OT: A Bandsaw and a Keyboard

[Silvan]

Life has been happening here too. I forgot all about it. No hurry. I still want it though.

[Mark Jerde]

In 1975 I was one of two boys in the high school typing class. I learned to touch type the number row pretty well. Unless I'm typing something like this const pi = 3.141592653589793238462643383279502884 I just use the keys conveniently placed above my fingers in the normal typing position. ;-)

-- Mark

[Mark Jerde]

Ok.

-- Mark

[Mark Jerde]

Ok. ;-) I can usually have the number mostly touch typed in before I'd get my hand repositioned on the keypad. But if it was quicker to use the keypad I would.

-- Mark

[Mark Jerde]

Seriously? They repeated the same bad decision that left us with the QWERTY keyboard?

-- Mark

[Mark Jerde]

P.S. I type 70+ WPM and 25-30 WPM on numbers. I shared an office this summer with one of those 20-something web-designer wiz kids. He averaged at least 120 WPM for normal typing and at least 80 WPM for special keys and numbers. He typed raw HTML faster than I can type "Now is the time for all good men..." And I'm better than a lot of secretaries I worked with. ;-) Humbling. !*@@#$%# kids!

[Leon]

Yeah I did that in HS and in college. Having been in the automotive business for most of my professional life I entered part numbers all day long for years.

[Leon]

Like you, I want to move my hand as little as possible from the key pad to the track ball.

[Leon]

My 17 year old son borders on typing that fast.

[Bruce Barnett]

I didn't use "had" describing phones.

I used "had" describing area codes:

[Bruce Barnett]

They did it to keep keys from jamming?

I thought they did it so Carrot Top could dial down the middle row.

[Tim Douglass]

When I was in school I had a friend whose mother was a civil service secretary. She could type over 120 WPM corrected - on a manual typewriter!

I used to do 90+ when I was programming, but that was a long time ago and I have arthritis problems in my fingers now, so it is a *lot* slower.

Besides, I don't think fast enough to type at that speed.

Tim Douglass

[Bernie Hunt]

In 1976 I started high school and the rumor in the freshman football team was that the varsity coach taught the typing class and it was an easy A. So we all signed up. What we didn't realize was that varsity football coaches are way to busy to teach a typing class. They have teachers assistants who don't like football players looking for easy As, to teach typing. We actually had to learn to type! I thought it was an complete waste of time. Then 4 years later I get into electronics and suddenly I'm on of the few guys who can type on a computer. Turned out to be a good thing, hahahhaha.

[Bernie Hunt]

Well actually its an accoustic thing. To be able to accurately identify the DTMF tones the circuit takes a curtain number of cycles to lock onto the tone. DTMF tones are mid range tones, the lower component is under 1,000 hz. This lenthens the amount of time it takes to detect. Basically the fastest you can dial is a 10 digit call in 1 second. The detectors can't work reliably much faster than that with all the other considerations, cost being a big one, hahahaha.

Bernie

[Robert Bonomi]

You might be surprised.

Early Hayes Smartmodem "clones" let you set the tone timing to _very_fast_. i.e., some units would allow settings as short as 15ms. *Default* setting was 50ms. The only place I ever found full 50ms timing _required_ was where the switch, itself, was _pulse_, and there was a DTMF-to-pulse front-end. Where the switch was "natively" DTMF -- at least in _my_ experience -- you could always shorten the timing to 35ms. And, often, even faster. Typically, I found the 'failure to capture' reliably point was around 20-22ms. I didn't want to run 'right on the edge', so I routinely ran with 25ms timing.

Bogon!

The telephone keypad arrangement was dictate by the fact that there are

*letters* on most of the keys, and "Standard English" reads left-to-right, TOP-TO-BOTTOM.

The '10 key' compatible arrangement, to wit:

PRS TUV XYZ [C]

GHI JKL MNO [B]

--- ABC DEF [A]

• 0 # [D]

Would *not* have made a whole lot of sense. Note: the _full_ touch-tone pad *IS* a 4x4(!!) grid. With the official labelling for the 'extra' column being "A B C D", in sequence of the 'row' tones.

The key-label to DTMF tone-pair mapping *could* have been done differently, such that one would not have the above 'silliness' with the 'column 4' labels. but you'd still have to deal with the lines of letters reading from bottom to top.

Note: Bell Labs *did* test both arrangements, and found the 'non-10-key' layout "worked better" for the vast majority of people. Especially so in NON-DESKTOP situations -- e.g. wall phones, pay phones, trim-line phones, etc.

[Bernie Hunt]

Robert,

You have to remember that our telephone system is designed for worse case, which we fortunately rarely see. Back in the 80s I ran across some wiring that was run exposed as the top strands of barb wire! I was always amazed at how our telephone system could keep running while maintaining all the embedded base. In the cities you can push the tolerances quite a bit, but getting out away from the new switchs and better wiring the toleranced matter!

Funniest thing I ever ran into was when I put some ACDs into offices colated with COs. The audio and loop current was so hot, we had to use altenuators and simluated line loss to keep from burning out line cards and causing accoustic shock for the users. We were sitting right on top of the CO and all their specs were assuming alot of loss in loop current and audio before it got to the customer premise. It actually took a whle to figure out the problem because you rarely ever hear of too much audio or too much loop current. Field guys are always looking for more, not less, hahaha.

Bernie

[Patriarch]

"Bernie Hunt" wrote in news:VJLPd.2505$ [e-mail]:

Today, we worry about the lasers being too 'hot'...

Patriarch

[Robert Bonomi]

Heh! I was _on_ some of that precise kind of wiring, for a while, back in the 60s -- a "4-party" fence, in point of actual fact. :) (call quality went to h*ll, when a cow decided to lean on the fence, too.) In the earlier days, there was a _lot_ of that kind of wiring. I'm somewhat surprised that it (fence wire) was still in use in the 80s, though. I'm curious, where was that?

As far as 'outside plant' goes, rural areas were _more_ likely to get upgraded than the big-city 'central city'. Downtown would get the newer C.O. switches, but the wiring to the customer was a "whole 'nuther story."

One of the *really* nice things about DTMF, however -- the wiring _mostly_ dropped out of the equation. DTMF transits long distances, and 'strange' wiring *better* than dial pulses.

The _big_ performance difference was tone detection via computerized DSP vs. the early PLL detectors. PLL detectors needed the 50ms, the DSP chips and algorithms were specced at Funniest thing I ever ran into was when I put some ACDs into offices colated

Yuppers. been there, done that. Local school district put in a private phone system -- leased 'transit' from the _gas_ company; optical fiber *in* the gas lines -- got a bargain rate, too; this was 'found money' to the gas company. This involved disconnecting *all* the telco lines, running to _all_ the schools, and substituting a relative handful of 'trunk' lines tied to the district's switch. The ILEC got shitty about the matter, and was demanding

*beaucoup* dollars 'per mile' for installing the trunk lines between the C.O. and the school district H.Q. The school system had *GOOD* telecom people. They went and rented a 'closet' *across* *the*street* from the telco C.O., and said 'terminate the lines *there*', Then they back-hauled on their gas- pipe fiber to the switch. Telco was *REALLY* pissed -- they got an entire *seventy-five*feet* of their 'many thousand dollar per mile' charge. At roughly '100 wire feet' from the switch, one was, for all practical purposes (*except* what one paid the telco, that is) co-located.

Telco had _really_ wanted the school district to co-lo their *switch* _at_ the telco. Then they could charge (recurring!!) for the space, utilities, etc. Over and above the line charges. As it was, they got the recurring line charges, and that 75' of one-time install. Renting that 'closet' was somewhere under 10% of what the telco wanted for colo space.

Insult to injury, *most* (like around 80%) of the call traffic for the schools was to/from *other* school facilities. When everything was interconnected through the private switch, the traffic level to the 'public' network plummeted. The half-a-dozen (or more) lines to a couple of hundred locations got replaced with an amazingly _small_ number of trunks.

The whole system paid for itself _incredibly_ quickly, out of the money that _used_ to go to the telco.