It's a bad practice to use unicode in math?

Question

Recently, I discovered the unicode-math package that permits to insert unicode characters inside math environment. This permits, even in its basic configuration, to use all the greek letters and most of the math operators directly in the source file reducing the amount of code like \alpha, \gg, \leq, etc. inside your equation making even complex equation readable directly in the code. This is obviously an huge advantage IMHO but is it without any risk and without drawbacks?

The risk that comes to my mind is to confuse one character for another very similar. While there isn't this possibility using macro, it could happen typing it directly.

The drawback that comes to my mind is that the symbol is fixed, in the sense that its appearance is glued to its semantic meaning. This is a things that usually one should tries to avoid, if possible. i.e. amsmath package gives us six different dots (\dots,\dotsc,\dotso,\dotsb,\dotsi,\dotsm) that have the same appearance but different semantic meaning and one can change the appearance of the symbol of one semantic meaning without affecting the others.

Probably I wrote something wrong or there are more things to say on the subject but my question still remains: is it a bad practice to use unicode-math? If not, what is the best way to use it? There are risks? If yes, what they are? There are drawbacks? If yes, what they are?

I know that that discussion should be posted on meta but this is a discussion about a package, so I think that it's better to post it here. If this isn't the place where asking such a things, please move the question or write in the comments where should I post it.

Edit: As pointed out in by Ulrike Fischer in her answer, I completely miss the whole picture about unicode-math. But my question still remain even if it should be reformulated into this one: What are the pros an cons of using unicode inside math instead of LaTeX macros?

As pointed out in the comments, this could be read as on opinion based question, it is not. I'm not asking which solution is the best in your opinion. I'm asking what are pros and cons of using unicode inside math instad of the classical macros that LaTeX offers.

Answer 1

Unicode input and unicode-math are not related like you claim it. You can use direct unicode input also with pdflatex:

\documentclass[]{article}
\usepackage[utf8]{inputenc} %not needed with texlive 2018/current miktex
\DeclareUnicodeCharacter{03B1}{\alpha}
\DeclareUnicodeCharacter{222B}{\int}
\begin{document}

  $
   α = ∫
  $
\end{document}

and you can use commands with unicode-math:

\documentclass[]{article}
\usepackage{unicode-math}

\begin{document}

  $
   \alpha = \int
  $
\end{document}

The point of unicode-math is that it uses one open type math font (here latin modern math) instead of a bunch of fonts like pdflatex.