Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Cartesian product of index family of sets

  1. Mar 8, 2014 #1
    Cartesian product of indexed family of sets

    The definition of a Cartesian product of an indexed family of sets [itex](X_i)_{i\in I}[/itex] is [itex]\Pi_{i\in I}X_i=\left\{f:I \rightarrow \bigcup_{i \in I} \right\}[/itex]

    So if I understand correctly, it's a function that maps every index [itex]i[/itex] to an element [itex]f(i)[/itex] such that [itex]f(i) \in X_i[/itex]…my question is, is there supposed to be a notion of 'order' implied in the definition here?

    The 'usual' Cartesian product is [itex]\Pi^n_{i=1}X_i=\left\{ (x_i)^n_{i=1}: x_i \in X_i \right\}[/itex] and here, there seems to be some notion of 'order' cause the first element is in [itex]X_1[/itex], second in [itex]X_2[/itex] etc.

    But for the first, if the index set is I={1,2,3}, then couldn't we have have [itex]X_1 \times X_2 \times X_3[/itex] or [itex]X_2\times X_1\times X_3[/itex] (any reordering of indices in the index set)? Then the definition of the Cartesian product could be [itex]((f(1), f(2), f(3))[/itex] or [itex](f(2), f(1), f(3))[/itex]...
    Last edited: Mar 8, 2014
  2. jcsd
  3. Mar 9, 2014 #2


    User Avatar
    Science Advisor
    Homework Helper

    hi chipotleaway! :smile:
    yes and no

    there's only an order if the index set, I, has an order :wink:
  4. Mar 9, 2014 #3
    hmm...so the case if I={1,2,3,4}, it would only reduce to the 'usual definition', n running from 1 to 4 to we somehow define the order to be 1,2,3,4
  5. Mar 9, 2014 #4


    User Avatar
    Science Advisor
    Homework Helper

    yes, eg spacetime coordinates are sometimes written (x0, x1, x2, x3), and sometimes (x1, x2, x3, x4)

    ie sometimes with t (or ct) at the beginning and sometimes with t at the end :wink:
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook