V.180114-2234	
                   Python V. 2.7.x  Installation  Notes 
       
      1. High-Level Programming Languages: Python and Better Ones 
      Be warned: Python is, in the opinion of this author, not the most user-friendly high-level(?) 
      language you can find, at least as far as programming for scientific computations is concerned. 
      The Python language was initially developed by people who had probably neither a clue nor an 
      interest in numerical work. That’s why the bare Python package does not even include the most 
      basic numerical data types and capabilities, such as arrays and vector syntax. Consequently, it is 
      unfortunately not enough for you to install just the basic Python package on your computer. Python 
      however  does  contain  (1)  senseless  constructs  for  binding  names  to  values,  as  well  as  (2) 
      oppressive rules of indentation and function definition, to name just a few of its flawed features, 
      all of which are at best useless and at worst confusing and cumbersome. 
       
      To do even the most elementary array-/vector-type operations or graphics, you need to install and 
      then import in your Python code all sorts of additional packages, such as “NumPy”, “MatPlotLib” 
      and “SciPy”. These add-on packages are sort of an afterthought to the original Python language, 
      running on top of the basic Python installation. Installing these packages on top of your basic 
      Python on your computer can be a harrowing experience, depending on which of the many possible 
      routes you take to do it. Doing scientific computations and graphics (plotting), even with these 
      add-on packages, can still be cumbersome, counter-intuitive and confusing, compared to other, 
      high-level languages, such as Matlab, Julia, Octave or R. 
       
      One advantage of Python is that it’s free. But you get what you pay for, see above! Another 
      advantage is that it is now thought of by many as the greatest invention since sliced bread and thus 
      widely used in many applications. Hence, it looks good on your resume if you can claim to be 
      proficient in Python. 
       
      By contrast,  the  advantage  of  Matlab  is  that  it  has  a  more  user-friendly,  intuitive  and  less 
      cumbersome syntax which is much better suited for scientific computations. You can do a lot – 
      and certainly everything you need for this course – with just a basic Matlab installation, without 
      any of its extra add-on packages. The reason is that Matlab was specifically designed by people 
      who know a lot about scientific computation, for people who want to do scientific computation. 
      Matlab is also well-recognized in the computational science community (physics, engineering 
      etc.), and hence not a bad entry on your resume either.  The big disadvantage of Matlab is that it 
      costs money; probably around $100 for a student license, including most of the add-on packages 
      – which you don’t need right now for this course, but could be useful for you later on.  
       
      The advantage of Julia is that it’s also more user-friendly than Python and well-suited for scientific 
      computations, just as Matlab. But, unlike Matlab, Julia is free! The Julia language syntax is 
      somewhat similar to the Matlab syntax and, just like Matlab, it provides a large, easy to use library 
      of functions for all kinds numerical applications, ranging from matrix / linear algebra operations 
      to differential equation solvers and lots, lots more. Installing any add-on packages you may need 
      is very easy and can be done with just a single standard command from within your basic Julia 
      installation. The disadvantage of Julia is that it’s been around for only a few years (since 2012?). 
      Hence, Julia is perhaps also not (yet) widely known or used in applications. So, it’s probably not 
      (yet) a big resume builder, but may well become one in a few more years.  Also, its native 
      documentation is rather terse and often hard to follow, especially for a beginner. However, there 
      are good tutorials around to make up for that. 
       
          V.180114-2234	
          For  alternative  opinions  on  the  relative  merits  and  demerits  of  various  high-level  languages 
          (Python, Matlab, Julia, Octave, R) just google them. I especially recommend this one: 
              https://lectures.quantecon.org/about_lectures.html#python-or-julia 
              https://lectures.quantecon.org/about_lectures.html#how-about-other-languages 
           
          Lastly, you might be curious about comparisons of the computation speeds of these different 
          languages. That’s not really important for the very small calculations you do in this course, but 
          probably more important if you later want to do large-scale numerical work. Here are some results 
          from such a comparison: https://modelingguru.nasa.gov/docs/DOC-2625.  
            
          2. Python Versions 
          There are probably many different ways(?) to install Python on your computer – and not each of 
          them is available for each major operating system (OS): Windows, MacOS and Linux.    
           
          However, regardless of which OS you have, it is strongly recommended that you install one of 
          the older Python versions, 2.x, not the newer versions, 3.x. As you will see and noted above, 
          compared  to  other  high-level  languages  (Matlab,  Julia),  Python  is  not  a  very  user-friendly 
          language, and in the newer versions, 3.x, they seem to have made it just a bit more unfriendly. One 
          of the latest Python 2.x versions is 2.7.14 (at time of this writing), and that would be a good one 
          to install. However, some earlier versions, 2.7.9 or later, will also work just fine for your purposes 
          in this course. Just do not install any version that starts with 3. … ! 
           
          If you have a Mac, Python may already be pre-installed with the Mac OS X, but it may not be the 
          version you want or may not contain all the add-on packages you want, see above. In the latter 
          case, you may still have to do some installation yourself, see below. The same may be true if you 
          have the Linux OS on your computer. 
           
          3. Checking for Already Installed Python Versions 
          Before you install anything, it could be helpful to check whether Python may be installed on your 
          computer already and, if so, which version.  To find out how to check which Python version, if 
          any, is installed on your computer, read this: 
              https://edu.google.com/openonline/course-builder/docs/1.10/set-up-course-builder/check-for-python.html 
          This is important because you need to first uninstall any existing older version of Python on your 
          computer before you can install a new one. Otherwise, the new version installation may interfere 
          with the old one, and vice versa, and the result of that will likely be that nothing works anymore, 
          at all.   
           
          4. Installing Python, SciPy, NumPy, MatPlotLib, etc: 
          4.1. Perhaps the best place to get an overview and learn some basics about Python, about its main 
          numerical and graphics add-on packages, SciPy, NumPy, MatPlotLib and how to install them all, 
          is the SciPy website:  
              https://SciPy.org/.  
          Read and follow:  
              https://SciPy.org/getting-started.html,  
              https://SciPy.org/install.html. 
          The latter “Install” page gives you many different installation options, some of which may, or may 
          not (?), work for your computer and the OS running on your computer. The Anaconda installer 
          seems to be the best one of the bunch right now. It is described in more detail below. 
           
              V.180114-2234	
              Once you’ve installed the basic Python, you can also install the add-on packages. For using basic 
              array data types and array/vector syntax, you need the NumPy package, described here 
                    http://www.NumPy.org/  
              For scientific computations beyond just basic linear algebra, you then the SciPy package, as 
              described above and  installed on top of NumPy. For doing graphics (plotting) from inside your 
              Python code, you need MatPlotLib, described here: 
                    https://MatPlotLib.org/ 
               
              4.2. The currently recommended way to install Python (see https://SciPy.org/install.html) and, 
              hopefully, all the ad-on packages you need is the Anaconda installer. Go to and then follow the 
              instructions at: 
                    https://www.anaconda.com/download/#windows    for Windows 
                    https://www.anaconda.com/download/#macos      for Mac 
                    https://www.anaconda.com/download/#linux      for Linux. 
              On each of these download pages, you have the choice to download the latest Python 2.7.x version 
              (likely 2.7.14 at time of this writing) or the latest Python 3.x version. As noted above, you’ll 
              probably be better off if you install the latest version of Python 2.7, not Python 3.  
               
              Once you’ve completed the foregoing Anaconda download, follow its instructions to complete the 
              Python installation on your computer. Chances are that Anaconda may also provide you with an 
              easy  way  to  install  the  add-on  packages  you  need  –  or  perhaps  even  install  them  for  you 
              automatically. But, just in case that fails, and you need to learn more about installing add-on 
              packages, go here: 
                    https://packaging.python.org/ 
                    https://packaging.python.org/tutorials/ 
                    https://packaging.python.org/tutorials/installing-packages/ 
               
              In case Anaconda doesn’t work for you at all, try some of the other options for installing Python 
              and/or its add-on packages, shown on https://SciPy.org/install.html, see above. 
               
              Finally, if all the above fails, try this: 
               
              4.3. Google for other options yourself. Good queries to enter on google, depending on your OS, 
              are 
                    how to install Python on mac 
                    how to install Python on windows 
                    how to install Python on linux 
              and, likewise for installing NumPy, SciPy, and MatPlotLib. 
               
              5. Getting Help: 
              The UGA Helpdesk is explained at 
                    https://confluence.eits.uga.edu/display/HDSH/Help+Desk+Support+%28HDS%29+Home 
              Contact them via  
                    Email:       helpdesk@uga.edu 
                    Or Phone:    706-542-3106  
               
              When you do contact them it helps them a lot to help you with your problem if you send them a 
              detailed description (incl. screenshots etc.) of  
                    (1) the task you are trying to accomplish,  
                    (2) the steps you were executing before and when you encountered the problem; and  
      V.180114-2234	
         (3) all error messages you received in the course of your actions.  
      The more carefully you describe your problem, the easier it is for them to help you find a solution.