[gnuastro-devel] [task #14697] cosmiccal documentation

[Mohammad Akhlaghi]

Follow-up Comment #1, task #14697 (project gnuastro):

Thanks a lot Boud. Indeed Wikipedia exists and there are also many other books
that provide the mathematical proofs. But the emphasis here is not on
mathematical rigour, its on visualization/intuition for someone who is not too
invested in theory, but needs to properly understand the outputs of the
program: an observer. 

Atleast when I was studying these subjects, I couldn't find any proper place
to help in the visualization, everywhere I could find would jump so directly
deep into the mathematics that even though I followed the equations and
appreciated their abstract beauty, I could never actually visualize what I am
doing or what all these concepts were. So I came up with this explanation
(many years before creating Gnuastro).

The thing about Gnuastro is that it is not just a reference for abstract
teaching or just a black-box software that throws out results. Gnuastro is
actually the book+software: not just a book or not just a software. 

I have fully described the philosophy/manifesto behind creating Gnuastro in
the introduction (link below), I would really appreciate your thoughts on it
(maybe as a separate task):

https://www.gnu.org/software/gnuastro/manual/html_node/Science-and-its-tools.html

You see this in many other contexts throughout the book. For example here is a
very long section on frequency domain convolution: 

https://www.gnu.org/software/gnuastro/manual/html_node/Frequency-domain-and-Fourier-operations.html

It starts with visualizing the complex plane and goes onto the fourier series
and transform and sampling theory and etc. Observers (users of Gnuastro's
convolve) need to understand these concepts (I have seen many who don't, they
just know the equations, but have no intuition). 

Another nice example in another context might be the introduction to libraries
in the book:

https://www.gnu.org/software/gnuastro/manual/html_node/Review-of-library-fundamentals.html

Like the discussion above for curvature, all this information is available in
other abstract teaching books (on digital signal processing and programming
respectively), but I don't think that makes these sections redundant. Given
the discussion in the "Science and its tools" section, does this convince
you?

Ofcourse, we don't want the Gnuastro book to be a full encyclopedia, we can
put links to references (books/wikipedia) for people interested in further
reading to avoid going into too much detail in what ever context we are
discussing.

    _______________________________________________________

Reply to this item at:

  <http://savannah.gnu.org/task/?14697>

_______________________________________________
  Message sent via/by Savannah
  http://savannah.gnu.org/