TODO

Here is a start of a list of ideas:
Please feel free to add to it or to start implementing stuff:

 1) The file, data/mgu74v2id
   contains an environment that has the affy number linked to (the 
  genbank accession number)

   It would be nice to be able to select a set of genes and
   get a webpage set up with the Affy #, and links to various
   databases (genbank in particular).

 2) We need to decide which databases we will store in data:
    -functional groups
    -basic mappings
    -proteins
    -chromosome mapping

 3) A set of tools that let us select data from a data frame (such as the 
  output of dChip) according to the data in these files.

 4) Do we want to explore a Tcl/Tk interface to selection?

 5) Need to store the data files compressed (.gz) and then use the appropriate
    readers. What would be even better would be to have stored binary forms
    of environments (or memory-mapped environments).

 6) Now have the ability to load up locus link webpages, but we
    would like to be able to do keyword searchs and to interpret the
    results directly in R.

 7) We need some form of class structure that will allow us to
   manipulate the objects in a more uniform fashion.

 8) We need to design the interface so it can be handled either by
   environments/hash tables or by a database -- either Access or
   Postgres

---
some comments after 26 oct meetings

options-like structure (or object-like, e.g., lm.object) for htmlTable
and htmlPage so that htmlTable( listOfContent, listOfMarkup, tableOptions )
and htmlPage( listOfContent, listOfMarkup, pageOptions ) have good defaults.
constraints on content elements of listsOfContent will be needed --
same number of elements per list element, e.g.  role of the container
class should be considered