Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Show HN: Rotary Phone Dial Linux Kernel Driver

gitlab.com

Post reply on HN
Print view View original

31–40 of 51 posts

Re: Show HN: Rotary Phone Dial Linux Kernel Driver

#31

Post by jmb99 »

gaudystead wrote: post #27
spicybright wrote: post #14

Earlier quoted context omitted.

That would have been a hilarious timeline if they just upgraded ipods with cell networking, at least for one model. Then you can use this for typing: https://www.youtube.com/watch?v=9BnLbv6QYcA

It's wild how that video came out 16 years ago and basically ends with "we'll see if Apple will make computers for business users and not just dicking around", and now in the current day, I'm working at a place where an update was sent out to our Macbook users that bricked the ones running an M1 chip, but not the M2 or M3 models. It almost feels _impressive_ how hard macOS still is to integrate into an enterprise set…

Maybe it depends on the business, but I’ve been using Mac’s in an enterprise environment for my entire career with 0 issues. And yes, fully managed devices, not just “here’s a Mac have fun.” If your company can’t test updates before rolling them out, it doesn’t matter whether your shop is Mac, windows, or something else. I mean hell, if you’re all on Apple Silicon MacBooks, there’s probably only what, 6-8 different SKUs to test? Not like windows shops where you might have a few generations of Dell/Lenovo/HP workstations and/or laptops, each of which will have its own distinct problems.

Re: Show HN: Rotary Phone Dial Linux Kernel Driver

#34

Post by niccl »

bacon_waffle wrote: post #33

Fun historical twist: New Zealand rotary phones used a reversed mapping between digit and pulse count - here the number of pulses was 10-digit.

There was a deep technical reason for it, too. I suspect I'm one of the very few living people that know what it was., so I'll share it with HN so the knowledge doesn't die :-)

A relatively early type of mechanical telephone exchange was the rotary exchange [0]. The pulses from the phone cause a clutch to connect the rotary driver in the exchange which then moved the switching stuff around (details can probably be inferred from the linked article). One of the issues with the rotary exchange is the pads of the clutch wear, leading to unreliable connections. Aotearoa/NZ had an existing number plan when they decided to install rotary exchanges. Some bright spark knew of the wear issue, and calculated that, given the existing number plan, if they had the 1 position on the dial giving 9 pulses (etc.) then the overall wear on the pads would be much lower and so the maintenance requirement would be less. And that's where it started.

And another fun fact. I believe Norway chose the same configuration for their rotary phones. I'm not sure if it was for the same reason, though.

[0] https://en.wikipedia.org/wiki/Rotary_system

Re: Show HN: Rotary Phone Dial Linux Kernel Driver

#37

Post by CodeBeater »

jermaustin1 wrote: post #7
CodeBeater wrote: post #6

This article pops up as I have a rotary phone disassembled right on my desk (rewinding the clock spring). Neat coincidence!

Real question: how long has the phone been disassembled on your desk? If you are like me: around 2 years ;).

Actually, just under two days. That phone is still in service, pulse dialing away on the rare occasions I gotta make a call through a landline.

Re: Show HN: Rotary Phone Dial Linux Kernel Driver

#38

Post by ztetranz »

niccl wrote: post #34
bacon_waffle wrote: post #33

Fun historical twist: New Zealand rotary phones used a reversed mapping between digit and pulse count - here the number of pulses was 10-digit.

There was a deep technical reason for it, too. I suspect I'm one of the very few living people that know what it was., so I'll share it with HN so the knowledge doesn't die :-) A relatively early type of mechanical telephone exchange was the rotary exchange [0]. The pulses from the phone cause a clutch to connect the rotary driver in the exchange which then moved the switching stuff around (details can probably be in…

Thanks for this. I'm a kiwi and wondered why we were different.

When friends started bringing pushbutton / cordless phones across from Australia, I was able to convert them to NZ numbering by reversing a few wires on the keyboard matrix. These generated pulses long before DTMF.

Re: Show HN: Rotary Phone Dial Linux Kernel Driver

#39

Post by anonymousiam »

jtwaleson wrote: post #30
anonymousiam wrote: post #10

Back in the late 70's, I made a rotary phone dialer for my HP41C calculator. I connected a NC reed relay to the piezoelectric beeper and put the NC contacts in-line with my telephone line. I used "synthetic programming" (undocumented opcodes) to get the short duration beeps needed for the dialing pulses. I could enter a name (alphanumeric!) and it would look up and dial the phone number. About 10 years ago, I met a g…

Cool story :) out of curiosity, do you program those kind of things on the calculator itself? I did that with the hp49g and was proud of the programs I was able to write in such a constrained environment, but the single line display of the 41c would be a real achievement!

Thanks for asking. Yes, the calculator itself was the first "computer" that I owned, and there was nothing else to connect it to, or to emulate it with. (I had my own PDP-11 at work, and I had previously done batch-mode programming on an IBM 370.) I bought my 41C in 1979 (made in the US!), and spent several thousand dollars on peripherals over a period of about three years. It was a big investment in a platform that did not interoperate with anything else, and I ended up giving it all away. I didn't keep any of it, including all the programs I had written and stored on magnetic cards (a fine example of extinct storage media).

I may still have a spiral notebook, in which I did all of the code development. It would be a hard-copy of all the code I wrote. If I do still have it, I have no idea where it is. It sort of reminds me of the end of Citizen Kane.

Later on, I also got rid of my Synertek SYM-1 and all my CP/M platforms, except for one IMSAI 8080 S-100 box, which I eventually gave away to someone else. I didn't realize at the time (nearly 30 years ago) that the IMSAI 8080 was quite collectable, and was going for about $800 on eBay. Still, I didn't regret it much.

The old systems evoke a sympathetic nostalgia, but actually using them is an exercise in patience, and eventually evokes a desire for more modern systems.

I've still got some emulated retro hardware and some old software emulators, but I don't really use them much.

Re: Show HN: Rotary Phone Dial Linux Kernel Driver

#40

Post by johnea »

anonymousiam wrote: post #10

Back in the late 70's, I made a rotary phone dialer for my HP41C calculator. I connected a NC reed relay to the piezoelectric beeper and put the NC contacts in-line with my telephone line. I used "synthetic programming" (undocumented opcodes) to get the short duration beeps needed for the dialing pulses. I could enter a name (alphanumeric!) and it would look up and dial the phone number. About 10 years ago, I met a g…

A friend and I also made this!

We used it as a speed dialer that worked on any phone.

I still have the hard copy of the synthetic programming guide!

Even outside of that technique, the calculator was a very powerful programmable tool for the early 1980s.

The same friend and I also wrote a program for performing all of the calculations for a grad class in S-parameter modeling of RF transmission. We were so proud, we showed off the program to our professor. Our "reward" was having calculator "programs" banned during all tests 8-/

We didn't realize until later, that there was a "turf war" going on between the RF section of our EE department, and the digital electronic section.

This is how the HP-41CX calculator helped me learn one of the most significant lessons of my EE degree: when it comes to human decision making, the tech is often far from the top priority.

Post reply on HN