Re: Issue 1752 in lilypond: redesigning G clef in our Feta font

[lilypond]

Comment #14 on issue 1752 by percival.music.ca: redesigning G clef in our  
Feta font
http://code.google.com/p/lilypond/issues/detail?id=1752

wait, let me rephrase that: it wouldn't be a Critical regression because it  
would be a deliberate change.

However, I think it could still be a bad idea.  I've been reading  
background about python, and one of their guiding phrases is "the principle  
of least astonishment".  In other words, when they had a choice of making  
python do A or B, they took the choice which would surprise future  
programmers the least.
(at least, so they claim -- I think there's a few counter-examples, such as  
len() vs. length())

Still, I like the general idea.  "We should avoid giving our users bad  
surprises".  IMO, changing the way the clef is drawn, without giving users  
any ability to select the old clef instead, would be a "bad surprise".

Besides, I really think that supporting the old clef would not be a great  
deal of work -- and that work would be beneficial to all future font work.   
I mean, if you actually made a separate "Oscypek" font, which was optional,  
then I would be 100% comfortable pushing patches from you (to that font)  
completely blindly.  No user discussion, no review, etc.  You could make  
whatever changes to that font you wanted.
(caveats: I'd need to get agreement from other developers, and your patches  
would still need to compile.  And you might want to survey users anyway.   
But you would be working from a position of strength (i.e. you would have  
the final say about what went into Oscypek), instead of trying to reach  
consensus with everybody for every single change!)