This is to me the correct behavior and was implemented at least a decade ago as a fraud prevention measure to prevent an endpoint from simply not providing answer supervision and conversing.
From what I recall, the forward audio path is *supposed* to be blocked until the call is answered. Reverse audio is enabled for ringback tones, intercept recordings, and the like.
For a toll-free end user to deliberately not provide answer supervision (and thus start billing) and expect to process DTMF into a menu or for that matter just carry on a conversation seems fraudulent.
Of course American Airlines and the like is kind of an 800-pound gorilla and trying to get them or their carrier to alter this behavior of a free ride navigating the menus would be an uphill battle.
So I could theoretically have a toll-free number that would have a menu, "Press 1 to hear an audiobook recording of 'War and Peace'; press 2 to hear the 1812 overture..." and run it for free? Doesn't seem kosher.
-- Jay Hennigan - CCIE #7880 - Network Engineering - jay at impulse.net Impulse Internet Service - http://www.impulse.net/ Your local telephone and internet company - 805 884-6323 - WB6RDV _______________________________________________ VoiceOps mailing list VoiceOps at voiceops.org https://puck.nether.net/mailman/listinfo/voiceops
_______________________________________________ VoiceOps mailing list VoiceOps at voiceops.org https://puck.nether.net/mailman/listinfo/voiceops
On 27-Jul-11 15:05, Jay Hennigan wrote:
On 7/27/11 6:28 AM, Stephen Sprunk wrote:
I've had numerous customers ask for longer ringback timeouts on the calling side; it turns out that the above is a /very/ common practice for toll-free numbers. The menus and hold music are all "ringback" messages, and the call isn't actually "answered" until a human agent gets on the line. As many of us are unfortunately aware from our own experience calling customer (dis)service lines, that can easily exceed five minutes. The menus? Really?
I have had issues with this and as far as I can tell, the menus should not be considered ringing.
We have had reports of DTMF failures navigating the IVR of toll-free numbers from time to time. American Airlines is one specific example.
The most common complaints I've seen involve airlines; so far, I know United, American and US Airways use this trick. Banks are a distant second, but most of them seem to do it too; they just generate fewer complaints since most folks are only calling to check balances and such via the automated system, rather than waiting for a human.
What we have found is that there really isn't a DTMF issue, but that the forward audio path is being disabled until answer somewhere enroute. This is to me the correct behavior and was implemented at least a decade ago as a fraud prevention measure to prevent an endpoint from simply not providing answer supervision and conversing.
We tried that and quickly got smacked down by our customers, and I suspect the same happened to other folks.
So I could theoretically have a toll-free number that would have a menu, "Press 1 to hear an audiobook recording of 'War and Peace'; press 2 to hear the 1812 overture..." and run it for free? Doesn't seem kosher.
The replacement "solution" is to put a hard limit on the amount of time spent in ringback state. We default to 180 seconds and get periodic complaints; changing that to 300 seconds seems to make the complaints go away. 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
Jeremy Kister wrote:
On 7/26/2011 2:09 PM, Carlos Alvarez wrote:
We have a customer who has asked us to let his main line ring "forever" [...] Your opinions requested.
Is he in control of the ringtone ?
No. Our business model is that we vet all change requests and do them ourselves. The customer has no access to make changes.
I'm wondering if he's trying to play a message to callers without having to pay for minutes.
It's none of the obvious things like toll fraud (avoidance?), since they are all local numbers and over 99% of calls are from other local numbers. The customer is on a completely flat-rate hosted PBX service with us, with no practical limits on number of channels or number of minutes. -- Carlos Alvarez TelEvolve 602-889-3003
participants (13)
-
abalashov＠evaristesys.com -
carlos＠race.com -
carlos＠televolve.com -
d＠d-man.org -
jackson.tim＠gmail.com -
jay＠west.net -
lindsey＠e-c-group.com -
LRiemer＠bestline.net -
michael＠simplesignal.com -
mm＠42com.com -
scott＠sberkman.net -
stephen＠sprunk.org -
voiceops-01＠jeremykister.com