# C++ or Python?

Up until this point, the commands for ROOT/C++ and Python/ROOT were
nearly identical.[^f61] I presented them in the context of using cling,
ROOT's C++ environment.

From this point forward, using ROOT/C++ is different from using Python
with ROOT extensions ("pyroot"). You have to decide: in which language
do you want to use ROOT? My initial advice is to ask your supervisor.
Their response, in ascending order of likelihood, will be:

-   A clear decision (C++ or Python).

-   "I don't know. Which do you feel like learning?"

-   "I have no idea what you're talking about."

If it's up to you, this may help you decide:[^f62]

In favor of Python:

-   Learning Python is easier and faster than learning C++.

-   Python can be more appropriate for "quick-and-dirty" analysis
    efforts, if that's the kind of work you'll be doing this summer.

In favor of C++:

-   All of the ROOT documentation, the {ref}`advanced exercises`, the
    {ref}`expert exercises` of this tutorial, and most of the
    {ref}`tutorials included with ROOT <references>` are in
    C++.

-   If you're going to be working with your experiment's analysis
    framework, it will almost certainly involve working in C++.

-   C++ code, when compiled, is faster than Python (see this
    {ref}`discussion <stand-alone-c>`).[^f63]

[^f61]: See {ref}`here <differences>` for the
    differences when using Python versus ROOT/C++.

[^f62]: Here are the areas in which neither has a clear advantage: 

    - Both
      C++ and Python are used worldwide, so knowing either one is useful.

    - Python's interactive development is usually cited as an advantage
      over C++, but ROOT offers the interactive C++ interpreter, cling.

    - Both languages have substantive numerical computing libraries (e.g.,
      SciPy in Python, GSL in C++). 

    - Rivals to C++ and Python include (respectively) the [Julia
      programming language](https://julialang.org/) and the [Ruby
      scripting
      language](https://en.wikipedia.org/wiki/Ruby_(programming_language)). As
      far as I know none of the particle-physics groups connected with
      Nevis use them.

[^f63]: There are various tricks for making Python run faster; e.g., the
    \%pypy cell magic, the Cython extension, list comprehensions, clever
    use of numpy. You'll learn about them if you choose to become a
    Python expert.