lauren@vortex.UUCP (06/30/83)
Right. Sixteen button touch-tone pads also have appeared/currently
appear with an "A" on the lower right button.
Overall, touch-tone pads have gone through a variety of rather
interesting transformations over the years. Anybody who
remembers when touch-tone first started showing up (probably seeing
it for the first time at a World's Fair or Disneyland), will remember
the early ten button pads that did not include "*" or "#". Ah, but
how many of you remember the ELEVEN button pads that were floating
around for awhile?
Western Electric had all of its bets covered when some of the later ten
button pads went into full production: an entire series of the pads had
holes perforated into the framework of the unit so that the "*" and
"#" buttons could be retrofitted later through the use of an appropriate
field modification kit. Presumably a craftsperson would install this
kit if you really needed it (and paid the appropriate fee, of course).
When touch-tone was originally under study at Bell Labs, a variety
of human-factors studies were performed to determine how to
best arrange the buttons. One of the big debates was whether:
1 2 3 7 8 9
4 5 6 OR 4 5 6
7 8 9 1 2 3
0 0
was the appropriate pattern. The former we all know from current
phones, but the latter was (and still is) the standard pattern
for calculators (well, "adding machines" back then...) It was
decided that "the masses" could more easily deal with the former,
though some users of adding machines would probably get some wrong
numbers until they got used to the new layout.
A variety of other designs were also considered, including the
buttons arranged like a conventional telephone dial:
3 2
4 1
5
6
7 0
8 9
Another possibility tested was the pattern used for the MF keypads of
switchboard operators (2 vertical columns). A considerable number of
other "bizarre" patterns were also studied.
--Lauren--
P.S. This is totally off the subject, but my mention above about charging
for an "upgrade" to a touch-tone pad reminds me of a particularly
annoying habit of some of the telephone Operating Companies some years
ago. For many years, the Code-A-Phone model 700 Answering Machine
was the workhorse answering machine of the Bell System (in fact, they're
still around at a variety of Bell System installations). This was the
only reasonable unit that you could rent from Bell System telcos (this is
in the days when hooking up your OWN answering machine without an
expensive coupler could result in doom, destruction, and death if
telco found out about it..) Anyway, the 700 (which was a fine, heavy-duty
machine, by the way) was tariffed in an interesting fashion. You paid
different amounts a month depending on which "version" of the machine you
wanted. The "best" version allowed for up to three minute outgoing
messages and up to two hours of incoming messages. If you paid less a
month, you'd get "versions" that had progressively less incoming or outgoing
message time. Subscribers were told that this was only fair, since different
length tapes had to be installed, and cost varying amounts. In
reality, there was only ONE version of the model 700 Code-A-Phone.
If you didn't pay for the "maximum version", the installer would set a
pair of little cams in the unit which would artificially limit the
incoming and outgoing message times! Talk about "creative" product
design...
--LW--leichter@yale-com.UUCP (06/30/83)
At least there really WAS a difference between the "minimum" and "maximum" Code-A-Fones - even if it was artificially produced! There is an interesting story told about the CDC 6600 that used to be at NYU (I think it was retired about a year ago.) I am told this story is true, but perhaps not; maybe someone at NYU will know and send corrections/details. Anyway...NYU got a very early 6600 - serial number 3 or thereabouts. Now, there were two available memory options at the time: The full 262K (60-bit words0; or \\\\\words); or the "half-size" 131K. NYU decided that 131K would be fine, thank you. CDC kept trying to convince them that they really should get the full memory; the extra cost wasn't that great, etc., etc.; but NYU was firm, and the machine eventually arrived with 131K. Sure enough, some hackers decided to see - probably after some program did it by accident - what happened if you addressed the non-existent memory. Oddly enough, writing to it caused no errors; reading back from it caused no errors, and in fact gave you what you had written there; and writing to the non-memory didn't seem to clobber any "real" memory locations either... Apparently, CDC had never gotten around to trying to build a machine with only 131K. When they tried it, it didn't work. So, since they had to get NYU's machine out the door, they "removed" 131K in the documentation only! -- Jerry decvax!yale-comix!leichter leichter@yale [If it's not true, it ought to be...]
norskog@fortune.UUCP (07/02/83)
#R:vortex:-6500:fortune:6700014:000:253 fortune!norskog Jul 1 13:38:00 1983 Apocryphal story about an early Amdahl imitation IBM/360: It came in 2 models: the slow and fast versions. The speed difference was exactly 2. The engineering difference that accounted for the speed difference was the presence or absence of 2 wires!
leichter@yale-com.UUCP (Jerry Leichter) (07/03/83)
TRUE story about Amdahl 470's: They came in two versions, with a significant difference in speed. You could upgrade to the fast version, then flip a switch and have it go back to the slow one. So Amdahl came up with the rather interesting idea that you paid, essentially, for CPU cycles used when you leased such a machine. Separate timers kept track of time in "fast" and "slow" modes and you were charged different rates. The idea was that you would run in "slow" mode most of the time, and, when that "end of the quarter" crunch came, switch to fast mode to get the work done. I have no idea whether there was any difference in expected time to system problems in fast rather than slow mode - about the only thing that would make a difference in Amdahl's costs; I find it hard to believe there would be. As far as I know, this policy is still in effect, and the companies liking the \\\\\\[companies] using the machines like it. -- Jerry decvax!yale-comix!leichter leichter@yale
minow@decvax.UUCP (07/06/83)
Rumor had it that the only difference between the Univac 1106 and the Univac 1108 was that the 1106 did all memory accesses twice. The person who told me this said that a Univac salesman was showing a prospective customer around an existing site when the host d.p. manager asked "Should I run it as an 1106 or an 1108". Speaking of artificially limited products, anybody know how to add variable rings to a Panasonic KX-T1505 (low-end) answering machine? Martin Minow decvax!minow
bhyde@inmet.UUCP (07/08/83)
#R:vortex:-6500:inmet:6400015:000:923 inmet!bhyde Jul 1 22:16:00 1983 About 12 years ago, at NYU, I heard that story about what I presume was the same machine. About the same time I heard two other stories. The precursor of that machine was rumored to have had an air lock on the cpu room door since it was very sensitive to temp. changes. The last story was about the 1100 series from univac, Univac had estimated a tiny market for the machine but then ibm canceled their 36 bit line. This left univac in a bind. You see they had these machines they had sold to the goverment which at the time had a clause in the standard contract that required a refund of any future price decreases. One presumes the goverment noticed they were always getting stuck paying for the develoment of esoteric super computers. Well univac's solution was the 1106, with the clock turned down 40 percent. Funny thing was that if you disconnected this wire near the back panel the thing would run faster.
bloom@inmet.UUCP (07/12/83)
#R:vortex:-6500:inmet:6400020:000:1144
inmet!bloom Jul 11 21:14:00 1983
16 button Touch-Tone :
(R)
Some years ago, I read about the DoD Autovan (spelling) network ...
they used 16 button sets, the 4th column was used to signal the priority of
the call (AA, A, B, & C, I think).
NYU's CDC 6600:
The serial number was 4, and the machine was actually owned by the
Atomic Energy Commission (who else?). It was installed at NYU's Courant
Institute of Mathematical Sciences near Greenwich Village. It was the first
machine I'd ever seen with a display console (it had two CIRCULAR CRTs), and I
recall that the disk drive was facinating too - I think it was about 3 feet in
diameter, went about 25,000rpm, and held close to nothing (by today's
standards).
They were, at the time (c. 1967-8) working on a timesharing system
(what was that?) call SHARER (get it?), which I believe Boeing later
turned into SHARER II. Somewhere I have an old manual printout....
Ray Bloom
{harpo, ima}!inmet!bloom
P.S. The resident computer scientist at that time was a fellow by the name
of Henry Mullish, who has since written a couple of books. Anybody
ever hear of him?? Is he still alive??russell@cmcl2.UUCP (07/27/83)
#R:vortex:-6500:cmcl2:7500003:000:1321 cmcl2!russell Jul 26 19:15:00 1983 Re: The CDC 6600 at NYU For a complete description of SHARER see: The Communications of the ACM Volume 10 / Number 10 / October, 1967 Pages 659 thru 665 As for the the machine, it has been turned off since the 10th of October, 1982. It is still sitting in the machine room, waiting for the junk dealers to take it away. The old Bryant disk was removed many years ago. I have one of the platters in my house. I took the clock disk, 5/8 of an inch thick, 4 1/2 feet in diameter. The entire disk unit took up about 100 square feet of floor space and held 8 million 60 bit words (the machine used 6 bit characters). The average access time was 170 milliseconds! To help out, we also had .5 million drum (17 millisecond access), that was used for program loading and swapping. This software was traded to CDC for the drum and some communications equipment. It was later worked on by Boeing and Leigh University, and became INTERCOM, the standard CDC Timesharing Service available under the SCOPE Operationg System. This information is being brought to by one of the CEs who worked on this system from 1969 to 1972, when I quite CDC and got out the hardware business and took up systems work here at NYU. This old machine gave us very good service for all those years, minus 6 or 7 months when CDC refurbished it. -- Bill Russell UUCP: ...!floyd!cmcl2!russell (212) 460-7292 ARPA: Russell@NYU
sullivan@cmcl2.UUCP (07/27/83)
#R:vortex:-6500:cmcl2:7500004:000:63 cmcl2!sullivan Jul 26 20:25:00 1983 ...And Henry Mullish stills works here teaching this and that. -- David Sullivan UUCP: ...!floyd!cmcl2!sullivan (212) 460-7287 ARPA: SULLIVAN@NYU