notes.txt

* Tricky plots to draw

* Other pitfalls

* Crazy ideas



Different layers may map different fields in the original
data to the same aesthetic. So we want to collect all the
values of all the fields that are mapped to the given
aesthetic and then choose the minimum of all of them.

Raw data:

 - Determine layers and how they map fields in the raw data to
   aesthetics.

 - Now can determine the minimum and maximum values for each
   aesthetic.

 -

Group data:

 - A layer may render the data by grouping (e.g. to draw multiple
   lines for different data sets.)

 - A faceting layer may group the data and then render each group in a
   separate facet which may have its own layers which may further
   group the data.




Facet knows what Scales it needs based on what Scales the Geometries
that are going to render on the facet use. It can set the domain on
the scales based on the Data it is going to render if it knows the
mappings from the raw data to the scales, possibly after being
transformed my some Statistic. E.g. if we're plotting box plots, we
have raw data with 'value' and 'group' fields which we then run
through BoxPlotStatistic to produce a box plot stat for each group.
Then we map the various fields in the stat object to either the x or y
aesthetic and use the minimum and maximum values across all fields in
all groups to set the domain. The other aesthetic will be set by the
group field.


Geometries should be fed scaled data.