shared rests between voice parts?

[Robinson P. Tryon 03]

I'm typesetting a book of college songs and came across some parts where two 
voices are on a single staff and share rests.

I did some digging in the archives and found that this issue had been raised 
before:

---- Start Quote ----
On Friday, Oct 25, 2002, at 10:35 US/Eastern, Klaus Zimmermann wrote:
I am just doing some SATB pieces and I wonder if there is a way
to share common rests between the two voices of one staff.


The default behaviour for something like the following is that there
are two rests generated, one typeset above the other.


I know of the possibility to use s instead of r in one of the voices,
but consider this more to be a workaround, then a solution because it
makes voice extracts a little uncomfortable.
---- End Quote ----


The answer that was given was to either put a spacer note in one of the two 
parts or put spacers in both parts and put the rests in a third voice.

Both seem like kludges; I'm already experiencing some issues with the midi 
output "omitting" parts in certain measures, so I'd really like to make my 
markup as "clean" as possible and not result to gross hackery.

Is there another way to accomplish this kind of markup?

Speaking of which, one concern I have with Lilypond is that to a certain extent 
it works as both a display and notation language; to tweak it to look "just 
right" one has to pervert the notes and the _music_ to change markup.  Is there 
a way to break up the intrinsic music properties (this is an eigth-note A flat, 
forte, staccato, with a slur to a quarter note B sharp, tenuto) with the markup 
(I want that to be a square-head, with a circle instead of a flat 
sign...etc...) ?  In my experience with lilypond files the two are integrated 
in a way that makes it very frustrating at times.