Plus42 (and Free42) minor cosmetic issues

[jonmoore]

(05-20-2022 02:08 AM)Thomas Okken Wrote:  Actually, I felt myself getting a bit annoyed at this whole line of questioning... but I bit my tongue, and after a while, it dawned on me that maybe a bit of perspective would be a good thing, so here goes:

When I started work on Free42, the app upon which Plus42 is based, it was 2004, and the first platform I had in mind for running it on was the Palm m100, a low-end PDA that was popular at the time. The m100 had a 16 MHz 68000 CPU and no hardware floating-point unit, and it had a 160x160 pixel 2 bpp (4 levels) grayscale screen.

For comparison, that's about the same number of pixels as one icon on a recent iPhone's home screen, and the CPU is about 100,000 times slower than current smartphones.

What this means is: back in 2004, it was all I could do just to get Free42 to run at roughly the same speed as a real HP-42S, and the keyboard and display barely fit. Retina screens and fancy fonts weren't on anybody's radar.

Of course, hardware development hasn't stood still in the nearly 18 years since then, but I had my reasons for not supporting every little UI innovation.

The big one was OS churn. I initially developed Free42 for PalmOS and Linux (the latter using the Motif widget set). Once they were done, I ported it to Windows before releasing 1.0. And in the years that followed, I added more and more functionality to the UI, like program import and export, and the print-out window, meaning every subsequent port was going to be more and more work. And there were ports: to Windows Mobile, to PalmOS on ARM, to Linux using GTK widgets, to Android, to iOS, and to MacOS, and I also added even more functionality to the UI, like customizable skins, portable state files and quick state file switching, audible and haptic feedback, landscape mode, Unicode support, 64-bit support, copy & paste for all data types including programs and matrices... And lots of core functionality as well, like the big stack, local variables, long strings, et cetera.

And I did all this while also living a life and working a full-time job.

I'm not complaining, I wouldn't have done it if it wasn't fun, but maybe this helps get the point across how it might be possible that features that may make perfect sense have nonetheless, somehow, not gotten done yet.

If Plus42 sells well, maybe I can make it my full time job and then I can implement everything anyone has ever suggested, but if not, the pace of progress will remain slow and unpredictable. I'll need to sell a lot more units to make that happen, though, so, spread the word!

I feel your pain, and fully understand why things have developed the way they have.

But I do have some sympathies also with the discussion ref Plus42 especially. With Plus42, there's an opportunity to improve upon 42S UX/UI conventions that can make it very difficult for those not versed in the world of HP calculator evolution to pick up Plus42 and realise its benefits before frustration stops them from feeling their money has been wasted. I feel the same way when it comes to the DMC42 in much the same way (I have a few friends who purchased DMC42's having heard me rave about it, for them only to sell them a few short months later as they found the UX cumbersome.

If developers continue to only target their creations to an ever-shrinking target audience (preaching to the already converted), they can't complain when their apps aren't profitable.

Take a look at the NCalc Fx app on iOS (and it's Android sibling that goes by another name). This is what you're going up against. It's an emulation of what is probably Casio's best education-focused calculator the humble/diminutive fx-991ex. As well as mirroring all of the functionality of the aforementioned Casio hardware, it includes a wealth of CAS functions that share the same nomenclature as the same functions in Mathematica and graphing capabilities (Casio have a couple of current graphing calculators, the CG50 and Classpad but the fx-991ex covers the majority of the same functionality minus the graphing and CAS features of the Classpad). And you get all of this for approx $3. As such, it has a universal appeal to both high school and college students as well as professionals working in the field when they may not have access to their laptops (with Maple, Mathematica or Matlab etc.).

I actually think the features of Plus42 stack up pretty well against NCalc Fx, but the UX/UI can be a barrier for those more used to modern calculator offerings.

I don't mean any of this as a criticism as I do understand how you got to now and the challenges of keeping your old customers happy whilst making a product appeal to new customers from outside of the HP calculator ecosystem/community. Many of the things you've discussed with regard to new v1.1 features will definitely help widen the appeal and hopefully, that will be enough to convince more paying customers to jump on board. Ultimately, continued innovation will only make commercial sense if you widen your base (classic chicken/egg scenario).