Re: pushproperty syntax

[Mats Bengtsson]

Every now and then I start complaining about the Mudela syntax,
I'm afraid it's time again. 

I think the current trend of the syntax development is to 
make Mudela more and more cryptic to people who don't have
a degree in computer science (I have one so I shouldn't complain).
The latest property syntax is one excellent example. 
>From the warning messages of the latest Lilypond version, 
it seems that you plan to remove the old property syntax 
completely. 

However, the change in 1.3.94 was a step in the right direction
since it reduced the number of keystrokes. Still, why use the 
concept "push" and "pop" in the syntax. Why not replace
\property Voice.Dots \push #'direction =  #-1
with, for example, 
\addproperty Dots= #-1 \to Voice
or
\addproperty Voice.Dots = #-1
The corresponding for \pop could be \removeproperty

The fact that this actually is a stack is easily described in
the documentation, something like "If several values have been 
added to the same property, the last one will be removed first 
and the last remaining value will be used".

Also, now the property setting involves three different 
entities, the engraver, the context and the actual property.
With the old mechanism, the information about engraver and
property was combined into a single string. I'm not if this
is better or worse, you probably have to look at the docs
or some example before you can use a property anyway. 

Has the type checking of the property values disappeared
not that generic-property is broken?

     /Mats