martillo@cpoint.UUCP (Joacim Martillo) (01/06/89)
Last month in Hamburg, West Germany, Joseph Weizenbaum chaired a three day meeting of the Computer Scientists' Forum for Peace and Social Responsibility. The meeting warned that "with ISDN a crucial prerequisite for the Orwellian police state has been created." Clearly, lots of computer scientists are losing it. ISDN is a ridiculous technology which does not provide anything which is not already available through the current public phone and data networks. The comparable ISDN service will probably be inferior but cost more. On the other hand by giving the end user direct access to the digital signaling stream, ISDN could well develop into a cracker's dream. Somehow Weizenbaum believe that ISDN will increase efficiency that will be used to wipe out jobs. Now a correct viewpoint is to assume that inefficient production methods threaten jobs because it puts the company at risk. If Weizenbaum is worried about the elimination of jobs, he should worry more about the existent tax structure of the Western industrial nations which encourage businesses to increase debt rather than increase the intrinsic value of the firm which is usually associated with job creation.
larry@pdn.UUCP (Larry Swift) (01/09/89)
In article <1943@cpoint.UUCP> martillo@cpoint.UUCP (Joacim Martillo) writes: >Clearly, lots of computer scientists are losing it. ISDN is a Losing what? >ridiculous technology which does not provide anything which is not >already available through the current public phone and data networks. >The comparable ISDN service will probably be inferior but cost more. This appears to be a nonsensical claim unless you can offer substantiation. Do you have any? The increased bandwidth alone makes the offering very interesting. Larry Swift UUCP: {peora,uunet}!pdn!larry Paradyne Corp., LG-129 Phone: (813) 530-8605 P. O. Box 2826 Largo, FL, 34649-9981 She's old and she's creaky, but she holds!
martillo@cpoint.UUCP (Joacim Martillo) (01/12/89)
In article <5291@pdn.UUCP> larry@pdn.UUCP (0000-Larry Swift) writes: >In article <1943@cpoint.UUCP> martillo@cpoint.UUCP (Joacim Martillo) writes: >>Clearly, lots of computer scientists are losing it. ISDN is a >Losing what? >>ridiculous technology which does not provide anything which is not >>already available through the current public phone and data networks. >>The comparable ISDN service will probably be inferior but cost more. >This appears to be a nonsensical claim unless you can offer >substantiation. Do you have any? Extension phones are not possible with ISDN. (The phone company always wanted to charge you for this.) For no good reason, the proposed ISDN numbering scheme is incompatible with the current numbering scheme which makes it very likely that ISDN users will be have difficulty setting up calls which interwork to a non-ISDN users. ISDN users will effectively be stuck on isolated desert islands. All the ISDN packet switches and modules which I have seen have miserable performance since they are designed and manufactured by people who really only understand voice. Look at it this way. AT&T products are low quality and ridiculously expensive. AT&T is a major player in standardizing ISDN. Worse the European PTTs are the major driving force behind ISDN. The PTTs are not comparable to AT&T but rather are comparable to the US Postal Service. The US Postal Service cannot deliver mail cheaply, efficiently and reliably. Would you really put the US Postal Service in charge of defing a major new communications technology? >The increased bandwidth alone makes the offering very interesting. What increased bandwidth? You can already lease 56kbps lines or T1 lines? An ISDN-PBX-LAN would be significantly lower bandwidth than most other LAN technologies. I suppose there might be some value to ISDN in providing remote connectivity but if I were a network administrator I suspect I could establish remote connectivity more cheaply by judiciously establishing point-to-point leased lines between remote networks. The bottom line is that ISDN because of the nature of the providers is mostly oriented to providing switched physical circuits, which really are not of terribly great use in genuine computer networking and resource sharing.
scott@attcan.UUCP (Scott MacQuarrie) (01/14/89)
In article <1960@cpoint.UUCP> martillo@cpoint.UUCP (Joacim Martillo) writes: >All the >ISDN packet switches and modules which I have seen have miserable >performance since they are designed and manufactured by people who >really only understand voice. Judging by the tone of the article I presume this statement is directed at AT&T. If you are implying that the technical and R&D staff of AT&T are only capable of understanding voice technology you are sadly mistaken. Many of the technologies developed by AT&T have a primary usefulness in the computer (or "data") environment. Three quick examples are the transistor, the UNIX operating system, and the C development language. If you require even more examples of contributions by AT&T Technical staff to the "Data" world, the list is quite long. Not to mention advances in area completely unrelated to Data or voice, except in the most indirect manner - Astrophysics for example. If you check the facts and the track record you will find that your statement is simply wrong - and insulting to those of us within AT&T who DO UNDERSTAND THINGS OTHER THAN VOICE. >Look at it this way. AT&T products >are low quality and ridiculously expensive. AT&T is a major player >in standardizing ISDN. While I am not prepared to argue price alone with you (pointless without reference to performance), I do take exception to your statement concerning "low quality". AT&T has a world famous reputation for Quality Assurance (One of the standard texts on the subject was written by an AT&T employee, thank you very much). Our products, regardless of what you think they can do (seperate arguement), are one of the most reliable products on the market - particularly in the voice environment. They have been known to survive an almost unbelievable amount of abuse and remain functional, as well as having an extremely long life expectancy. In reference to the Data product line, which I am the most familiar with, I have personally seen a 3B2/600 get dropped off a loading bay, with no packaging around it, unto solid concrete 3.5 meters below. This demo unit is still functioning two years later in our facilities (although the entire unit has a noticable twist to it). That really doesn't sound like low quality to me - how about you? What this boils down to is that your statement are insulting and ungrounded in fact. -- ------- Scott T. MacQuarrie -====------ Senior Technical Consultant, Toronto Branch, AT&T Canada -======------ -====------- Phone: 416-756-5124 UUCP: ---------- CompuServe: 73677,102 uunet!attcan!scott or ------- ATTmail: !smacquarrie uunet!attcan!strider!scott P.S. Opinions expressed are my own and represent neither statement nor policy of AT&T or AT&T Canada Inc.
hsc@mtund.ATT.COM (Harvey Cohen) (01/14/89)
I am posting this followups to the technical newsgroups but not talk.politics, since anybody should be free to say any stupid thing he pleases in talk.politics. In article <1960@cpoint.UUCP> martillo@cpoint.UUCP (Joacim Martillo) writes: >Extension phones are not possible with ISDN. There is a sense in which this is true; ISDN phones will not work in parallel when simply wired in parallel. However, the Basic Rate Interface (BRI) spec does permit daisy-chaining. Also, an ISDN phone could be designed with a BRI-to-tip/ring converter that support standard extensions on existing wiring. >(The phone company always >wanted to charge you for this.) For no good reason, the proposed >ISDN numbering scheme is incompatible with the current numbering scheme >which makes it very likely that ISDN users will be have difficulty >setting up calls which interwork to a non-ISDN users. ISDN users will >effectively be stuck on isolated desert islands. Will everyone who believes that ISDN users will have trouble telephoning non-ISDN users, and vice versa, please send me your name and credit card number. >All the ISDN packet switches and modules which I have seen have miserable >performance since they are designed and manufactured by people who >really only understand voice. Look at it this way. AT&T products >are low quality and ridiculously expensive. AT&T is a major player >in standardizing ISDN. Worse the European PTTs are the major >driving force behind ISDN. The PTTs are not comparable to AT&T >but rather are comparable to the US Postal Service. The US >Postal Service cannot deliver mail cheaply, efficiently and reliably. >Would you really put the US Postal Service in charge of defing >a major new communications technology? Trust this guy - He must know something or he wouldn't have such strong opinions! (This is the first time I have ever seen a protocol spec criticized ad hominem.) Could we please confine ourselves to technical discourse, or at least decorate our heated opinions with some relevant facts and logic? -- Harvey S. Cohen, AT&T Bell Labs, Lincroft, NJ, mtund!hsc, (201)576-3302
perl@step.UUCP (Robert Perlberg) (01/18/89)
Thank you for the company line. Now let's cool down a bit and examine
what Joacim was talking about:
The discussion involved one single AT&T device: a switch. You assert
that AT&T knows what "it's" doing in the data area because they have
"given" us UNIX and C. UNIX & C are software products which were
developed by people who probably aren't even allowed into the labs
which develop data switches. Just because a company produces a good
product doesn't mean that the people who worked on that product work on
all of the other products that come out of that company. In a company
the size of AT&T, that's just not possible.
Your inference that AT&T knows a lot about data because they invented
the transistor gave me a good laugh. Besides my point about UNIX and C
being applicable here as well, it's like saying that the first caveman
who discovered fire knew as much about building cars as Henry Ford. A
lot of companies try to impress people by pointing out that they were
the first to use a certain technology, but my observation has been that
the first is seldom the best since it is always the competition who
improve on the original design.
What Joacim was trying to say is that the AT&T data switches were
designed by people who know more about voice than data. While AT&T may
have some very talented data communications people, they apparently
were not involved in the design of the data switches in question.
Robert Perlberg
Dean Witter Reynolds Inc., New York
phri!{dasys1 | philabs | mancol}!step!perl
-- "I am not a language ... I am a free man!"frg@jfcl.dec.com (Fred R. Goldstein) (01/20/89)
In article <1960@cpoint.UUCP> martillo@cpoint.UUCP (Joacim Martillo) writes: [various flames deleted] Yakim, you're always entertaining, even if your facts aren't always straight! C'mon, you went to a few T1D1 meetings yourself -- do you really think that AT&T was running the show? Just because Don Stanwyck is very effective doesn't mean that the rest of us don't have a significant say. Besides, Don _does_ understand more than voice. (Even if some of the delegation doesn't...) >Extension phones are not possible with ISDN. (The phone company always >wanted to charge you for this.) This is not exactly true. Extension phones are indeed a casualty of the ISDN S-bus interface. The Passive Bus is truly brain-dead, to use a net phrase, but then the User-Network Interface is at U or T, not as S. I rather like the idea of putting a Terminal Adapter in the basement, keeping the analog phones in the house, and running the ISDN digital interface up to the computer! I wonder what interesting hardware work-arounds will be made for this. >For no good reason, the proposed >ISDN numbering scheme is incompatible with the current numbering scheme >which makes it very likely that ISDN users will be have difficulty >setting up calls which interwork to a non-ISDN users. ISDN users will >effectively be stuck on isolated desert islands. Just not true. E.164 is a superset of E.163. In North America, the ISDN Numbering Plan will include the telephone numbering plan, and all ISDN phones will be able to dial all POTS phones. SOME POTS phones won't be able to dial SOME ISDN devices prior to 1996 or so, IF the numbering plan (interim) for non-LEC carriers isn't done right, but I don't even think that'll be the case. Now Swissnet, the national data ISDN in CH, will be an island at first, but then Swiss people aren't supposed to _use_ telecommunications, they're supposed to look at the fine Swiss craftsmanship and don't touch or else! (Ever price a call from CH to USA?) >>The increased bandwidth alone makes the offering very interesting. > >What increased bandwidth? You can already lease 56kbps lines or T1 >lines? An ISDN-PBX-LAN would be significantly lower bandwidth >than most other LAN technologies. I suppose there might be some >value to ISDN in providing remote connectivity but if I were a >network administrator I suspect I could establish remote connectivity >more cheaply by judiciously establishing point-to-point leased lines >between remote networks. Absolutely backwards, with regard to WANs. Today, 56 kbps lines cost about 3 times as much as bananalogue voice-grade circuits, and typically take months to get, since they're special-engineered on the pole. Comes the Revolution, every POTS line will provide 64 kbps for voice, so it'll take three days to install 64 kbps, and it'll cost little more (or same as) voice. SWITCHED 64 will cost like switched voice, more than private lines, but the economic basis of high 64 kbps leased line rates will go away too. As a LAN technology, of course, ISDN is as useful as socks on a snail. (Gee, isn't it fun to have Yakim to flame at? Hmmm, this isn't supposed to be a flame group. Sorry.) fred