HP Prime - worthy successor to HP-71B ?

[brickviking]

(09-27-2017 08:18 PM)Vincent Weber Wrote:  ...but I came to dislike its FORTH side: everthing is in the stack, you must remember where, and do acrobatic DUP, ROT, ROLL, OVER...statements just to place things right.

The "problem", if indeed you could call it that, is that RPN (and to a very large degree RPL) both require time spent on thinking about problem shuffling. I.e. for an expression or function, you have to think more about where you want stuff while you're running the function. It tends to focus you on the order of solving rather than relying on someone else's model of precedence for mathematical and programming operators. I've seen variables used—sometimes liberally—to ease the impact of stack rotates.

(09-27-2017 08:18 PM)Vincent Weber Wrote:  As someone once said: RPL makes complicated things simple, but simple things complicated
And complicated things can be done even easier than RPL in advanced BASIC.

I don't have enough BASIC or RPL experience to be able to evaluate that effectively, but I do know RPL's definitely not as readable as BASIC. As time goes on I'm getting more amazed just how "baroque" RPL can be, even with its efficiencies borrowed from Forth and RPN. I compare it to a C that has syscalls embedded into the language as keywords, inline processor assembly language and a built-in debugger. All on a calculator. Ahhh....Reminds me of Smalltalk.

(Post 102, revision 19)