HP Prime Emulator and inverse matrix

[Joe_Horn]

Whoa, chill out dude. I'm sorry if I offended you; it was not my intention.

My intention was not to "defend HP", but was rather to help you avoid the anguish which is always caused by unfilfilled expectations.  I understand now that you believe that HP should change Prime to fit your expectations, instead of you changing your expectations to match reality.  I wish you the best of luck in that endeavor, especially since (as I wrote above) I agree with you on your other points.

Again, my apologies.

-Joe-

[BartdB]

Hi Joe,

Sorry I misunderstood.

I understand that HP cannot make a calculator to fit my expectations. I know they have to make it as general as possible (one of the reasons I have liked HP's is the customize-ability, e.g. re-definable keys & programmability so I can add my own custom functions).

I also am a firm believer that just because something has "always been done this way", does not mean it should stay that way. Change is inevitable.

However, change should be justified.

At the moment, I do not believe that the change to "do something" when inverting a non-square matrix, instead of the usual "error", is justified. Especially for M^-1, where the inverting of the elements already has a function that does it.

Also, if two representations are described as doing the same thing, they should do the same, i.e. M^-1 and inv(M) (compare to e.g. x^2 and SQ(x) ). But M^-1 and 1/M are described as doing the same, and do so for square matrices, but do not do the same thing for non-square matrices.

Dennis Handly mentioned documentation. Currently matrix inverse for non-square matrices it is not documented.

HP is welcome to keep this behaviour, BUT they must document it clearly. Maybe the 1/M function is useful for non-square matrices, but I don't know that because the manual doesn't say it.

Perhaps HP want users to understand that giving invalid inputs will result in strange outputs - then they must document that.

Suddenly providing answers to something that all other calculators give errors on (because it has generally been accepted to be an invalid operation)  is likely to cause confusion with users. Thus HP must make some clear statements about it in the manuals.

What I fear is that when people see this kind of behaviour, they may wonder what other "funnies" the calculator has and not take the Prime seriously.

Sometimes change, even if it could be considered for the better, might be a step too far.

[BartdB]

Update: emulator version 5447 dated 2013-11-25.

There is now a warning message when inverting a non-square matrix. I think this is a good solution.

However, three instances of the warning message are displayed: