Path: utzoo!attcan!uunet!cs.utexas.edu!csd4.csd.uwm.edu!bionet!agate!ucbvax!pro-exchange.cts.com!sschneider From: sschneider@pro-exchange.cts.com (The RainForest BBS) Newsgroups: comp.sys.apple Subject: Explanation of HAYES/v.42 Protocol Message-ID: <8909021256.AA03609@trout.nosc.mil> Date: 30 Aug 89 21:55:42 GMT Sender: daemon@ucbvax.BERKELEY.EDU Organization: The Internet Lines: 93 *************************************************************************** Recently Toby Nixon a Principal Engineer with Hayes and also our representitive to the CCITT answered a question to a user who wasn't truely understanding just what V.42 was, it's history, and it's future. The following message was written by Toby as a reply and it may help answer any questions that you may have about the subject as well. **************************************************************************** Let me first state my qualifications. I am a Principal Engineer in the Systems Engineering Department at Hayes Microcomputer Products. I am Hayes' representative to national and international standards committees, including CCITT Study Group XVII which developed the V.42 international standard for modem error control and is now finalizing the V.42bis international standard for data compression (which applies to V.42 modems only). I personal contributed to the development of V.42, and am making similar contributions to the development of V.42bis. In other words, what I say here is not guessing or what I've read in some trade journal, but from personal knowledge. Some seem to have the misconception that "V.42" is a proprietary data compression scheme from Hayes. That is not true. Recommendation V.42 is the CCITT's international standard for modem error control. It defines two protocols: a "primary" protocol in the body of the standard known as LAPM, which is based on long-existing and widely-used international standard protocols such as X.25 LAPB and Q.921 LAPD; and, an "alternative" protocol, for backward compatibility only, which provides for interworking in error-control mode with MNP Class 2-4 modems. All future extensions to V.42, including data compression, will apply to LAPM only; the alternative protocol is frozen and will not be extended. Although Hayes V-series V.42 modems do not include the proprietary, trademarked "MNP" protocol, they DO support the V.42 Alternative Protocol, and thus fully interwork with MNP Class 2 through 4 modems. In fact, since V.42 Annex A is the most correct and complete definition of "MNP 2 through 4" in existence (Gregory Pearson, the Microcom employee who was the "father" of MNP, made several corrections to the text in V.42 that have never appeared in the "official" Microcom documentation of MNP), one could say that Hayes' V.42 modems are the _best, most correct_ MNP modems in existence. But because of trademark and licensing reasons, Hayes does not call the modems "MNP" modems, but "V.42" modems which are fully compliant including the Alternative protocol. I won't dispute your contention that "MNP7 will bite the dust soon". Hayes, AT&T, IBM, and British Telecom have made major contributions to the development of a Lempel-Ziv-based data compression scheme that will become Recommendation V.42bis after the September 25-29, 1989, meeting of Study Group XVII in Geneva (at which I will be present, and presenting several papers concerning V.42bis). You seem to think that Hayes and BT have been adversarial in this process, but that is not at all true: Hayes, BT, AT&T, and IBM have been cooperating in support of a Lempel-Ziv based scheme since early in the process; Microcom and ACT have been presenting alternatives, and have been defeated. Hayes has announced its intention to add V.42bis support to its V.42 products as soon as possible after the standard is agreed. I personally know that several other modem manufacturers have similar plans, and some have delayed introduction of V.42-compilant modems awaiting V.42bis' adoption. The statement that only Hayes has V.42 now is also incorrect. ATI and MultiTech are among the other companies that are shipping, NOW, modems for which they claim full V.42 compliance. Many other companies have announced the intention to do so soon. The two major modem firmware consulting firms, R. Scott Associates and Michael Mooradian Associates, are both in the final stages of testing. Also, Silicon Systems has announced the imminent availability of a MODEM CHIPSET that wil include full V.42 compliant error control -- and Hayes has worked closely with them in the development of this set. With this availability of V.42 from RSA, MMA, and SSI, we will quickly see many other companies, including relatively low-cost sources, providing V.42 modems. I would be pleased to offer any further clarifications on the V.42 and V.42bis standards that you might desire. -- Toby Nixon, Principal Engineer, Hayes Microcomputer Products Inc. P.S. With V.42bis, a V.32 modem will be able to transmit up to 30,000bps or more on normal dial up phone lines. ***************************************************************************** ----------------------------------------------------------------------------- | UUCP: crash!pro-exchange!sschneider COMPU$ERVE : 75166,2544 | | ARPA: crash!pro-exchange!sschneider@nosc.mil GENIE : sschneider | | INET: sschneider@pro-exchange.cts.com * My son is a Georgia Tech freshman | | I work for Xerox Corporation for decent bucks but dream of Palto Alto RC | | The RainForest @ 305-434-4927 / PO Box 841422, Pembroke Pines, Fl, 33084 | -----------------------------------------------------------------------------