W8SDZ@simtel20.arpa (Keith Petersen) (12/12/86)
The FCC is considering a ruling which may threaten low-cost modem
access to many on-line services, perhaps including Arpa/Milnet TACs
and Usenet Unix systems. Here are the details from a copy of a file
just uploaded to my Remote CP/M system.
--Keith Petersen
Arpa: W8SDZ@SIMTEL20.ARPA
Uucp: {bellcore,decwrl,harvard,lll-crg,ucbvax,uw-beaver}!simtel20.arpa!w8sdz
GEnie Mail: W8SDZ
RCP/M Royal Oak: 313-759-6569 (300, 1200, 2400 bps)
--cut here--BADNEWS.PCP--cut here--
The FCC is considering reregulating the packet-switching networks like Telenet,
Tymnet, Compuserve, The Source and PC Pursuit. This could result in additional
costs to the user. This is excerpted from Infomat magazine which is available
for downloading.
====================================
COMPUTER AND SOFTWARE NEWS -- PART 1
====================================
by Tim Elmer
------------------------------------
FREE LOCAL ACCESS TO PACKET
SWITCHING NETWORKS MAY BE ELIMINATED
------------------------------------
(BPS) -- The Federal Communications Commission (FCC) will vote on a proposal
to reregulate packet switching networks that, if approved, would eliminate
free local telephone access to those networks.
"If this occurs, it might eventually double or triple the costs to those
using packet switching networks to access commercial on-line databases and
information services and triple or quadruple the costs to those using
Telenet's PC Pursuit," said Philip M. Walker, vice president and regulatory
counsel for Telenet Communications Corp.
Predictably, the initiative to reregulate packet switching networks comes
primarily from the Bell Operating Companies (BOCs) and secondarily from AT&T.
These companies provide local telephone service to vast majority of telephone
customers throughout the U.S. and will benefit the most from FCC reregulation
of the packet switching networks.
Under current FCC rules formulated in 1980 in the FCC's Second Computer
Inquiry, called Computer II, a distinction is made between "basic services"
and "enhanced services."
"Basic services" are those that don't offer protocol conversion such as local
and long-distance voice telephone services. "Enhanced services" are defined
in an open-ended fashion as computer-based services that are more than a
"basic service," in other words, services such as packet switching networks,
database and on-line type services, and remote computing services that offer
protocol conversion, according to Walker.
Under the 1980 Computer II Inquiry, the FCC ruled that "basic services" would
continue to be regulated as they had always been. However, the FCC also ruled
that "enhanced services" would be deregulated, which opened up the industry to
competition. This resulted in numerous companies entering the packet
switching business, including BOCs, AT&T and at least a dozen others. The
competition resulted in significant price reductions for packet switching
services.
To prevent monopolization of the packet switching industry by the Big Boys
(the BOCs and AT&T), the FCC ruled that they had to keep separate accounting
figures for their "basic services" and for their "enhanced services," and that
they could not use revenues from their lucrative "basic services" to cross-
subsidize their "enhanced service" packet switching networks.
The FCC also ruled that if the BOCs and AT&T used their "basic service"
telephone lines for packet switching services, then they must let their
competitors have access to those lines on the same basis, which would preserve
true competition in the industry.
"Now, under the FCC's Computer Inquiry III, the FCC is asking, should we
redefine protocol conversion services as 'basic services' rather than enhanced
services? Should we redefine all those companies as common carriers? This
would, in effect, subject them not only to federal regulations but, even
worse, to state regulations," Walker said.
The result would eliminate comparable interconnection requirements currently
imposed on BOCs and AT&T, allowing them to charge their packet switching
competitors local dial-in fees to access packet switching long-distance line
networks.
It would also allow BOCs and AT&T to offer their own packet switching
services on a non-compensatory basis and, finally, allow them to cross-
subsidize those services with revenues from their much more lucrative voice
telephone service revenues. In short, it would allow BOCs and AT&T to
monopolize the packet switching industry and probably drive out most
competitors.
"In terms of cost impact," Walker said, "if we had to pay local access
charges, it would cost us about $3.60 an hour at the originating end, for
calls made by users to on-line databases and information services like
CompuServe and The Source.
"And with PC Pursuit, for which we have out-dial modems, we would have to pay
not only 3.60 per hour access fees at the originating end but also $4.80 at
the terminating end, a total of about $8 or $9. Obviously, to survive, we
would have to add those additional charges to our current fees and pass them
on to our consumers," Walker said.
That would almost certainly spell the end of PC Pursuit, and it would likely
put out of business not only many independent packet switching networks but
also many on-line databases and information services.
FCC approval of changes being considered in Computer III, Walker said, "would
really have a major impact on anyone using a packet switching service to
access online bulletin boards, databases, or information services aimed at the
residential user. They are just going to get creamed if this happens."
Walker said that is was not clear exactly when the FCC would vote on the
proposal, but that it would probably be the latter part of January or early
part of February, 1987. "They are moving very fast on this," he said.
For additional information, be sure to read Alan Bechtold's editorial in this
issue.
==========END>>>
Copyright (C) 1986, by BBS PRESS SERVICE, INC.
=================
THE EDITOR SPEAKS
=================
"Low-Cost packet switching Service Threatened"
by Alan R. Bechtold
As described in our lead news story this issue, the FCC is now considering a
major change in the way packet switched phone services are defined. This
change is likely to lead to the demise of many of these services, and to much
higher prices for the use of the few that will eventually remain in business.
At the risk of over-simplification, I think I should first describe just what
a packet switched networking service is. These are the services you use to
access online databases and commercial online services, such as CompuServe and
The Source, with just a local telephone call. Once you call the local Telenet
or Tymnet number, for example, and a connection is made, you are then
connected with a computer that puts you in communication with the online
services with which you wish to communicate.
This computer is handling a number of calls into the main system computer at
the same time. It takes information you send and delivers it in "packets" to
the proper destination, picks up information from the online service computer
you called, and sends it, also in "packets," back to you. All of this
communicating is done in these so-called "packets" because this allows the
network's computers to offer protocol conversion and handle several ongoing
communications sessions at the same time.
FCC regulations allow AT&T and Bell Operating Companies (BOCs) to engage in
packet switching network operations, but they must also maintain completely
separate accounting of their voice and packet switching operations. They must
also offer free local-calling access to their lines to any competitors engaged
in the packet switching service industry.
The above regulations have allowed Telenet and Tymnet, among others, to
operate at a reasonable cost in a competitive atmosphere. This is a case of
regulation of a business actually RESULTING in increased competition and lower
prices to consumers.
As things stand now, you can call any local Telenet or Tymnet access number
and use these services to inexpensively access such online services as
CompuServe, The Source, Delphi, and countless others. In addition, GTE's new
PC PURSUIT service now offers you access, through their Telenet packet
switching service, to literally hundreds of local bulletin boards in cities
all across the country--for a flat charge of $25 per month.
But, the FCC is now being asked to REREGULATE this segment of the
communications industry, eliminating the FCC requirements that AT&T and BOCs
keep separate accounting records of their voice and packet switching services,
and eliminating the stipulation that the BOCs and AT&T must offer their
competitors in the packet switching business free access to their local
telephone connection lines.
The idea is patently ridiculous.
Mark Fowler, Chairman of the FCC, has been hailed by the press as a "fair-
market zealot." The chances are very good that he views this proposed
reregulation as the magic road to increased competition and fairer pricing for
consumers.
Unofficially, the word is out that the FCC advisory committee now considering
this matter is indeed leaning in favor of the proposed reregulation of the
packet switching industry. If the committee recommends these changes, it's
likely that a majority of the five voting members on the Federal
Communications Commission will vote in favor of the changes.
I have talked to sources within the industry who say it is the BOCs who are
pushing VERY HARD for this reregulation, because they want to get into the
packet switching service business in a big way, and they would like to rid
themselves of needless competition on their way to success.
What's that? RID themselves of competition? But--the proposed reregulation
is supposed to FOSTER competition! Why would a group of companies (BOCs)
hoping to eliminate their competition PUSH for this reregulation? I hope the
answer to THAT question is entirely clear.
Here we have an industry that is currently populated with plenty of
competition. Prices are already reasonable. Reregulation of the packet
switching service industry will IMMEDIATELY give giant corporations the upper
hand, and will allow them to cut off free access to their local access phone
lines to their competitors, namely Telenet and Tymnet and other similar
services that now offer you high-quality service, in a competitive
marketplace, at reasonable prices.
The proposed reregulation, however, would force all packet switching services
to compete with the BOCs and AT&T, companies that would be able to use the
enormous profits they earn with their voice telephone services to cross-
subsidize their packet switching services and offer them on a non-compensatory
basis, at least until their competitors are eliminated. When that happens,
they are then sure to jack up their fees to any level they want.
It would also force their packet switching competitors to pay access fees for
connection to local phone lines. The access fees alone could add as much as
$4.00 per hour to the fees packet switching companies would be forced to pass
on to their customers. This will be added to your hourly connect-time charges
for accessing ALL online databases through these services.
The proposed reregulation could very well spell the death of PC PURSUIT.
Because GTE also uses dial-out modems at the other end of their Telenet
connections for PC PURSUIT service, the company would be forced to pay an
hourly charge at BOTH ends of the phone line--totaling up to $8 or $9 per
hour. These fees would have to be added to the flat $25 per month that GTE
now charges for access to PC-PURSUIT. It would simply make the final cost to
PC-PURSUIT customers too high for the service to remain practical and
affordable.
So--this is ONE TIME you MUST use your word processor to produce some letters
opposing this proposed reregulation! Write to:
Honorable Mark Fowler
Chairman of the Federal Communications Commission
Washington D.C. 20554
Refer to Computer Inquiry III in your letters. State clearly, in your own
words, that competitive packet switching services should not be reregulated or
subjected to carrier access charges, and then explain why not. Tell Mr.
Fowler that reregulation of packet switching services will completely destroy
the existing fair market for these services, and eventually increase costs,
not DECREASE them.
And hurry! I have heard this matter will be going before the FCC for a vote
in the latter part of January or early part of February. Time is running out.
==========END>>>KFL%MX.LCS.MIT.EDU@mc.lcs.mit.edu (Keith F. Lynch) (12/16/86)
From: Keith Petersen <W8SDZ@SIMTEL20.ARPA>
... Mark Fowler, Chairman of the FCC, has been hailed by the press as
a "fair market zealot." The chances are very good that he views this
proposed reregulation as the magic road to increased competition and
fairer pricing for consumers.
In a free market, it would not matter to users whether this
legislation was passed or not. The legislation does not COMPELL local
phone companies to charge four dollars or more per hour for a local
phone call to a long distance data service (e.g. PC PURSUIT) it merely
ALLOWS them to do so.
Since it doesn't cost local phone companies any more to complete a
local call to such a service than it costs them to complete any other
local call, phone companies would not lose money by not adding this
charge. And since any local phone company which chose NOT to charge
extra for such calls would get plenty of business from users who
formerly used any local phone company which DID decide to add the
extra charge, there would certainly be local phone companies which
choose not to add this charge. This is how the free market works.
HOWEVER, we unfortunately do NOT have a free market in local
telephone service. Since each user has no choice which local phone
company to use, thanks to a pernicious government-mandated monopoly,
most local phone companies probably WILL add this charge if they are
allowed to. They know they won't lose any customers to competing
firms, since there are no competing firms allowed.
In an ideal world, this legislation would be a good thing. Phone
companies like any other company should be allowed to charge whatever
they wish for their services, subject only to the constraints of the
marketplace. But in the context of the captive marketplace, this
legislation would be a very bad thing. If phone companies are given
a monopoly, their prices have to be regulated by the government, since
they are not regulated by the free market. Without regulation, they
would be able to charge as much as they could without people abandoning
phone service for bicycle messengers or carrier pigeons.
Phone service ought to cost the user just a few percent more than
the cost to the phone company of providing the service. In a free
market, it would. In a regulated mandated monopoly, it might (how
could anyone ever tell?). But given an unregulated mandated monopoly,
i.e. the worst of both worlds, the local phone companies will sell
their services for slightly less than the cost to the user of doing
completely without phone service.
If Mark Fowler is indeed an advocate of the free market system, this
is how it should be explained to him.
Write to:
Honorable Mark Fowler
Chairman of the Federal Communications Commission
Washington D.C. 20554
Refer to Computer Inquiry III in your letters. ...
And hurry! I have heard this matter will be going before the FCC for a
vote in the latter part of January or early part of February. Time is
running out.
I completely agree. Write today!
Please reply to me, I am not on most of these lists.
...KeithOCONNORDM@ge-crd.arpa (OCONNOR DENNIS MICHAEL) (12/16/86)
Date: 16-DEC-1986 13:30 Sender: OCONNORDM Subject: Re: Re: Pending FCC ruling threat to modem users To: Info-Micro@BRL.ARPA@SMTP, Unix-Sources@BRL.ARPA@SMTP, To: Info-XMODEM@simtel20.arpa@SMTP, Info-Cpm@amsaa.arpa@SMTP -------- People have written lots on this new FCC rule change ( threat? threat?? ). Lots of stirring propaganda about regulators who beleive in fairy tales, monopolies that will screw you to the wall given the chance, and legislated monopolies. Lots of misconceptions. First: MODEM calls DO NOT cost the phone company the same amount as other calls. They tend to be longer, and don't tolerate noise as well. Anyone who knows ANYTHING about the phone system knows it can only handle some fraction of the possible calls that might be happening at any one time, and the longer the average phone call is, the more equipment will be needed to meet this fraction. If the phone company always charged for local service BY THE MINUTE, well, no problem, but phone companies usually charge BY THE CALL or BY THE MONTH. So heavy modem users are currently being SUBSIDIZED by the rest of the users. Sounds like it MIGHT be UNFAIR. 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. The problem is you and I would lose big money trying to compete with the phone system. And our users would be annoyed at people next door using BELL being a long-distance call. But you can do it, in fact, General Electric HAS done it, for both its local and long-distance telephone needs ( known as DIALCOM ). If ANYBODY needs something explained to him, it is probably NOT Mark Fowler. Before people rush to pester him, why don't you all invite someone from the Phone Companies to give THEIR SIDE. This rush to get half-informed people to rise up and make trouble is simply electronic rabble-rousing, NOT what democracy thrives on : informed opinion from people who have been exposed to ALL SIDES of an issue. ( DISCLAIMER : I'm not neccesarily disagreeing with anyone, nor do these opinions represent anybody elses. I could even be wrong. But then again, so could you. 'Cause remember ... No matter where you go, There you are. ) Dennis O'Connor -------- --------
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?
frank@dvm.UUCP (Frank Wortner) (12/17/86)
In article <1575@brl-adm.ARPA> OCONNORDM@ge-crd.arpa (OCONNOR DENNIS MICHAEL) writes: > > >First: MODEM calls DO NOT cost the phone company the same amount as >other calls. They tend to be longer, and don't tolerate noise as well. >... If the phone company always >charged for local service BY THE MINUTE, well, no problem, but >phone companies usually charge BY THE CALL or BY THE MONTH. So heavy >modem users are currently being SUBSIDIZED by the rest of the users. > Dennis O'Connor > Correct me if I'm wrong, but some phone companies (I'm thinking of New York Telephone in this case) charge their customers by the MESSAGE UNIT. This method of billing takes both TIME and DISTANCE into account. I don't notice New York Tel giving me a discount if I keep my calls short, sweet, and voice-only. In fact, their advertising seems to encourage me to make numerous and lengthy calls to my family, friends, and business associates. New York Tel gets its due from me for my voice-based calls, and I don't see the company failing to obtain fair payment for my modem-based calls. I also don't see (hear?) the company striving to make sure that my modem calls receive a better connection than my voice calls. If telephone noise were significantly lower, there would be little need for error- correcting modems. Phone companies would be much better off if they upgraded their billing systems than if they imposed unfair charges against only one class of users (-- modem users). This is only logical. Afterall, doesn't a company which charges users by the month lose just as much money on a lenthy call to Grandma as on a lengthy call to a BBS? In fact, I would wager that more people have grandmothers than have computers. OK, enough ranting -- off the soapbox I go! -- Frank ...!inhp4!allegra!phri!orville!dvm!frank
chris@mimsy.UUCP (Chris Torek) (12/18/86)
In article <1575@brl-adm.ARPA> OCONNORDM@ge-crd.arpa (OCONNOR DENNIS MICHAEL) writes: >First: MODEM calls DO NOT cost the phone company the same amount as >other calls. They tend to be longer, and don't tolerate noise as well. >... If the phone company always charged for local service BY THE >MINUTE, well, no problem, but phone companies usually charge BY THE >CALL or BY THE MONTH. I may well be wrong---I am no expert on telephone systems---but as I understand it, in most areas there is a twisted pair of wires running from your telephone all the way to the local Central Office. When you make a local call that does not require routing between different COs, this ties up only the two twisted pairs and whatever it takes to cross-connect them. It is only when you use a long- distance trunk that you begin using a shared resource. (I am not at all certain that the connection at the CO is `free', though.) As for noise, I have never had a modem exhibit noise trouble unless there was quite audible line noise, which I, too, would not tolerate well. -- In-Real-Life: Chris Torek, Univ of MD Comp Sci Dept (+1 301 454 7690) UUCP: seismo!mimsy!chris ARPA/CSNet: chris@mimsy.umd.edu
jbs@mit-eddie.MIT.EDU (Jeff Siegal) (12/19/86)
In article <126@dvm.UUCP> frank@dvm.UUCP (Frank Wortner) writes: >In article <1575@brl-adm.ARPA> OCONNORDM@ge-crd.arpa (OCONNOR DENNIS MICHAEL) writes: >>First: MODEM calls DO NOT cost the phone company the same amount as >>other calls. They tend to be longer, and don't tolerate noise as well. >>... If the phone company always >[...] >I also don't see (hear?) the company striving to make sure that my modem >calls receive a better connection than my voice calls. [...] You're both wrong. Modem calls do cost the phone company more and length has nothing do do with it. When you call your friendly BBS (or your grandmother), the phone company's equipment does not simply throw a few switches and connect you. Your call gets routed over inter-office trunks, which are shared between many calls. One of the ways they make finite trunk capacity go further is by multiplexing calls. For example, in the time between the words "hello" and "Granny", someone else is using the channel you're not using (because you're being silent). For normal conversation, this "silent time" is a substantial percentage of the call time. With modem calls, there is no idle time. Even when no data is being transfered (and data is often transfered continuously for long periods of time), the channel is being used by the modems which need to exchange carrier signals. Jeff
rb@cci632.UUCP (Rex Ballard) (12/20/86)
In article <4851@mimsy.UUCP> chris@mimsy.UUCP (Chris Torek) writes: >In article <1575@brl-adm.ARPA> OCONNORDM@ge-crd.arpa (OCONNOR DENNIS >MICHAEL) writes: >>First: MODEM calls DO NOT cost the phone company the same amount as >>other calls. They tend to be longer, and don't tolerate noise as well. Length is the critical factor here. Noise is a factor of the switching system. It is only a cost factor if you call to complain about the noise. >>... If the phone company always charged for local service BY THE >>MINUTE, well, no problem, but phone companies usually charge BY THE >>CALL or BY THE MONTH. >I may well be wrong---I am no expert on telephone systems---but as >I understand it, in most areas there is a twisted pair of wires >running from your telephone all the way to the local Central Office. >When you make a local call that does not require routing between >different COs, this ties up only the two twisted pairs and whatever >it takes to cross-connect them. This was the case until about 5 years ago. Today, however, most of those twisted pairs are run into digital concentrators at the earliest possible time. In addition, whether concentrated at a codex, or at the central office, the call has to be routed through a switch, either a 5ACD (old analog), a #5ESS (new digital) or a DMS-200 (new digital). These switches are essentially very fast computers. Ever seen what happens when you put 100 users, all using 'VI' on a VAX 11/750? The same thing happens to those switches when they get a few hundred users running 2400 baud modems for a few hours of "hacking". >It is only when you use a long- >distance trunk that you begin using a shared resource. (I am not >at all certain that the connection at the CO is `free', though.) Actually, a "dedicated line" costs about $7000/month, but because of the concentrators, these lines can be shared and the costs can be split across quite a few users. >In-Real-Life: Chris Torek, Univ of MD Comp Sci Dept (+1 301 454 7690) If the phone companies are allowed to charge for "voice modem" links, they will probably still be required to provide lower cost ISDN digital links. I haven't seen the actual proposal yet. Will access charges be allowed *only* if the BOC is able to offer digital service to the "back end" system *and* the subscriber? If this is the case, then expect to be offered 9600 baud async links, or 56KB X.25 type links within a few months after the proposal is adopted. The BOCs will still be required to provide digital service to "digital service nodes" for very reasonable costs. The difference is that the interface has to be changed so that the concentration is done by the BOC rather than the carrier. There has been lots of information on ISDN published, and there is also the CCITT "yellow book" (or is it red this year :-). Disclaimer: CCI provides certain equipment to the Bell Operating Companies (among others) but the opinions expressed above are mine, and not those of CCI or the BOCs. Rex Ballard.
parker@epiwrl.UUCP (Alan Parker) (12/20/86)
In article <4334@mit-eddie.MIT.EDU> jbs@eddie.MIT.EDU (Jeff Siegal writes: >When you call your friendly BBS (or your grandmother), the phone >company's equipment does not simply throw a few switches and connect >you. Your call gets routed over inter-office trunks, which are shared >between many calls. One of the ways they make finite trunk capacity >go further is by multiplexing calls. For example, in the time between >the words "hello" and "Granny", someone else is using the channel >you're not using (because you're being silent). For normal >conversation, this "silent time" is a substantial percentage of the >call time. With modem calls, there is no idle time. Even when no >data is being transfered (and data is often transfered continuously >for long periods of time), the channel is being used by the modems >which need to exchange carrier signals. Bull. Very few (if any) trunks work this way. For the most part Digital trunks are TDM mutiplex using either 64K bps per call (for regular PCM) or 32K bps (for APCM). Older analog carrier trunks (of which there are still plenty of out there) assign a piece of the frequency domain to each call. These do not have the smarts to move calls around during silence periods. A scheme like you describe might be used by a carrier on long haul (particularly satellite) circuits. I think the original SBS system worked this way (and it didn't work well).
barber@rabbit1.UUCP (Steve Barber) (12/23/86)
In article <4851@mimsy.UUCP>, chris@mimsy.UUCP (Chris Torek) writes: > When you make a local call that does not require routing between > different COs, this ties up only the two twisted pairs and whatever > it takes to cross-connect them. It is only when you use a long- > distance trunk that you begin using a shared resource. (I am not > at all certain that the connection at the CO is `free', though.) > -- > In-Real-Life: Chris Torek, Univ of MD Comp Sci Dept (+1 301 454 7690) > UUCP: seismo!mimsy!chris ARPA/CSNet: chris@mimsy.umd.edu Well, you are correct that all it takes are the twisted pairs and the cross-connect stuff, but the cross-connect stuff *IS* a shared resource also. Each CO has a finite number of switches and that number is very much smaller than the (number of lines / 2) figure that would be required if everyone called someone else at the same time. The number of switches in an office is based on the average frequency and duration of calls through that office: modem calls are longer (on the average) than voice calls (no matter how many teenage girls there are out there). They consume more resources (switches), causing the phone company to either install more switches or tolerate a lower successful connection rate. Since reliability, or rather perceived reliability, is important, they usually will install more switches, causing a capital outlay to be passed on to consumers. These economics have been true since the beginning of the Bell System up through and including the ESS 5 switch (I don't know if there have been any since then). So, regretfully, data calls do cost the phone company more, and equal rates for voice and data cause voice-only users to "subsidize" data users if someone wishes to perceive things that way. Too bad. Not that I relish paying extra (and I won't if I don't have to), but that's the way it looks from where I sit. -- Steve Barber Rabbit Software Corp. ...!ihnp4!{cbmvax,cuuxb}!hutch!barber ...!psuvax1!burdvax!hutch!barber (215) 647-0440 7 Great Valley Parkway East Malvern PA 19355
sjl@amdahl.UUCP (Steve Langdon) (12/23/86)
In article <886@epiwrl.UUCP> parker@epiwrl.UUCP (Alan Parker) quotes jbs@eddie.MIT.EDU (Jeff Siegal) describing how the silences in speech can be supressed to allow more efficient use of trunks. He then says: > Bull. Very few (if any) trunks work this way. For the most part > Digital trunks are TDM mutiplex using either 64K bps per call > (for regular PCM) or 32K bps (for APCM). Older analog carrier trunks > (of which there are still plenty of out there) assign a piece of the > frequency domain to each call. These do not have the smarts to move > calls around during silence periods. > > A scheme like you describe might be used by a carrier on long haul > (particularly satellite) circuits. I think the original SBS system > worked this way (and it didn't work well). Alan is correct in his assertion that normal interexchange trunks do not use the technique Jeff described. The standard acronym associated with this technique is TASI (Time Assigned Speech Interpolation?). It was originally developed for use on high cost routes such as trans-Atlantic cables (not satellites). Given the cost of the equipment involved use is still restricted to this type of application. A related (but different) way of achieving the bandwidth compression is packet switched voice which can be found in new products just starting to enter the market (e.g. StrataComm <sp?>). -- Steve Langdon ...!{decwrl,sun,hplabs,ihnp4,cbosgd}!amdahl!sjl +1 408 746 6970 [I speak for myself not others.]
phil@amdcad.UUCP (Phil Ngai) (12/24/86)
In article <4815@amdahl.UUCP> sjl@amdahl.UUCP (Steve Langdon) writes: > The standard acronym associated with >this technique is TASI (Time Assigned Speech Interpolation?). It was >originally developed for use on high cost routes such as trans-Atlantic >cables (not satellites). Given the cost of the equipment involved use >is still restricted to this type of application. Steve, would you care to comment on whether the availability of TAT-8 (a fiber optic trans-Atlantic cable) will affect the use of TASI? Will the new cable reduce the cost of bandwidth enough to make the use of TASI equipment uneconomical? I smell a "data calls cost the phone company more" debate here. All the situations I've seen involved Regional Bell Operating Companies trying to eliminate unlimited local calls. For local calls, particularly at night, which is when most users are home and on their modem, there really is no additional cost to the RBOC. All that is used is their local loop, which exists and is economically non-sharable, and a piece of the switch, which has to be designed for daytime peaks anyway. I don't think there are any long distance carriers who care, at least, that I have heard of. And that is where things like echo suppression and TASI come into play. -- If you had everything, where would you put it? Phil Ngai +1 408 749 5720 UUCP: {ucbvax,decwrl,hplabs,allegra}!amdcad!phil ARPA: amdcad!phil@decwrl.dec.com
G.MDP@score.stanford.edu (Mike Peeler) (12/31/86)
In article <975@chinet.UUCP> magik@chinet.UUCP (Ben Liberman) writes: >Seem that I recall a figure like 20% trunkage. If you have 100 subscriber, >no more that 20 can be calling out of the exchange (no more than 40 within >the exchange - 20 trunks). The local phone company estimates its average and peak loads and allocates capacity as a business decision. The percentage of calls that can be trunk calls thus varies from place to place. Trunk capacity increases exponentially with the number of trunk lines, not linearly, as at first you might expect. Figure out how much traffic a highway can bear as a function of its width in lanes. That's basically the same thing, and some fancy math helps (if you think statistical analysis is fancy math). Cheers, Mike -------
G.MDP@score.stanford.edu (Mike Peeler) (12/31/86)
In article <14178@amdcad.UUCP> phil@amdcad.UUCP (Phil Ngai) writes: >I don't think there are any long distance carriers who care, at least, >that I have heard of. And that is where things like echo suppression >and TASI come into play. Right. Echo suppression does show, however, that data calls must be detectable... For those who don't know what we're talking about, there's a little device on long-distance lines called an echo-suppressor that briefly cuts off your signal so you don't get an echo, which telephony research has found tends to be a bit unsettling to the person speaking. On a data call, this loss of signal tends to be interpreted as carrier drop, so naturally there's another little device called an echo-suppressor suppressor. The moral is, when you make a data call, TPC knows... -------
dmt@mtunb.UUCP (Dave Tutelman) (01/01/87)
In article <2066@brl-adm.ARPA> G.MDP@score.stanford.edu (Mike Peeler) writes: >In article <975@chinet.UUCP> magik@chinet.UUCP (Ben Liberman) writes: >>Seem that I recall a figure like 20% trunkage. If you have 100 subscriber, >>no more that 20 can be calling out of the exchange (no more than 40 within >>the exchange - 20 trunks). > >The local phone company estimates its average and peak loads and >allocates capacity as a business decision. True. It's a business decision based on telephone calling pattern history. The decision tends to be made company-wide; that is, the decision is embodied in a standard that says the probability of your call being blocked due to no more capacity. A "typical" such standard might be: Less than 1% blocking during the busy hour of the ten busiest days of the year. >The percentage of >calls that can be trunk calls thus varies from place to place. Close. The 20% figure above is, as you correctly imply, a number that varies from place to place. However, it accounts for not only the percentage of calls that are interoffice (trunk) calls, but also the fraction of subscribers that have a connection RIGHT NOW. For instance, where the number is 20%, this could arise from: 1. Everybody talks all the time; 20% of them are on interoffice calls. 2. Subscribers talk only 20% of the time, but always call other offices. 3. Some linear mix of (1) and (2). > >Trunk capacity increases exponentially with the number of trunk >lines, not linearly, as at first you might expect. Figure out >how much traffic a highway can bear as a function of its width >in lanes. That's basically the same thing, and some fancy math >helps (if you think statistical analysis is fancy math). Not really true. The highway analogy deals with a very small number of servers in the queueing system (say, 1 to 3). In this range, I agree that small increases in servers give large increases in capacity, IF your standard is based on blocking probability. (I noted above that this IS the standard used by phone companies. But it's not the only one possible, and probably isn't the one used by highway engineers. If, for instance, you used an average server occupancy as your standard, capacity would vary linearly with number of servers, for ALL server sizes.) In any event, interoffice telephone trunk groups are much larger (tens to hundreds of servers). For this size of group, capacity varies pretty nearly linearly with number of servers, for any required blocking probability. > >Cheers, > Mike Likewise, and Happy New Year! Dave
dmt%mtunb.uucp%BRL.ARPA%taurus.BITNET@wiscvm.wisc (Dave Tutelman) (01/05/87)
In article <2066@brl-adm.ARPA> G.MDP@score.stanford.edu (Mike Peeler) writes: >In article <975@chinet.UUCP> magik@chinet.UUCP (Ben Liberman) writes: >>Seem that I recall a figure like 20% trunkage. If you have 100 subscriber, >>no more that 20 can be calling out of the exchange (no more than 40 within >>the exchange - 20 trunks). > >The local phone company estimates its average and peak loads and >allocates capacity as a business decision. True. It's a business decision based on telephone calling pattern history. The decision tends to be made company-wide; that is, the decision is embodied in a standard that says the probability of your call being blocked due to no more capacity. A "typical" such standard might be: Less than 1% blocking during the busy hour of the ten busiest days of the year. >The percentage of >calls that can be trunk calls thus varies from place to place. Close. The 20% figure above is, as you correctly imply, a number that varies from place to place. However, it accounts for not only the percentage of calls that are interoffice (trunk) calls, but also the fraction of subscribers that have a connection RIGHT NOW. For instance, where the number is 20%, this could arise from: 1. Everybody talks all the time; 20% of them are on interoffice calls. 2. Subscribers talk only 20% of the time, but always call other offices. 3. Some linear mix of (1) and (2). > >Trunk capacity increases exponentially with the number of trunk >lines, not linearly, as at first you might expect. Figure out >how much traffic a highway can bear as a function of its width >in lanes. That's basically the same thing, and some fancy math >helps (if you think statistical analysis is fancy math). Not really true. The highway analogy deals with a very small number of servers in the queueing system (say, 1 to 3). In this range, I agree that small increases in servers give large increases in capacity, IF your standard is based on blocking probability. (I noted above that this IS the standard used by phone companies. But it's not the only one possible, and probably isn't the one used by highway engineers. If, for instance, you used an average server occupancy as your standard, capacity would vary linearly with number of servers, for ALL server sizes.) In any event, interoffice telephone trunk groups are much larger (tens to hundreds of servers). For this size of group, capacity varies pretty nearly linearly with number of servers, for any required blocking probability. > >Cheers, > Mike Likewise, and Happy New Year! Dave