johnl@ima.UUCP (John R. Levine) (12/17/86)
In article <1575@brl-adm.ARPA> OCONNORDM@ge-crd.arpa (OCONNOR DENNIS MICHAEL) writes: >Second: the goverment does not "mandate" a "pernicous" monopoly, it >simply allows it. You or I can go out, get right-of-way on the >utility poles like the cable companies, and start our very own >telephone system. ... Sorry, but that's just not true. Your local phone company has a monopoly franchise to run phone wires down the street. As a case in point, a few years ago at Yale, we wanted to run an Ethernet cable from one building on the campus to another, running down the street for a block. We could not legally do it -- it fell into the phone company's franchise. Eventually we ended up getting a microwave link because the phone company had no reasonable service to offer. (High speed? You mean 56KB? No, we mean 10MB. Uhh...) The cable company has a separate franchise to run cable TV. Some cable companies want to use their spare capacity for switched data, but that's legally interesting at best. Some comanies have private internal phone nets but that's legally quite different from being a public phone company. I suspect the real wave of the future is in things like ConnNet, which is a packet switched network run in Connecticut by SNET, the local telephone company. You can gateway from ConnNet to other networks. The same logic which says that local voice service is a natural monopoly would suggest that local data service is similarly a natural monopoly and services like Telenet are more analogous to MCI or Sprint than to and end-to-end phone company. The local monopoly on data service also addresses the issue that data calls are more expensive than voice calls. AT&T, for example, has a 1PSS packet switch exchange designed to sit next to the voice exchange and to pull the data calls off the voice circuits as close to the user as possible, to avoid tying up voice circuits with data. The standard rate review process is supposed to ensure that the rates charged for such service are reasonably related to the costs. We'll see. -- John R. Levine, Javelin Software Corp., Cambridge MA +1 617 494 1400 { ihnp4 | decvax | cbosgd | harvard | yale }!ima!johnl, Levine@YALE.EDU Where is Richard Nixon now that we need him?
larry@kitty.UUCP (Larry Lippman) (12/25/86)
In article <403@pixar.UUCP>, aaa@pixar.UUCP (Tony Apodaca) writes: > In article <3454@curly.ucla-cs.UCLA.EDU> stiber@zeus (Michael D Stiber) writes: > >>First: MODEM calls DO NOT cost the phone company the same amount as > >>other calls. For all practical intents and purposes, this statement is INCORRECT. Read on for more details. > >2) Modems use the same lines as voice. The assertion that they tolerate > >noise less well is irrelevant, since they do not get special priveleges. This statement is correct. > I'm sure that this discussion is going hot and heavy in net.dcom.etc but > I'll answer here anyway. Flame off, Mike. The point is true even if the > rationale is messed up. Modem calls DO cost the phone company more, for > several reasons: > 1) They are continuous. The dual carrier never stops. Therefore, > the phone company must supply bandwidth to the call continuously even if > there is no "valid" data. They cannot time-multiplex their signals. > There are small breathing and thinking pauses in all voice conversation, > and 99% of voice is half-duplex, even a teenage girl's conversations. If you are alluding to most telephone calls being statistically multiplexed (for lack of a better simple term) by a process known as TASI (Time-Assigned Speech Interpolation), you are wrong. TASI was first developed by Bell Labs during the 1950's to improve the effective circuit capacity of transoceanic cables where the number of channels was limited, and the per channel cost was very high. TASI does make use of the fact that an intertoll circuit is 4-wire, with separate transmit and receive channels; during normal human conversation each of the two channels is used only about 45% of the time. Therefore, TASI apparatus can almost double the effective channel capacity by assigning a given intertoll trunk to a transmit or receive channel only when speech energy is detected. HOWEVER, TASI apparatus is complex, expensive, and has some objectional human factors (like clipping of syllables). Operating telephone companies have used TASI only for transoceanic circuits, and for some early satellite circuits where TASI was cost-effective. As far as I know, TASI has never been used within the continental U. S. by an operating telephone company. In recent years, there has been a resurgence in interest for TASI apparatus, and a few vendors have offered TASI apparatus for use on private line communication facilities - however, this does NOT apply to use by operating telephone companies for the DDD switched network. So the point is: operating telephone companies do NOT vary the "bandwidth" of telephone call by TASI or other means (with the exception of transoceanic cable use). > 2) They are high bandwidth. The phone line was designed with human > voices in mind, and they are pretty low bandwidth, as everyone knows. Also, > everyone knows that their modems strive to get the most out of it, so they > use it all up (if they didn't you'd buy a new one!). However, the phone > company "counts on" the signals being voice-like, so they can cram as many > signals into one wire as possible, and a modem transmission screws up their > frequency-division multiplexing. This is sick. For all intents and purposes, the bandwith of any dialed telephone call is _limited_ to being well under 4 KHz for several reasons: 1. Many central office subscriber loops are loaded to minimize the effects of capacitive loop attenuation. The result of placing loading coils in the subscriber loop is a distributed L-C low-pass filter which drastically limits bandwith to less than 4 KHz. 2. Frequency-division multiplex apparatus (like N-carrier) effectively limits channel bandwith to less than 4 KHz by virtue of bandpass filters. 3. Time-division multiplex apparatus (like T1-carrier) effectively limits channel bandwith to less than 4 KHz since the T1 sample rate is 8 KHz. 4. Digital central offices effectively limit channel bandwith because they too sample at typically 8 KHz. So the point is: there just ain't no way to achieve more than 4 KHz in bandwidth on the switched DDD network (you're lucky to get 300 to 3,000 Hz). There is no bandwith allocation scheme being used by an operating telephone company that is somehow defeated by a modem. Furthermore, the bandwith required to transmit intelligible speech is the same as is used by any dial-up modem. > 3) The carrier on some modems just happens to overlap a critical > region of part of the phone company's equipment's frequency allocation. > The circuits known as "echo suppression" use it to monitor themselves, and > kick in higher bandwidth and better circuits if there is too much echo on the > line. If they didn't have it, you'd bitterly complain about the quality of > your long-distance connections. Voice doesn't have much of these freqs, > but the carrier flips it out, causing it to allocate too much signal to > your call. Echo supressors do NOT affect channel bandwith. On a voice call an echo suppressor detects speech energy (say on an E-W direction) and inserts a large amount of attenuation (as much as 40 dB) in the opposite direction (say, W-E in the above example). This is why one person may not be able to interrupt another until they stop talking on a DDD circuit with an echo suppressor. Echo suppressors are used sparingly in the DDD network, and generally only on regional intertoll trunks. Only ONE echo suppressor can ever be in a given DDD connection. Echo suppressors are intentionally designed to be disabled by a SF tone between 2,000 Hz and 2,200 Hz which lasts for at least 400 ms; it is a necessary requirement that echo suppresssors be disabled for any data call. The primary purpose of the "answer tone" furnished by a modem is to disable any echo supressor for the duration of the call. Your description of echo suppressor operation is totally wrong. > I don't work for the phone company ... Perhaps you should - then you'd learn something about telephony. I don't enjoy being harsh on you - but your article contained some serious misinformation than has to be corrected. I try to consider Usenet as an educational forum, and it always troubles me to see articles such as yours. <> Larry Lippman @ Recognition Research Corp., Clarence, New York <> UUCP: {allegra|bbncca|decvax|nike|rocksanne|watmath}!sunybcs!kitty!larry <> VOICE: 716/688-1231 {hplabs|ihnp4|mtune|seismo|utzoo}!/ <> FAX: 716/741-9635 {G1,G2,G3 modes} "Have you hugged your cat today?"
qeds@mtuxo.UUCP (e.schulz) (12/27/86)
I was a member of the TASI-E development team at AT&T Bell Labs
from 1979 to 1982, and can confirm that as far as I know, Larry
Lippman's comments about TASI are correct. Advances in digital
radio, fiber-optic cable, and other technologies keep driving the
cost of long-distance facilities down; this keeps TASI from proving
in domestically. There are at least two dozen TASI-E systems in
use on undersea cable (last I heard), each with a capacity of about
240 conversations on 120 long-distance circuits.
The front-end clipping of speech bursts is barely noticeable to the
trained ear on TASI-E. How we achieved this and tested it in field
trials is interesting, but not for comp.dcom.modems!
TASI-E does detect the echo-suppressor disable tone at the start of
a data call, and "pins up" the connection (does not do TASI) for the
duration of that call. If overseas modem (and fax and music on
hold) traffic exceeded 5% or so, the TASI advantage would be
reduced, those routes would need to be re-engineered, and eventually
overseas phone service would cost more. Not domestic service,
though.
--
Ed Schulz, AT&T Information Systems, Middletown, NJ
(201)957-3899 {at&t}!solar!edsrab@well.UUCP (Bob Bickford) (12/27/86)
In a previous article, Larry Lippman writes: > There is no bandwith allocation scheme being used by an operating > telephone company that is somehow defeated by a modem. [note: mucho context deleted] It is my understanding that long-distance calls may be (and frequently are) switched in mid-call between available circuits by automatic equipment. I am also informed that some of this equipment is designed to disable if the channel to be switched is carrying a modem tone. (Because the switchover can take long enough to cause older modem designs to lose carrier lock.) Thus, a modem call could inhibit some of the normal adaptability and flexibility of a network. -- Robert Bickford {hplabs, ucbvax, lll-lcc, ptsfa, msudoc}!well!rab terrorist cryptography DES drugs cipher secret decode NSA CIA NRO IRS coke crack pot LSD russian missile atom nuclear assassinate libyan RSA The above is food for the NSA line eater. Add it to your .signature and you too can help overflow the NSA's ability to scan all traffic going in or out of the USA looking for "significant" words. (This is not a joke, sadly.)
joel@gould9.UUCP (Joel West) (12/29/86)
BizComp and Hayes may have the biggest threat to modem users.
BizComp claims a patent on auto-dialing technology, controling
dialing with the data line (unlike the earlier UNIX modems, which
required a separate calling unit.)
Apparently Hayes paid about $2 million to license (or buy off, if you like)
BizComp's patent threat, and also claims its own patent on the
1 second before and after the +++ escape.
BizComp has sued Prometheus, a modem clone company. Needless to say,
if Hayes and BizComp win, cheap competition in the
modem market would be threatened. US Robotics has filed a
pre-emptive lawsuit alleging antitrust violations against both
companies, but if the patents are valid, it doesn't stand a chance.
--
Joel West MCI Mail: 282-8879
Western Software Technology, POB 2733, Vista, CA 92083
{cbosgd, ihnp4, pyramid, sdcsvax, ucla-cs} !gould9!joel
joel%gould9.uucp@NOSC.ARPA