Occasionally I encounter messages like

Power::indet: Indeterminate expression (0. + 0. I)^0 encountered.

(or just 0^0). I believe I can safely declare these to be 1 for all of my purposes. Is there a way to build this in?

  • $\begingroup$ @MichaelE2 Oh I see thanks. Could not find that for some reason, I honestly tried searching for 0^0. Let us mark it as duplicate. Although I already learned from the answer below some additional possibilities, but still... $\endgroup$ Jul 18, 2017 at 5:14
  • 1
    $\begingroup$ @m_goldberg The way it was before your edit looked nicer with all that color-coding, are there specific reasons why your version is better? $\endgroup$ Jul 18, 2017 at 5:38
  • $\begingroup$ The formatting I applied to the error message is the formatting we have been using for such on this site for all the years I have been a member. The idea is that messages should be distinct from code. $\endgroup$
    – m_goldberg
    Jul 18, 2017 at 6:05
  • $\begingroup$ @m_goldberg I see. Pity. I wonder if somebody could be asked to extend color coding to messages too. Besides, it is just a random choice of convention which ones should go with four spaces and which with >!, no? $\endgroup$ Jul 18, 2017 at 6:18
  • $\begingroup$ Messages issued in Mathematica do not have color coding. Messages are are mixture of text and code. It is probably more trouble that it would be worth to try to get lexical scanner to distinguish what is text and what is code. $\endgroup$
    – m_goldberg
    Jul 18, 2017 at 7:11

1 Answer 1


Operator overloading is one way.

Power[0, 0] = 1;

You should add also the cases like Power[0.,0.]. When you want to clear your own definitions, Unprotect, Clear, Protect.

  • 1
    $\begingroup$ It should of course be stressed that you do this only if you know what you're doing (e.g. combinatorial as opposed to analytic operations). $\endgroup$ Jul 30, 2017 at 3:52
  • $\begingroup$ I 100 % agree with you. While the list of things you should do even if you don't know what you are doing is very short, this one might have deserved an explicit warning. $\endgroup$
    – Kiro
    Jul 31, 2017 at 5:35

Not the answer you're looking for? Browse other questions tagged or ask your own question.