Path: utzoo!utgpu!jarvis.csri.toronto.edu!cs.utexas.edu!tut.cis.ohio-state.edu!mailrus!accuvax.nwu.edu!nucsrl!telecom-request From: KLH@nic.ddn.mil (Ken Harrenstien) Newsgroups: comp.dcom.telecom Subject: Communications with the Deaf Message-ID: <4261@accuvax.nwu.edu> Date: 21 Feb 90 23:29:11 GMT Sender: news@accuvax.nwu.edu Organization: TELECOM Digest Lines: 137 Approved: Telecom@eecs.nwu.edu X-Submissions-To: telecom@eecs.nwu.edu X-Administrivia-To: telecom-request@eecs.nwu.edu X-Telecom-Digest: Volume 10, Issue 123, Message 1 of 5 Naturally, as soon as I get bored and look away for a couple weeks, something interesting finally pops up! I see that the basic points have been covered pretty well by Curtis Reid (howdy!) and others, so I just have some more-or-less scattered comments to make. Regarding: [Moderator's Note: Your comments about AT&T pretty well sum up what I said earlier: Both AT&T and the Bell Companies have always been very responsive to the needs of their handicapped customers. Neither MCI or Sprint seem concerned at all. AT&T spends *a lot* of money staffing and maintaining 800-855-1155 around the clock 365 days per year. PT] I would have to respectfully disagree, particularly where the "Bell Companies" are concerned. In most cases that I'm aware of, the telcos had to be dragged kicking and screaming into providing handicapped services (let's see, is that "services for the handicapped" or "services that are handicapped"?). You just don't hear about that part because the instigators don't have an advertising budget. After the fact, of course, the existence of these services is always good copy. The 800/855-1155 TDD operator service is STRICTLY an operator service. They do not relay calls, and as far as I know, no one is currently providing an interstate relay service, which is a source of great frustration as I personally must make a lot of voice calls out of state (how many of you do this daily?). Without access to interpreters at work, I would have no recourse other than to ask friends to make the calls for me, which has a greasy feeling akin to begging for quarters in the street. The California Relay Service here is reportedly working with the telcos and LD carriers to figure out how they can incorporate interstate LD charges into their billing system. It's obviously not a new idea, but no one knows when it will finally happen. So AT&T spends *a lot* of money on 800/855-1155? So what? I pay AT&T *a lot* of money for my toll calls. Remember that they spend *many megalots* of money on everything else, and at any rate the connections between services and charges have historically been exceedingly tenuous (a famous example is LD subsidized by local calls). ----------------------- Hmmmm. Anyway, about TDDs themselves. There is no shortage of ideas for them, but the big problem is the lack of a market, or rather the perception thereof. At SRI International in 1977-78 we (primarily Wolfram & Fylstra) developed a portable TDD, about a handset-breadth wide and a third that in depth, which could handle both Baudot/Weitbrecht and ASCII/103 (the latter at either 110 or 300 baud) -- they used an ingenious trick to fool a single modem chip into supporting both. This even included a tiny integral printer cannibalized from a HP calculator, and unlike many "modern" TDDs it could generate and represent the complete ASCII character set. Four working prototypes were built. This development was funded by a HEW grant, and the main idea was to come up with something useful enough to the commercial world that the device could be mass-produced in enough quantity to be available at low cost to deaf users. A secondary motivation was to start shifting the TDD standard from Baudot/Weitbrecht to conventional datacomm protocols as soon as possible; I trust the reasons are obvious. Unfortunately, not one company in the US or abroad, including TDD manufacturers, was interested in producing it. Why? Many piddly reasons, but basically, no one was convinced that a "sufficient" market existed for that particular device. (SRI does non-profit R&D only, by the way.) It wasn't until some clever people slipped bills through state legislatures forcing the telcos to provide TDDs to hearing-impaired subscribers (and collect the money from other rate-payers) that manufacturers suddenly sat up and saw dollar signs. In California, at least, there are very good reasons to suspect that one struggling TDD manufacturer was responsible for the original bill, in hopes that the company would reap a windfall in telco orders. Ironically, they're long gone by now, while their good deed (such as it was) has survived and been propagated to a number of other states. But you (being a technically literate group) wouldn't believe what kinds of nonsense the telcos, and even some TDD manufacturers, trotted out at the PUC hearings over the issue of just what a "free" TDD should consist of. At the time all this happened (ca 1980) I sent long reports of the proceedings to, let's see, I think it was HUMAN-NETS at the time; I don't think TELECOM existed that far back. Basically the groups were divided into two camps: in one, the minions of Evil (Pacific Bell, GTE, Plantronics, others) opined that Baudot was "good enough for 'em", and in the other, the forces of Good (deaf organizations, Novation, SRI, others) argued for requiring dual-capability devices. At the time, the ability to someday access VANs (Value-Added Network) was an important issue, as this would permit TDD users to pay for LD-like services on a data rather than time basis, and we were already heavily engaged in experiments with "Deafnet" (e-mail services accessible via either standard). To simplify a complicated story, the final outcome was a flawed compromise which resulted in a small selection of TDDs that advertise a poorly defined "ASCII capability", more or less useless for anything but talking to other TDDs. Meanwhile, Baudot-only TDDs are still being churned out and sold. All of the few manufacturers left really and truly believe that this is what the market wants -- simplified, 3-row, rock-bottom Baudot/Weitbrecht TDDs. If adding ASCII/103 capabilities means just one more key, or just one more chip, or even just one more diode, they won't do it because it would be "too expensive". If not that, then it's "too complicated". The depressing thing is that, for the most part, the market appears to agree with them. There's an interesting backflow effect of technological advance here. It's true that better technology and a broader market has made it much easier and cheaper to use ASCII/103 standards in TDDs. However, it has also made it just as easy and cheap to perpetuate the Baudot/ Weitbrecht standard, whether or not it makes sense! A truly delicious example of this, which ranks right up there with narrow-necked ketchup bottles and stinging antiseptics, is the LETS/FIGS shift keys. Because the original TDDs were actually discarded Baudot TTYs (I used to have a Model 28 -- an amazing beast) with LETS and FIGS shifts as described by a recent message, they were much more awkward to use than a regular typewriter, even for (especially for?) anyone with some typing skills. You'd think that the advent of electronic TDDs would mean that the little CPU inside could take care of the LETS/FIGS shift state, right? Right. But, as you must have guessed by now, that's not what they did. For a long time the great majority of new TDDs persisted in retaining a LETS/FIGS key pair, sometimes disguised as "upshift" and "downshift". I no longer keep close track of the TDD market, but I'm sure you can still find many TDDs still being made with this amazing tribute to backwards user compatibility. ------------------------------------ One last thing. How relevant is any of this to most readers? Well, let me toss in one final statistic for your consideration: the median age of the hearing-impaired population is over 60. How long do you plan to live? Ken