3 ms·
Science left the limitations of the pocket calculator behind, ages ago. Any field of analytic research, be it science, engineering, finance, etc. the values you
by mahmud 15y ago
Science left the limitations of the pocket calculator behind, ages ago. Any field of analytic research, be it science, engineering, finance, etc. the values you manipulate on a day to day basis are no longer scalars, but tend to be matrices, often very large ones.
TI got this part right, sort of: the TI calculators while clumsy for data entry have 2D screens and make apparent the various interpretations of mathematical objects, as equations, matrices, and graphs.
- runningdogx 15y agoReally? I think the n-spires are going in the wrong direction for post-academic users, trying to approach the input (full alphabetic keyboard) and UI of a computer on a severely size-constrained piece of hardware, rather than focusing on what I think stand-alone calculators should be about: an easily manipulated tool for back of the envelope calculations and first-try graphing of equations. By allowing easily accessible alphabetic entry, it makes typing in complex equations easier, but anyone who is not constrained by academic requirements will be using matlab/R/mathematica/etc. for that, rather than trying to type using a small non-qwerty keypad, and getting results on a too-small screen. I understand the logic of something like the n-spire for academics (though I'd prefer an HP 50g): get a piece of hardware that can do enough math for testing purposes, but limited in usability and/or IO, to hopefully reduce cheating, and have students use it in class and for homework so they are familiar with it by the time they have to use it on a test. However, it's sad that an entire industry (calculator hardware) requires poor and time-wasting data and formula entry characteristics (mostly based on required form factor) simply to avoid having students cheat on tests (both in-classroom and standardized tests). How much time and effort would be saved if students could use smartphones or tablets with a customizable calculator UI? Calculators' limited screen size, inflexible input, and limited market size compared to smartphones/tablets ensures they will remain hamstrung. And TI still hasn't gotten it through their heads that RPN should be a basic supported calculator mode, not left to a 3rd party to implement as an optional program. Disclosure: I grew up with a TI 30+, 34, 68, 81, 83+, and then a HP 48gx -- which got me through most of HS and college, hence my strong preference for RPN. My HP48 broke (screen damage, no fault of HP's) and I never replaced it. I've used a calculator probably a single-digit number of times since then.
- sliverstorm 15y agoPocket or table calculators can still be useful for quick calculations. Of course you won't use a 15C for any seriously heavy lifting anymore- you have MATLAB and Mathematica for that now- but M & M are poor tools for spur-of-the-moment computation.
- w1ntermute 15y ago> but M & M are poor tools for spur-of-the-moment computation. I don't really see how this is the case. The only advantage of a stand-alone calculator, now that I'm done with standardized test taking, is its portability. And when most of my work takes place in front of a computer anyway, I might as well just keep a copy of Mathematica humming so I can use it as soon as I need to do a calculation. In fact, having to pull out a standalone calculator, turn it on, and adjust my frame of mind from "computer" to "standalone calculator" would probably be very bad for spur-of-the-moment calculations.
- sliverstorm 15y agoWell, I was thinking of more when I'm talking over my whiteboard with a co-worker, or something like that. Also, M & M are not light on the footprint... I don't especially enjoy leaving them open.
- w1ntermute 15y agoThat's definitely true when you're in a conference room working over something, though nowadays I'm just using a scientific calculator app on my Android phone instead, since that means I've got one less device to keep track of.