Dan Bernstein <brnstnd@kramden.acf.nyu.edu> (08/21/90)
In this city, a local phone call of unlimited length has fixed cost. Call waiting is a lot cheaper than an extra line. It occurs to me that modems could reasonably recognize the call waiting beep. Then the user could answer the call and talk normally, without dropping the computer connection. In other words, for a monthly call waiting charge, you could get a permanent, basically free connection to the computer of your choice. What would it take to get this working in practice? The modem technology would be relatively simple: the hardest part would be convincing the modem on the other side not to hang up during a regular phone conversation. (It shouldn't be hard to make this work with answering machines either.) But what would the phone companies think of people getting connections so cheaply? Dan [Moderator's Note: Suppose you could set your modem to never time out; to never drop carrier, meaning you could flash your switchhook to take a call and your modem would just sit there waiting. If you could do that, how would the other end know you were on a call-waiting and had not disconnected abrubtly? What would prevent the other end from dropping carrier after it found your carrier was lost? Now if the one on the other end was fixed like yours, to ignore loss of carrier and just sit there humming away waiting for someone to return, then what would happen if some other user called and got accidentally cut off? How would the distant modem recover from that? What you are asking for is not as easy as merely fixing your own modem to ignore loss of carrier while you are on another call. And if your modem did work that way, would you want to sit there and try to converse with someone over the carrier tone (which was still there since you told it not to leave)? I don't think it would work out at all. And do not think that the telco is very concerned 'about people getting calls so cheaply', since most modem owners probably already have a second line to start with, and a phone bill double what a non-modem user is paying. PAT]
mje@gargoyle.uchicago.edu> (08/28/90)
In article <11160@accuvax.nwu.edu> brnstnd@kramden.acf.nyu.edu (Dan Bernstein) writes: X-Telecom-Digest: Volume 10, Issue 585, Message 6 of 14 >In this city, a local phone call of unlimited length has fixed cost. As used to be the case in Johannesburg ... (Until 'local' meters were installed ... Local calls in most other areas are still unlimited by time.) >.... In other words, for a >monthly call waiting charge, you could get a permanent, basically free >connection to the computer of your choice. >... But what would the phone companies think of people getting >connections so cheaply? As one customer used to do ... dedicate a phone line just for modem use. His cost was 2 X monthly rental which was 18 Rand X 2 which is about $13 a month for the total cost of the line - etc (both ends). When the PO found out - they were not too happy - they developed some manual routine of disconecting him some time about 5pm each night - so he was forced to make at least one call a day. They don't like people doing this, as it competes with their own 'data-line' service which costs a lot more per month. Olivetti Systems & Networks, Unix Support - Africa UUCP: {uunet,olgb1,olnl1}!olsa99!mje (Mark Elkins) mje@olsa99.UUCP (Postmaster) Tel: +27 11 339 9093
David M Archer <v116kznd@ubvmsd.cc.buffalo.edu> (08/29/90)
In article <11444@accuvax.nwu.edu>, mje99!mje@gargoyle.uchicago.edu (Mark J Elkins) writes... >As one customer used to do ... dedicate a phone line just for modem >use. His cost was 2 X monthly rental which was 18 Rand X 2 which is >about $13 a month for the total cost of the line - etc (both ends). >When the PO found out - they were not too happy - they developed some >manual routine of disconecting him some time about 5pm each night - so >he was forced to make at least one call a day. They don't like people >doing this, as it competes with their own 'data-line' service which >costs a lot more per month. Are they, the phone company(s), allowed to do this? I've never heard of a maximum period of time for a phone call. I consulted my phone book, and the term it uses is untimed. I'd call the customer representatives once a day and request a credit. After all, my phone call was interrupted by them, and so their "equipment failure" required me to make a second phone call. Seems appropriate to me. I ask about this, because I can quite easily see myself doing something similiar in the future.
John Higdon <john@bovine.ati.com> (08/29/90)
Steck Thomas <steck@aplcen.apl.jhu.edu> writes: > PAT - I beg to differ with you on this one. On the older switching > systems, this may have been true. However, on the new ATT 5ESS and > the Northern Telecom DMS-100, call waiting is not implemented as a > voltage changee - simply a tone introduced to the line. There is no > true interuption of the line and no 'switching' sounds (clicks, etc..) > like there used to be. But even on these digital switches, the talk path is interrupted for the duration of the "beep". In most cases, this is enough for a modem to consider that there has been carrier loss and to hang up. There may be no clicks, but there is definately interruption of the line. If you don't believe me, call someone on a 5ESS or DMS100 who has call waiting and have them hum into the line. Then call them on another phone and see if you don't hear the person disappear for a moment. But that's all quite moot. There are so many 1AESS switches that will be around for so long as to make it impossible to discount their presence in the telecom world. Even the ancient 1ESS "serving" my home phone is not scheduled for replacement. ("Hell, we wouldn't want to waste money replacing THAT -- it still completes calls sometimes, doesn't it?") John Higdon | P. O. Box 7648 | +1 408 723 1395 john@bovine.ati.com | San Jose, CA 95150 | M o o !
Dan Bernstein <brnstnd@kramden.acf.nyu.edu> (08/30/90)
In article <11160@accuvax.nwu.edu> TELECOM Moderator replies: > [Moderator's Note: Suppose you could set your modem to never time out; > to never drop carrier, meaning you could flash your switchhook to take > a call and your modem would just sit there waiting. If you could do > that, how would the other end know you were on a call-waiting and had > not disconnected abruptly? A timeout is okay. Anyway, I envisioned something like this: Local modem hears call waiting beep. Local modem somehow communicates to remote modem that it's call waiting time. Remote modem acknowledges. Local modem shuts up, dropping carrier. You talk. Eventually the line flashes back to the remote modem. Local modem hears remote's carrier again. Local modem starts generating carrier. Remote hears this and undoes whatever flow control it might have done before. > What you are asking > for is not as easy as merely fixing your own modem to ignore loss of > carrier while you are on another call. Certainly; I don't see this sort of thing working unless both modems are modified to take positive action upon the call waiting beep. Note that once you've gotten over the technical hurdle of recognizing the beep, you can use that for the wait-ack sequence (sort of like a connect). > And if your modem did work that > way, would you want to sit there and try to converse with someone over > the carrier tone (which was still there since you told it not to > leave)? The local carrier would disappear. You wouldn't hear the remote one. > I don't think it would work out at all. On the contrary: all your technical objections are answered by newer phone systems; there's nothing inherently difficult about the idea; and I think most modem users would jump on it in an instant. > And do not think that > the telco is very concerned 'about people getting calls so cheaply', > since most modem owners probably already have a second line to start > with, and a phone bill double what a non-modem user is paying. But a large number don't. Even the ones who do probably wouldn't mind turning one line plus one modem connection into two lines plus one modem connection, for just the cost of call waiting. And as Mark Elkins points out (10/603/10 of 12), the phone company could very well be concerned about this. Dan
dricejb@husc6.harvard.edu (Craig Jackson drilex1) (09/02/90)
In article <11470@accuvax.nwu.edu> v116kznd@ubvmsd.cc.buffalo.edu writes: X-Telecom-Digest: Volume 10, Issue 605, Message 8 of 10 >In article <11444@accuvax.nwu.edu>, mje99!mje@gargoyle.uchicago.edu >(Mark J Elkins) writes... >>As one customer used to do ... dedicate a phone line just for modem >>use. His cost was 2 X monthly rental which was 18 Rand X 2 which is >>about $13 a month for the total cost of the line - etc (both ends). >>When the PO found out - they were not too happy - they developed some >>manual routine of disconecting him some time about 5pm each night - so >>he was forced to make at least one call a day. They don't like people >>doing this, as it competes with their own 'data-line' service which >>costs a lot more per month. >Are they, the phone company(s), allowed to do this? I've never heard >of a maximum period of time for a phone call. I consulted my phone >book, and the term it uses is untimed. I'd call the customer >representatives once a day and request a credit. After all, my phone >call was interrupted by them, and so their "equipment failure" >required me to make a second phone call. Seems appropriate to me. I >ask about this, because I can quite easily see myself doing something >similiar in the future. Years ago the way to handle play-by-play coverage of a college basketball game, etc was for the radio station (or network) to order a 'radio loop' from the arena to the origination point. This was a dedicated circuit, which was valid during the period of the game. (Typically, it actually was put up the preceding business day.) Around about 1970, the tariffs in Virginia (where I worked at a student radio station) changed so that it became cheaper to have a POTS line installed in the press box, and send the game back via an ordinary long-distance call. (I mean a POTS line installed *just* for the event -- I never saw the tariffs, but this is certainly true.) This use of an ordinary long-distance call was not a subterfuge--the TELCO craft people knew all about it, and I believe that the business office recommended it. The service for the POTS long-distance call was generally just as good as a radio loop offered in the way of bandwidth, etc. However, it came with fewer guarantees. Regular radio loops had their punchdowns marked and other steps to ensure continuous service. One time during a football game, the coverage was interrupted partway through the game. It was restored in just a few minutes. What had happened was that the long-distance call simply dropped during the middle of the broadcast, and had to be re-dialed. Obviously someone was upset, because we found out what had happend. It seemed that this game was coming from some place with older equipment in the frame. After the call had been live for several hours, some part of this equipment was overheating. The person manning the frame noticed the problem, and disconnected the call manually. (Of course, he didn't listen in -- that would be an unnecessary violation of privacy.) So at least in this case, the 'equipment problem' was real. I don't know if any monetary relief was in order due to the interruption. But I'm pretty sure that the TELCO has themselves covered so they don't have to fork out, or forgo charges, when a call runs so long that it causes equipment failure. Craig Jackson dricejb@drilex.dri.mgh.com {bbn,axiom,redsox,atexnet,ka3ovk}!drilex!{dricej,dricejb}
Kenneth R Crudup <kenny@world.std.com> (09/03/90)
In article <11444@accuvax.nwu.edu>, mje99!mje@gargoyle.uchicago.edu (Mark J Elkins) writes: >As one customer used to do ... dedicate a phone line just for modem use. >When the PO found out - they were not too happy - they developed some >manual routine of disconecting him some time about 5pm each night - so >he was forced to make at least one call a day. In article <11470@accuvax.nwu.edu> v116kznd@ubvmsd.cc.buffalo.edu says: >Are they, the phone company(s), allowed to do this? He lives in the Republic of South Africa. Obviously the government and its agencies (which I bet the phone company is) do anything they want. Kenny Crudup, Unix Systems Consultant nubian!kenny@ima.ima.isc.com 14 John Eliot Sq. #2B, Roxbury, MA 02119-1569 (617) 442 6585
John Higdon <john@bovine.ati.com> (09/03/90)
Craig Jackson drilex1 <drilex!dricejb@husc6.harvard.edu> writes: > One time during a football game, the coverage was interrupted partway > through the game. > So at least in this case, the 'equipment problem' was real. I don't > know if any monetary relief was in order due to the interruption. But > I'm pretty sure that the TELCO has themselves covered so they don't > have to fork out, or forgo charges, when a call runs so long that it > causes equipment failure. With the rapidly advancing cost of equalized lines, most broadcasters have been looking to other methods of carrying program material in both temporary and permanent situations. Most radio stations now use 950 MHz equipment to carry program to the transmitter. But for remote broadcasts, there are a number of options. With the advent of frequency agile, wideband 450 MHz equipment, and large number of broadcasters are turning to the airwaves to solve their remote broadcast woes. Unfortunately, in major metro areas the available frequencies for 450 MHz remote use are scarce. Even with coordinating committees, the scramble for channels frightens many away. A semi-popular alternative is to use a dial up telephone line with a "frequency extender". This device operates under the assumption that what makes a phone line sound bad is the lack of low (yes -- low) frequency response. To correct this, the audio channel is shifted up about 500 Hz. That would mean that a tone of 1000 Hz would travel over the phone line as 1500 Hz. It would also mean that a sound occurring around 50 Hz would travel as 550 Hz, well within the response capability of any phone line. There are also multi-line models that split the band up into parts and send 3000 Hz wide "slices" over each line. Obviously, there is appropriate decoding equipment at the receiving end in all cases. I am personally unimpressed by these devices and feel that if a broadcast is going to travel over a dialup line, it might as well go barefoot for all the improvement you get with "frequency extenders". One of my clients just bites the bullet and buys 8 KHz and 15 KHz dedicated lines and builds them into the cost of the remote as billed to the advertiser. A good telco equalized line is still the champ when it comes to quality -- even over wideband 450 MHz equipment. With the amount that they spend with Pac*Bell (they do MANY remotes), they have been provided with a virtually permanent on-site installer. The service with those lines has been VERY reliable and they sound VERY good. Which brings us to the question about telco liability for service interuptions on dialup. Most tariffs call for a credit of one day's worth of the monthly service charge for each day that the service is unusable after the first 24 hours of service loss. So for the lost call, there would be no liability whatsoever. Only if no calls could be made for at least the next 24 hours would there be any remuneration from telco. And that would be minimal. John Higdon | P. O. Box 7648 | +1 408 723 1395 john@bovine.ati.com | San Jose, CA 95150 | M o o ! [Moderator's Note: His message and your response reminds me of years ago when once a week I would see a very large truck parked in the alley behind Orchestra Hall on Michigan Avenue. The truck was inscribed "Illinois Bell Telephone Co". A large cable coming out the stage door of Orchestra Hall ran into the back of this truck. Then it came out of the truck, and down into a manhole nearby. The weekly broadcast of the Chicago Symphony Orchestra to how many ever radio stations carried it started out at that point. This would have been the early 1950's; I was around ten years old, and fascinated by the inside of the truck, and the fellow who worked inside backstage wearing an operator's headset into which it seemed he was constantly talking to someone, somewhere. PAT]
Donald Krapf <dkrapf@atropos.acm.rpi.edu> (09/04/90)
In article <11471@accuvax.nwu.edu> John Higdon <john@bovine.ati.com> writes: X-Telecom-Digest: Volume 10, Issue 605, Message 9 of 10 >Steck Thomas <steck@aplcen.apl.jhu.edu> writes: >> voltage changee - simply a tone introduced to the line. There is no >> true interuption of the line and no 'switching' sounds (clicks, etc..) >> like there used to be. >But even on these digital switches, the talk path is interrupted for >the duration of the "beep". In most cases, this is enough for a modem >to consider that there has been carrier loss and to hang up. There may Most modems can be instructed to ingore carrier loss for a brief period. The only problem here is that the modem at each end must be so instructed. When I have control of both ends of a line I typically instruct the modems to tolerate a carrier drop of up to three seconds. Don