I can’t leave well enough alone. Tonight I changed the keyboard slightly and added a bit of additional functionality.

The keyboard was updated to change the INV key to 2nd, replacement of the x><t (x exchange t(est register)) with l_nx, and the addition of the 3rd key where l_nx used to be. You can see the changes above.

The code change was to tie into the 2nd key’s click event and to change its color to indicate when it was toggled into an active state. It goes from light gray to orange and back again. But what’s significant about this is that keys that can change their state when the 2nd key is toggled also change their color and the text on the key itself. This, the l_nx key under normal mode conditions becomes the exponential key e^x and changes its color to a matching dark orange as you’ll note above.

This addresses a major gripe I always had, and continue to have, with standard key calculators. The keys are “dead” in that they are always a fixed color with fixed text. But with something like a phablet or tablet Android or iOS device, that’s not longer the case. This is a proof of concept for me. Time permitting I’ll add more secondary functions, and even tertiary functions, for the 3rd key.

The idea of an orange/yellow and blue secondary key is an idea as old as the HP-65, which first showed up in January 1974. I’ve just taken the idea and extended it a tiny bit to make it easier for the operator to find additional functions without having to use tiny text above and below a key, or to fill up the screen with lots of single function keys.

If I had the money I’d create a keyed calculator with OLED keys that would do precisely what this app is doing.

Here’s what the 3rd key looks like toggled. There are at present no other function keys it can modify.

The 2nd and 3rd function keys are set back to their primary colors by toggling the respective key, by hitting Clear or Clear Entry, or by performing the indicated operation, such as e^x.

I’ve decided to actually move forward to keystroke programmability, but not the way that either TI or HP did it. But first, I need to finish making all the keys fully functional. More to follow…

It would appear I’ve been channeling the TI-59 after all. Searching online brought up the keyboard for the TI-58/59 calculators. It didn’t take much for me to realize I’d replicated the core keypad in pretty much the same order as the original calculators. I quickly added two more rows of buttons/keys, right beneath the top row, with their proper symbology, and pretty much replicated the primary core keys.

The only keys not on Alternative Calculator are the far left row (the fifth row) that are dedicated to keystroke programming, a hallmark of those calculators. A final row is missing across the very top as well, the keys A-E that were meant to start programs loaded into the calculator.

A final key, the 2nd key, is also on that far left fifth column. It enabled a 2nd function for every key on the calculator.

If I want to follow the path of the TI-59, how far to I want to follow it? Do I want full emulation, quirks, bugs, and all? Do I want to replicate all the design decisions?

As much as I liked the TI-59 I recognized it had it’s problems and limitations, which were the product of the technology of the times (mid- to late-1970s). For example, a number of calculators show every keystroke entered, including all the operators. This gives a complete visual indication of where you are in formula entry, which is powerful especially if you’re interrupted and then come back to what you were doing. I can’t count the number of times I had to start over on my old TI-59 because I was called away from using it, only to have to clear everything and start over when I came back to it after an interruption.

There’s also the concept of keystroke programmability. The TI-59 (along with the equivalent Hewlett Packard calculators) allowed you to enter arbitrary sequences of keystrokes that combined the built-in functionality of the calculator in new ways for solving complex mathematical problems. With the TI-59 you could save them off on small chewing-gum sized magnetic cards. I was so happy to have this capability in the late 1970s and early 1980s, but by the mid-1980s I’d completely stopped using it and had moved on. True personal computers were all the rage and the spread sheet was the power math tool.

How far do I really want to go with this? I started down this path because I wanted a better way to perform math on my Nexus 7 tablet. How much time and effort do I want to sink into this? This project now has all the makings of a detour time sink with little to no real return. I have a lot to think about…