Path: utzoo!utgpu!jarvis.csri.toronto.edu!cs.utexas.edu!uwm.edu!rutgers!mcdchg!ddsw1!karl
From: karl@ddsw1.MCS.COM (Karl Denninger)
Newsgroups: comp.unix.wizards
Subject: Re: Re^2: FCC doing it again...
Summary: Huh?  Cost of maintaining a connection is smaller than setup/teardown
Message-ID: <1989Nov29.191956.10500@ddsw1.MCS.COM>
Date: 29 Nov 89 19:19:56 GMT
References: <1989Nov28.011514.4193@virtech.uucp> <246@cfa.HARVARD.EDU> <5464@internal.Apple.COM> <34661@cornell.UUCP>
Reply-To: karl@ddsw1.MCS.COM (Karl Denninger)
Followup-To: comp.dcom.telecom
Organization: Macro Computer Solutions, Inc., Mundelein, IL
Lines: 60

In article <34661@cornell.UUCP> murthy@alsvid.cs.cornell.edu (Chet Murthy) writes:
>earlw@Apple.COM (Earl Wallace) writes:
>
>>In article <246@cfa.HARVARD.EDU> wyatt@cfa.HARVARD.EDU (Bill Wyatt) writes:
>>>...
>>>I don't want extra charges either, but in addition to the above 
>>>consideration, modem calls are not the same simply because they
>>>usually last much longer than a voice call. Somewhere I read an
>>>estimate that if only 20% of household had modems in regular use,
>>>the phone system would be hoplessly bogged down.
>>>...
....

>For phone calls, though, the cost of setup and teardown is small, and is
>easily dwarfed by the cost of maintaining the connection and data transfer.

Does it?  Not really.

If so, then the "first minute" charge would be less than the recurring
minute charge.  It isn't.

To set up a call you need use of a module in the switch to parse and
interpret the digits you (or your modem) dial, as well as the resources of
the switch to actually process the call.  Once the line is open, all you are
using is physical plant; the switch has done it's job, and you are connected.
Tearing down the call again involves the switch's electronics.

>So it makes sense to charge more for time spent.  You might say that in
>that case, they should charge per-call.  But voice calls, as pointed
>out, are shorter than modem calls, hence the modem calls incur more
>load on the network.

Not if you have a teenager in the house!  I know people who do, and they
regularly get phone bills that dwarf mine -- yet I have (and use) a modem in
a measured service area, and they DO NOT have a modem.  They do have a teen
who is on the phone from 3:30 in the afternoon until 10:00 pm at night,
stopping only for dinner!

Modems are not normally used THAT extensively.  When the FCC also surcharges
anyone who has a 12-18 year old in the house, then I will accept a modem
surcharge without complaint.  Not until.

>All in all, it makes sense.  The system _does_ get more load from modems,
>so why not charge them more?  However, the _reason_ the system gets more
>load from modems is because it's designed for voice. 

Actually this is not strictly true.  Modems put their energy in a narrow
band of frequencies (Telebits excepted).  Voice is all over the 300-3khz
band that phones work with, and is thus actually uses more bandwidth.  This
is a moot point with the current multiplexing schemes in use, as they
digitize the entire spectrum -- thus there is >no< technical difference in
bandwidth use between the two.

What in the dickens is this doing in comp.unix.wizards anyway?

--
Karl Denninger (karl@ddsw1.MCS.COM, <well-connected>!ddsw1!karl)
Public Access Data Line: [+1 708 566-8911], Voice: [+1 708 566-8910]
Macro Computer Solutions, Inc.		"Quality Solutions at a Fair Price"