I don't understand, musical notation is text though so how can it be superior to itself?

I think they mean staff notation, not a textual notation like "B♭".

Although, one could make the argument that staff notation is itself a form of text, albeit one with a different notation than a single stream of Unicode symbols. Certainly, without musical notation, a lot of music is lost (although, one can argue that musical notation is not able to adequately preserve some aspects of musical performance which is part of why when European composers tried to adopt jazz idioms into their compositions in the early twentieth century working from sheet music, they missed the whole concept of swing which is essential to jazz).

> one could make the argument that staff notation is itself a form of text, albeit one with a different notation than a single stream of Unicode symbols

Mostly this is straightforwardly correct. Notes on a staff are a textual representation of music.

There are some features of musical notation that aren't usually part of linguistic writing:

- Musical notation is always done in tabular form - things that happen at the same time are vertically aligned. This is not unknown in writing, though it requires an unusual context.

- Relatedly, sometimes musical notation does the equivalent of modifying the value of a global variable - a new key signature or a dynamic notation ("pianissimo") takes effect everywhere and remains in effect until something else displaces it. In writing, I guess quotation marks have similar behavior.

- Musical notation sometimes relates two things that may be arbitrarily far apart from each other. (Consider a slur.) This is difficult to do in a 1-D stream of symbols.

> although, one can argue that musical notation is not able to adequately preserve some aspects of musical performance

Nothing new there; that's equally true of writing in relation to speech.

How is that not text? Surely if we consider Arabic to be text (lots of ligatures, grouping, right-to-left notation) then music notes must be, too?

"I cannot read A, and I cannot read B. Therefore, A and B must be identical".

They didn't say that, maybe they can read both Arabic and musical notation.

The replied to comment seemed skeptical to treat musical notation as text. But any reasonable definition of "text" should include musical notation.

Otherwise it would be hard to include other types of obvious text, including completely mainstream ones such as Arabic. They are all strings of symbols intended for humans to read.

Feel free to disagree but I don't understand the argument here, if there is any. Lots of people read both Arabic and musical notation, it's a completely normal thing to do.

any reasonable definition of "text" should include musical notation

Then many a dictionary must be unreasonable [0]:

  text
    1. A discourse or composition on which a note or commentary is written;
       the original words of an author, in distinction from a paraphrase, annotation, or commentary.

    6. That part of a document (printed or electronic) comprising the words [..]
  
    7. Any communication composed of words

    n 1. the words of something written

Musical notes do not form words, and therefore are not text. (And no, definition 1 does not refer to musical notes). The written down form of music is called a score, not a text.

[0] e.g. http://dict.org/bin/Dict?Form=Dict2&Database=*&Query=text

Anything that can be turned into a string programmatically is by definition text.

I agree with you, I am disagreeing with the one that replied to you.

For complex music, sure, but if I'm looking up a folk tune on, say, thesession.org, I personally think a plain-text format like ABC notation is easier to sight-read (since for some instruments, namely the fiddle and mandolin, I mainly learn songs by ear and am rather slow and unpracticed at reading standard notation).

Yes. And I create and manage the musical notation for over 100 songs in text, specifically Lilypond.

If we accepted the validity of this argument, then literally everything that can be represented by a computer can be referred to as text.

It renders the term "text" effectively meaningless.

To be fair, in Lilypond's case, it is an ASCII interface that renders to sheet music (kind of like openSCAD).