Re: The hel-arabic.ly file story...

[Amir Czwink]

Thanks for your responses.

First and foremost: I'm not trying to accuse or hold someone
accountable. I'm trying to understand (and idealistically improve) the
process.

Okay, maybe my expectations were a bit too high considering the need of
comparing a contribution against the documentation in a musical context
one is not familiar with.

But come on, you don't need to be experts to discover code duplication.
Even if you're not familiar with Arabic music, the same blocks of code
in a single file from a single contribution can possibly be spotted and
it should be in your best interest to have as few copy-pasted code as
possible. I mean the first contribution to hel-arabic.ly even redefined
the "major" and "minor" key-signatures (and copies of that too) with
exactly these names to the exact definition they already have.
Ok, but as far as I understood it, people simply don't look at the
content of contributions about music styles they are not familiar with.
Well then, here I am to do it.

But I wonder, how the process will work then? Every one simply commits
what he thinks is best and we regularly overwrite each others
contributions? That can't be in the best interest of nobody...

Best Regards,
Amir

Am Freitag, dem 30.12.2022 um 11:54 +0100 schrieb Jean Abou Samra:
> Le 28/12/2022 à 23:09, Werner LEMBERG a écrit :
> > Exactly.  We are *not* experts for music from Arabia/Turkey/Iran. 
> > We
> > take what we get.  It was our mistake then to not asking you or
> > other
> > guys for for an expert opinion – this time I've directly contacted
> > Hassan for comments on your MR (hopefully his e-mail address is
> > still
> > valid).
> 
> 
> 
> Our having few experts of Arabic music around is definitely the main
> problem(and no, in spite of bearing the name أبو سمرا, I'm not
> knowledgeable about Arabic music theory).
> 
> Judging from the comments written in French in hel-arabic.ly, my
> impression
> is that the patch author thought his personal file would be useful
> for
> LilyPond but without really realizing the expectations from a
> contribution,
> James picked it up as we was helping with getting patches from
> unexperienced
> people through the contribution process although not a developer
> himself, and nobody really tried to compare it to arabic.ly because
> nobody felt qualified on Arabic music.
> 
> However, I would like to point out that even as people with zero
> experience in Arabic music , we do have a simple way to assess
> whether
> a contribution looks good. I mean the commit message.
> 
> Suppose I see an MR improving support for a type of music I know
> nothing
> about (say: Arabic music), and the commit message reads like this,
> where
> 'blabla' stands for stuff I don't understand:
> 
>    Add blabla style support to arabic.ly
> 
>    Previously, the only supported blabla style was blabla, which
>    is the most common style in blabla.  However, for blabla music
>    from blabla, a different style is used, where blabla. This allows
>    using this style by setting the property blabla. The data blabla
>    was filled according to the reference book blabla by blabla
>    (page 42), which is the closest to a standard that exists.
>    Note that this necessitates the change blabla to blabla
>    because blabla. In theory, this also affects the default style;
>    however, users of the default style should not notice it because
>    it only affects the situation where blabla, which does not make
>    musical sense because blabla.
> 
> Reading this, even with no understanding of the 'blabla' parts, I
> am confident that the change is well thought through. As a reviewer,
> I will check the idioms used, but I will trust the author on the big
> picture.
> 
> This should take under 5 minutes to write for the patch author,
> if indeed they understand what they are doing -- and if they don't
> understand what they're doing and nobody else does, the change
> should not be accepted!
> 
> Bottom line: in such cases, we can and should ask the patch
> author to provide details on what the patch does.
> 
> Best,
> Jean
>