Thu Aug 27 08:58:51 1981 TELECOM Digest V1 #3 From JSol@RUTGERS Thu Aug 27 08:53:59 1981 TELECOM AM Digest Thursday, 27 Aug 1981 Volume 1 : Issue 3 Today's Topics: Modem Vs. Codec - a Clarification Direct Dialed Credit Card Calls - Fact or Fiction IDDD & ESS - The "#" is a return key ---------------------------------------------------------------------- Date: 26 August 1981 01:21 edt From: JSLove at MIT-Multics (J. Spencer Love) Subject: Modem vs Codec Sender: JSLove.PDO at MIT-Multics To: DJC at MIT-XX I have been asked to clarify this distinction since I mentioned the use of CODEC's briefly in TELECOM V1 #1. I will try to avoid the use of technical terms. MODEM is short for MOdulator/DEModulator. The purpose of a modem is to send a bit stream over a voice grade (or other frequency domain) channel. The primary thing is the bit stream, we don't care what it sounds like as long as it fits in the bandwidth of the voice channel. CODEC is short for enCOder/DECoder. The purpose of a codec is to send a voice over a bit stream channel. The primary thing is the voice, we don't care what the bit stream looks like as long as the bit rate is within the capacity of our digital network. I have tried to emphasize the mirror symmetry of the concepts. The reason that codec's are interesting is that we can do a much better job with digital transmission than with analog transmission. Modern information theory lets us engineer the necessary redundancy into bit streams and we can even use error correcting codes so we know that the voice will be "perfectly" reconstituted at the other end. Anyone who has consumed frozen orange juice will understand that even though the reconstitution is less than perfect, it can be done with a uniform and fairly high quality, and it can be delivered more cheaply. The contract of a codec is to reproduce the voice as faithfully as possible with as few bits per second as possible, like the modem's contract is to get as many bits per second as possible through the crummy analog channel. -- Spencer ------------------------------ Date: 26 Aug 1981 0530-PDT Sender: LEAVITT at USC-ISI Subject: Direct Dial charging of calls From: Mike Leavitt As a regular business traveler, I often make operator-assisted calls instead of direct calls just to charge those calls to a credit card or to my home number. Are there any mechanisms or plans for a system where I can dial the number I am calling, followed by a *, followed by the number I want to charge the call to? It would surely make things easier. Mike ------------------------------ Date: 26 Aug 1981 0926-PDT From: Ian H. Merritt Subject: IDDD Lauren: Why do you say 'Usually TSPS'? The TSPS is not usually consulted unless the caller is on a STEP line, of which there are a rapidly decreasing number. Any such call from ESS is sent directly to the 4E, which takes care of getting the call to the overseas sender. STEP doesn't have the intelligence to parse the variable length sequences, so anything starting with a '0' is immediately connected to the TSPS which then receives the remaining digits as the customer dials them. By the way, for those of you who don't already understand (there must be some) the trailing '#' character is used as a carriage return on any call from ESS in which there is any ambiguity in length. In the case of the IDDD calls, there is *ALWAYS* ambiguous length unless the maximum number of digits (I think it's about 17) is dialed. <>IHM<> [Of course if you don't type a carriage return to the phone, the timeout deamon will insert one for you in 4 seconds -JSOL] ------------------------------ End of TELECOM Digest **********************