Einar Stefferud wrote:
>}>I have some serious problems with the propose polling questions below.
>}>
>}>The problem si that they deal with lower level details while the
>}>higher level meta issues regarding common values have not yet been
>}>touched. So, the cart is in front of the horse here.
>}>
>}>Also, the poll does not alow for enough choices, such as "NONE OF THE
>}>ABOVE", and do not allow for comments to be inserted to
>}>qulify answers.
>}>
>}>In short, it is much to limited in band width. We shodul be able to
>}>say something about why YES or why NO, or why SOMETHING ELSE?
>}
Stef,
Its not just your serious problem - its the essence of any abstraction/
concretization process. What we've got here is a *design problem (there
is no cart yet ;-), and the tried and true method is to take all the bits
and scraps and toggles and sheer details you can get ahold of, shuffle
them around until patterns (which we might call subassembies) start to
emerge, and then do the same with them until 'assembies' become apparent,
and finally you can see how some 'whole thing' might come out of it. But
even then, it may not be the whole thing you really want, so you start
back through with feedback -- that is, the knowledge of how well the
(sub)assemblages work together -- and trim and snip and cobble and kludge
so the patterns are clearer, the joints tigher, the action smoother, the
whole thing neater and tidier.
Then, yes, *after* a situation has been fully analyzed, one can write out
the instructions, make neat hierarchical decision trees, so the *builder
(or the voter) can deal with 'higher-level' issues first, and then move
down into greater and greater detail of implementation.
In short, this DNS mess is entirely backwards from the way lots of
people are accustomed to working together. It wont get turned around in
one fell swoop, but how are we going to turn it around at all if we cant
sort out these two fundamental perspectives?
kerry
