randolph@cognito.Eng.Sun.COM (Randolph Fritz) (03/10/91)
I'd like to talk about another area of politics which I hope is of interest to the readers of this groups: the politics of public data communications services. So far, all our high-speed data services are both very expensive and very limited in interconnection. I can, for instance, order 64 Kbps (or faster) private circuits to interconnect any chosen points within the USA, and within most of Europe. I can also, if I meet the criteria, get an NSFnet connection at very high expense and reach academic and a small number of commericial users. Or, I can use a voice-band modem (now up to 9600 bps) to connect to most of the USA. All of these services have different problems: the high-speed services reach very few people and are exclusive besides; the low-speed services are widespread, but slow. It turns out that existing telephone circuits can be used to provide wide-spread 64 kpbs services; this is what the telephone companies call Integrated Services Digital Network (ISDN). Taking advantage of the vast amount of transmission facilities already in place for digitized voice, ISDN offers the possibility of inexpensive, widespread medium-speed services to most of the USA and Europe. The potentials are enormous: having the kind of performance we now expect only from expensive private networks available publicly, and at prices perhaps twice that of voice telephone services -- very cheaply, in other words. In Europe, the government-run telephone companies are moving rapidly to offer ISDN. In the USA things aren't going nearly so smoothly. Here in California, ISDN is trial-tariffed as a Centrex service and from my brief discussions with Pacific Bell representatives it appears that Pacific Bell is more interested in selling this service to their largest users -- who are actually the ones who have the least use for it -- than to the vast range of small-to-medium-size business and residential users, who are the people who have the most use for it. On the other hand, I have heard Pacific Bell representatives claim that the California Public Utilities Commission is reluctant to authorize the needed capital expenditure, which after all has to come out of someone's telephone bill. There are also some purely technical problems: so far ISDN switching software written by ATT does not talk to ISDN switching software written by Northern Telecomm, though both companies are planning on making their switches compatible eventually. So what I'd like to see would be some effort to get ISDN service tariffed and widely available, at least within our biggest computing metropoli: say Boston, the SF Bay Area, LA, New York. The effort would have to be two-pronged: on the one hand, the telephone companies have to be persuaded that offering this service widely would be profitable and, on the other hand, the state public utilty commissions (and perhaps the Federal Communications Commission as well) would have to be persuaded that this is a service in the public interest. What do you think? Do you agree that this is a worthwhile idea, and one worth pursuing? Does the EFF perhaps feel that this might be an area worth exploring, and perhaps worth organizing support for? nd t ou ui R Press T __Randolph Fritz sun!cognito.eng!randolph || randolph@eng.sun.com ou ui Mountain View, California, North America, Earth nd t
curt@cynic.wimsey.bc.ca (Curt Sampson) (03/10/91)
randolph@cognito.Eng.Sun.COM (Randolph Fritz) writes: > It turns out that existing telephone circuits can be used to provide > wide-spread 64 kpbs services; this is what the telephone companies > call Integrated Services Digital Network (ISDN). I talked to the BC Tel rep. who is in charge of ISDN implementation in Vancouver, recently. He says that they will be introducing ISDN lines to most modern switches in the next year or two. Anybody can get one; the cost will be a little more than that of a standard business line ($90 per month or so, I gather). More interesting yet, he said that BC Tel intends to set up a "modem pool." The idea is that people would call up the modems and say "connect me to ISDN number 987-6543." Interesting, no? If anybody is interested, I can try and dig up any written reports on this that may be available. cjs curt@cynic.wimsey.bc.ca | "Sometimes it's like a party you go to where curt@cynic.uucp | there are no lights and everyone is doing {uunet|ubc-cs}!van-bc!cynic!curt | animal impressions." -Phillip Evans on usenet
louisg@vpnet.chi.il.us (Louis Giliberto) (03/12/91)
In article <9508@exodus.Eng.Sun.COM> randolph@cognito.Eng.Sun.COM (Randolph Fritz) writes: >What do you think? Do you agree that this is a worthwhile idea, and >one worth pursuing? Does the EFF perhaps feel that this might be an >area worth exploring, and perhaps worth organizing support for? Part of the problem is that the places that would REALLY REALLY use it are the universities. Take U of Illinois for example. They have a Cray (more than 1?) , and I know that Loyola can access it. The PLATO system there can also be accessed, and the Chicago campus is linked to the Champaign-Urbana campus. How? LEASED LINES. They could fiber optic the whole thing if they wanted to, probably. The Universities can usually get it if it's needed, and the local users (businesses and individuals) ususally have to pay dearly out of their pocket. I really have no use for it (though I would like it!), and not too many BBS's carry the kind of traffic that they would need it. It seems kind of business-oriented, and that won't generate too much support, IMHO, Louis -- --------------------------------------------------------------------------- ! "As above, so below; as below, so above" -- The Kybalion ! ! "I don't trust him; he has dark hair" -- My girlfriend's mother ! ! "So I'm stupid; what's your point?" -- Me !
randolph@cognito.Eng.Sun.COM (Randolph Fritz) (03/16/91)
Louis Giliberto points out that large organizations can afford internal leased lines, that BBS's don't seem to need ISDN, and that it's pretty business-oriented and that's not too useful. Your point that large organizations can afford leased services is exactly right -- the use of ISDN is in interconnecting smaller organizations, individuals, and establishing temporary links between large organizations (consider fax). Because ISDN shares transmission lines with regular telephone service, most of the cost of an ISDN network has already been paid; the price comes in reach of (at least) small businesses and, from tariffs I've seen, individual users as well. It's potentially quite cheap -- after an initial installation fee, the rate is about twice the cost of a regular phone line and the data transmission rate is over six times that of our fastest (voice-band) modems, which would probably pay for itself after only a few loads. As for it being business-oriented -- that was also once true of the plain old telephone and the plain old modem. That works in our favor. The technical improvements in pc's and modems have been driven largely by business applications -- and we get to make use of the results. As for what we could do with it -- ISDN would enable BBS's and commercial services to offer the kinds of things currently only available on the internet -- real-time conversations, animated messages, audio. Something closer to what we wish cyberspace were, rather than what it currently is. The broader range of media will, I hope, draw in more users. So I think it's very much worth working for. nd t ou ui R Press T __Randolph Fritz sun!cognito.eng!randolph || randolph@eng.sun.com ou ui Mountain View, California, North America, Earth nd t
emv@ox.com (Ed Vielmetti) (03/19/91)
In article <1991Mar18.235612.16233@agate.berkeley.edu> cliff@garnet.berkeley.edu (Cliff Frost) writes:
But I don't think six times faster is really true now, what with V.32bis/
V.42bis modems hitting the market for less than $400 each. Suppose
you can get 38.4Kb/s with a $300 modem over voice grade lines, would you
still be so anxious for ISDN (with its $200 install and $1,500 Terminal
Adaptor)?
can you keep that v.32 line nailed up the whole day to someone who
will give you a reasonably priced internet connection? figuring that
you're going to be paying someone else $100-$300/mo for the right to
pass packets across the internet, the ISDN $200 install doesn't look
too bad. the $1500 terminal adapter, on the other hand, ugh, esp if
the concentrator on the regional network's end is similarly expensive.
I'm used to working in a large network, so I don't really know what your
concerns are, but I'd guess that the major problem facing small business
folks is network management rather than throughput.
speaking (sorta kinda) from a small business standpoint, the major
problem is finding someone who you can get connected to without making
a huge up-front investment in equipment and administrivia. network
management is tertiary; at the beginning, it doesn't matter if it's
slow or even slightly unreliable, as long as it's there. if the links
can be shown to be worth it, the extra money can be found to pay for
reliability, but people are willing to go through some non-trivial
amount of personal hassle to get it off the ground.
--
Msen Edward Vielmetti
/|--- moderator, comp.archives
emv@msen.comcliff@garnet.berkeley.edu (Cliff Frost) (03/19/91)
In article <9928@exodus.Eng.Sun.COM>, randolph@cognito.Eng.Sun.COM (Randolph Fritz) writes: |> well. It's potentially quite cheap -- after an initial installation |> fee, the rate is about twice the cost of a regular phone line and the |> data transmission rate is over six times that of our fastest |> (voice-band) modems, which would probably pay for itself after only a |> few loads. Actually, the tarrifs around here are $29.50/month for 2B+D, and since that gives you both a voice and a data line it is not much more than two regular service lines, one for voice and one for data. But I don't think six times faster is really true now, what with V.32bis/ V.42bis modems hitting the market for less than $400 each. Suppose you can get 38.4Kb/s with a $300 modem over voice grade lines, would you still be so anxious for ISDN (with its $200 install and $1,500 Terminal Adaptor)? I'm used to working in a large network, so I don't really know what your concerns are, but I'd guess that the major problem facing small business folks is network management rather than throughput. This may be totally off-base, though. It would be interesting to know exactly what kinds of things you envision doing with ISDN lines. Cliff Frost UC Berkeley
randolph@cognito.Eng.Sun.COM (Randolph Fritz) (03/19/91)
Cliff, the problem of small business in data networking is no network, no standards, and no-one to help with the problems of networking. ISDN could help with that. Also, the additional speed would be immediately useful for fax. The six times figure is correct -- you're comparing V.32 with data compression with ISDN without data compression. As far as I know, without compression V.32 only goes 9600 bps. Conversely, if comparable compression techniques are used with ISDN, you get data rates in the 100Kbps - 200Kbps range. nd t ou ui R Press T __Randolph Fritz sun!cognito.eng!randolph || randolph@eng.sun.com ou ui Mountain View, California, North America, Earth nd t
emv@ox.com (Ed Vielmetti) (03/20/91)
In article <1991Mar19.182935.17931@infoac.rmi.de> rmohr@infoac.rmi.de (Rupert Mohr) writes: >still be so anxious for ISDN (with its $200 install and $1,500 Terminal >Adaptor)? Terminaladapter here: DM 850,00 = $ 566 (Install = just plug in) TCP/IP with 64kB + header compression is fun. FTP 150 k in seconds... this is tcp/ip into an MS-DOS system, a VME card, ? do you have prices, ordering details, etc? any confirmed working with USA ISDN offerings? list of users who are happy with it? links which are using it? this isn't strictly appropriate for efftalk, I'm cross-posting to (hm) the modems group; drag the discussion there if you think it's worth pursuing. -- Msen Edward Vielmetti /|--- moderator, comp.archives emv@msen.com
rmohr@infoac.rmi.de (Rupert Mohr) (03/20/91)
cliff@garnet.berkeley.edu (Cliff Frost) writes: >still be so anxious for ISDN (with its $200 install and $1,500 Terminal >Adaptor)? Terminaladapter here: DM 850,00 = $ 566 (Install = just plug in) TCP/IP with 64kB + header compression is fun. FTP 150 k in seconds... Rupert -- ***************************************************************** ___ ____ ___ _ _ ___ ___ ___ ___ ___ ___ _ _ /__/ / / / / /\ / /__ / /__//__// /__//__ /\ / / \ / / __/_ / / /__ / / // //__ / //__ / /
hutch@ticnj.UUCP (-Robert Hutchison) (03/25/91)
There are, IMHO, a few major stumbling blocks that need to be resolved before we see ISDN gain widespread acceptance. 1) Non-Centrex ISDN tarriff. Currently, ISDN lines are very expensive to install - not because they have to be, but rather because of how the tarrifs are set up. To install, let's say, 3 ISDN lines in the same Centrex group, it may cost a few thousand dollars (US) and might cost up to a few hundred a month for basic service (plus toll charges). When Non-Centrex ISDN (BRI) is approved, we will see in the order of $50 installation and rates comparible with analog service. That should halp a bit. 2) Lack of connectivity. Currently, you can only talk ISDN-ISDN between machines that are connected using SS7. If your's is not connected to others as such, you can ISDN-ISDN with others on your same CO switch (might not be too bad for a University or large company, but then again they probably have their own PBXs anyway). The telcos are starting to fix this. 3) Lack of APIs. There are several ISDN terminal adapters on the market from companies such as AT&T, Hayes, Teleos, NCR and others. Some use an extended INT14 interface (under DOS), some NetBIOS, some Hayes AT command set), and they are all fairly different. Also, there is no widespread UNIX support (a couple of companies have made limited attempts at supplying UNIX drivers, but nothing comprehensive so far). For me, when problem 1 is fixed, I'll be ordering my lines. For many other applications developers (the ISDN pioneers), they will probably wait for progress on problem #2. For many applications developers, they will probably wait for #3. Currently, ISDN boards can be found from about $500 and up, but if you want 64Kb-B-channels to send data, you will probably have to spend about $1000 or more. Boards typically come with support tools for applications developers. None that I am familiar with use data compression, although V.42bis contains instructions for implementing compression over ISDN. If you would like more information, send me mail at ...!princeton!sinc!hutch or contact me at the address/phone below. I will no longer have access to this account as of the end of this week (3/31/91), so don't reply to this posting address. Robert Hutchison Softright, Inc. (609) 443-3028 ISDN Software Development ..!princeton!sinc!hutch SMALL AD: (hope nobody objects) If anyone reading this needs (of knows of someone who needs) ISDN drivers (UNIX or DOS) or ISDN applications, please let me know.
randolph@cognito.Eng.Sun.COM (Randolph Fritz) (03/31/91)
Robert, generally I agree with your list of ISDN stumbling blocks; I think the APIs are less of block than apparent -- we have enough APIs to get started with. As for the problem of tariffs and ISDN switch connectivity -- these are problems with potential political solutions. If we users get together and lobby the telephone companies and public utilities commissions for them, we can probably persuade them to offer the services. Do you know if the telcos have ever actually applied for non-Centrex tariffs, Robert? In California I have the impression they have not. nd t ou ui R Press T __Randolph Fritz sun!cognito.eng!randolph || randolph@eng.sun.com ou ui Mountain View, California, North America, Earth nd t