bunda@cs.utexas.edu (John Bunda) (01/08/89)
> ... The crushing problem with things > like TDD's (Telecommunications Device for the Deaf) is that they are > totally incompatible with everything else (they use an ancient protocol > predating even 110 baud modems) - the other party has to have one in > order to talk to them, and they aren't very common except for the > deaf and certain "core" social services (911 and similar services, at > least in some cities). I have run into this problem before, and I'm wondering how difficult it would be to get my PC to emulate a TDD - I'm guessing that it's trivial but would require hardware - I'm supposing that it wouldn't be possible to convince my modem to talk to the things. Probably easier to send a PC and modem to my deaf relative... -John -- ................................... John Bunda UT CS Dept. bunda@cs.utexas.edu Austin, Texas
emv@a.cc.umich.edu (Ed Vielmetti) (01/08/89)
I have a 1986-vintage pointer to TDD software and hardware for the IBM PC. It's the "Handicapped Educational Exchange", 11523 Charlton Drive, Silver Spring MD 20902 and the HEX BBS at (301) 593-7033. Richard Barth is the person, and his voice phone two years ago was (301) 681-7372. I have no current abiding interest in this, but if someone else checks things out please post or e-mail or fax me. If I were to lose my hearing tomorrow, I'd buy a fax machine -- you're more likely to be able to communicate with lots of people than via TDD. --Ed Edward Vielmetti, U of Michigan Computing Center electronic communications group fax # 313-763-4434
bcw@rti.UUCP (Bruce Wright) (01/08/89)
In article <4521@cs.utexas.edu>, bunda@cs.utexas.edu (John Bunda) writes: > > > ... The crushing problem with things > > like TDD's (Telecommunications Device for the Deaf) is that they are > > totally incompatible with everything else (they use an ancient protocol > > predating even 110 baud modems) - the other party has to have one in > > order to talk to them, and they aren't very common except for the > > deaf and certain "core" social services (911 and similar services, at > > least in some cities). > > I have run into this problem before, and I'm wondering how difficult > it would be to get my PC to emulate a TDD - I'm guessing that it's > trivial but would require hardware - I'm supposing that it wouldn't > be possible to convince my modem to talk to the things. > > Probably easier to send a PC and modem to my deaf relative... TDD's use a modem signalling protocol known as "simplex". There is no carrier signal (as you may already know). There are two separate frequencies used for Mark and Space, which are only generated when one side or the other wants to send something. The frequencies are the same whether you are the Answer or the Originate side of the connection - and there is no provision for relinquishing the line (as there is in half duplex), so if both parties try to talk at once you get garbage. The frequencies used for Mark and Space are different from the frequencies used by any other modem signalling technology that I'm aware of. Simplex was abandoned by most communications carriers back in the 50's and 60's because it is just too primitive to be useful for most purposes. Now once you get the signal demodulated, the problem becomes easy: they use a Baudot-type character code (actually, there is no single "Baudot" character code, there's a whole family of them; I forget which one the TDD's use. The differences are relatively minor, though they sometimes move around the Shift-In and Shift-Out codes so you may not get anything meaningful by applying the wrong Baudot-ASCII converter). If you have a fair amount of ability with signalling hardware it would probably not be too difficult to build a Simplex modem. They don't run at any kind of speed (something like 50 baud) so there isn't any heavy timing constraint. Unfortunately I don't remember enough specs on the subject to be able to tell you what you would have to do to recognize the signals, but this may give you enough to try to look up references. Better yet, it is likely that one or another TDD manufacturer has a Simplex -> ASCII (or at least Simplex -> RS-232) converter; if this exists it would simplify your task considerably. Good luck! Bruce C. Wright
darrelj@exec.sm.unisys.com (Darrel VanBuer) (01/08/89)
TDDs differ from more common computer communications as follows:
Baudot instead of ASCII (handle in software with a couple of tables
and shift state flags)
45 baud (60 WPM in the Baudot 5 level code) instead of 110, 300, 1200,
2400 ... (IBM PC serial I/O chip can be programmed for this, but not
via the BIOS)
5 data bits instead of 7 or 8, no parity (IBM PC serial chip can do it)
1.5 stop bits instead of 1 or 2 (any serial I/O chip which does 5 data
bits usually provides 1.5 stop capability, even those that don't will
interoperate by programming 2 stop bits since the receiver only needs
one, and sending 2 instead of 1.5 justs costs 7% of speed)
Modem uses FSK half duplex signalling which uses frequencies
incompatible with any other modem (something like 1400 Hz mark, 1800
Hz space, silent when idle). I faked it once by using a Motorola
Modem IC, a fast crystal (to shift freq up) and a couple of inverters
to fix the data sense [it was a little off spec, but just fine with
the wide tolerance detectors used at such a slow data rate].
The deaf community is beginning to use dual standard machines (old
plus ASCII/Bell 103), but large numbers of existing equipment in the
hands of many who cannot afford new equipment make this a very slow
process. (And the need to change is low for them since even 60 WPM is
faster than most people can type).
Darrel J. Van Buer, PhD; unisys; 2400 Colorado Ave; Santa Monica, CA 90406
(213)829-7511 x5449 KI6VY darrel@CAM.UNISYS.COM or
..{allegra,burdvax,cbosgd,hplabs,ihnp4}!sdcrdcf!darreljbrent@rtech.rtech.com (Brent Williams) (01/10/89)
From article <4521@cs.utexas.edu>, by bunda@cs.utexas.edu (John Bunda): > >> ... The crushing problem with things >> like TDD's (Telecommunications Device for the Deaf) is that they are >> totally incompatible with everything else (they use an ancient protocol >> predating even 110 baud modems) - the other party has to have one in >> order to talk to them, and they aren't very common except for the >> deaf and certain "core" social services (911 and similar services, at >> least in some cities). > > I have run into this problem before, and I'm wondering how difficult > it would be to get my PC to emulate a TDD - I'm guessing that it's > trivial but would require hardware - I'm supposing that it wouldn't > be possible to convince my modem to talk to the things. > > Probably easier to send a PC and modem to my deaf relative... > I have gone through the same version of hell trying to talk to my deaf cousin. The TDD's use *45* baud acoustic couplers. I think Teletypes went to 110 Baud in the 60's. This means you can out-type TDD's relatively easily. Also, the things are incredibly poorly designed from an ergonomic standpoint. There's a 1-line, ~24 character screen. Unfortunately, the California Deaf Relay Service isn't that much help, since they type too fast for many deaf people (including my cousin) to read. My cousin told me he prefers to write letters rather than trying to use the Deaf Relay because it's so frustrating for him. The company that makes these TDD terminals is undoubtedly making mondo $$ selling them at goverrnment contract prices to Pac Bell. For about $800 you could equip deaf people with a Toshiba T-1000 (or other low-end laptop) and they would have a real computer as well as a powerful communications device. If you've got the hardware to spare, a PC w/modem is the way to go. A fax machine sounds tempting, but 2-way communication with a fax would probably be somewhat annoying. -brent -- -brent williams Relational Technology, Inc. 1080 Marina Village Parkway {amdahl,sun,mtxinu,cpsc6a,hoptoad} Alameda, CA 94501 !rtech!brent (415)-769-1400
neal@druhi.ATT.COM (Neal D. McBurnett) (01/10/89)
in article <5604@sdcrdcf.sm.unisys.com>, darrelj@exec.sm.unisys.com (Darrel VanBuer) says: | TDDs differ from more common computer communications as follows: | Baudot instead of ASCII (handle in software with a couple of tables | and shift state flags) | | 45 baud (60 WPM in the Baudot 5 level code) instead of 110, 300, 1200, | 2400 ... (IBM PC serial I/O chip can be programmed for this, but not | via the BIOS) | | 5 data bits instead of 7 or 8, no parity (IBM PC serial chip can do it) | | 1.5 stop bits instead of 1 or 2 (any serial I/O chip which does 5 data | bits usually provides 1.5 stop capability, even those that don't will | interoperate by programming 2 stop bits since the receiver only needs | one, and sending 2 instead of 1.5 justs costs 7% of speed) | | Modem uses FSK half duplex signalling which uses frequencies | incompatible with any other modem (something like 1400 Hz mark, 1800 | Hz space, silent when idle). I faked it once by using a Motorola | Modem IC, a fast crystal (to shift freq up) and a couple of inverters | to fix the data sense [it was a little off spec, but just fine with | the wide tolerance detectors used at such a slow data rate]. | The deaf community is beginning to use dual standard machines (old | plus ASCII/Bell 103), but large numbers of existing equipment in the | hands of many who cannot afford new equipment make this a very slow | process. (And the need to change is low for them since even 60 WPM is | faster than most people can type). (I included the whole article because I'm cross-posting this to misc.handicap.) It seems to me that BBS's, on-line library collection databases, etc. are a great technology for the deaf. Are there any on-line dial-up services that interface via TDDs? -Neal McBurnett, att!druhi!neal
msa@toybox.UUCP (Mark Anacker) (01/13/89)
In article <5604@sdcrdcf.sm.unisys.com>, darrelj@exec.sm.unisys.com (Darrel VanBuer) writes:
: TDDs differ from more common computer communications as follows:
: Baudot instead of ASCII (handle in software with a couple of tables
: and shift state flags)
:
: 45 baud (60 WPM in the Baudot 5 level code) instead of 110, 300, 1200,
: 2400 ... (IBM PC serial I/O chip can be programmed for this, but not
: via the BIOS)
:
: 5 data bits instead of 7 or 8, no parity (IBM PC serial chip can do it)
:
: 1.5 stop bits instead of 1 or 2 (any serial I/O chip which does 5 data
: bits usually provides 1.5 stop capability, even those that don't will
: interoperate by programming 2 stop bits since the receiver only needs
: one, and sending 2 instead of 1.5 justs costs 7% of speed)
:
: Modem uses FSK half duplex signalling which uses frequencies
: incompatible with any other modem (something like 1400 Hz mark, 1800
: Hz space, silent when idle). I faked it once by using a Motorola
[ text deleted...]
: Darrel J. Van Buer, PhD; unisys; 2400 Colorado Ave; Santa Monica, CA 90406
Well, if you absolutely MUST make your PC talk to a TDD, it appears
that you could fairly easily modify a ham-type RTTY interface to do
it. For instance, I have a Heathkit RTTY box which takes the audio
FSK tones from my shortware receiver and puts out RS-232. From there,
it's all software. I have no trouble converting the BAUDOT to ASCII
on a PC or Model 102, at a variety of speeds. The beauty of this
particular box is that it's easy to modify the filters for whatever
mark/space tones you are using. All you would need to couple to the
phone line is a simple isolation and gain adjust circuit, or even an
acoustic coupler (remember those?).
It's a pity Heathkit doesn't sell this kit anymore, but they should
be available from ham's who've upgraded to more sophisticated
units. Might be something to look into.
--
Mark Anacker, US West Network Systems Inc., Bellevue, WA
{...uw-beaver!tikal,...!thebes}!toybox!msa -or- msa@toybox
"Their mission read, To Boldly Go,
But how ill-bred, To phrase it so." - Phil Garland