R-alpha: This weeks bugs and requests for enhancements

Peter Dalgaard BSA p.dalgaard@kubism.ku.dk
18 Oct 1997 20:32:02 +0200

Here's a number of items that came up in connection with my course for 
medical ph.d. students this week:

        - deals ungracefully with empty groups and factor levels not
	present in grouping variable.
        - no indication of what variable is being plotted

data.entry() (mostly W95 rseptbeta problems)
        - Entry of first variable name doesn't take before 2nd
        name is entered.
        - data.entry(BW,RMR) gives two columns named BW. Quitting
	dataeditor at that point *overwrites* BW with RMR!
        - inserting NA is only possible by switching to character
	values, entering invalid number, and switching back to numeric
        - blanking a field, then exiting dataeditor crashed one system
	(I *think*)

        - insert/delete rows (in single column *and* range of columns)
        - undo facilities and "panic button"
        - use comma as decimal point (Nat.variant of W95 maps Keypad
	period to comma)
        - switch between vertical/horizontal cell cursor movement on

	quantile() labels results as % rounded to integer, which is
	kind of confusing when 2.5% and 97.5% is one of the first
	things people want to try:
	> quantile(rnorm(1000),c(.025,.975))
	       2%       98% 
	-2.067519  1.891768 

        plot() still has inconsistency with docs in xlog/ylog vs
	log="xy". Which one do we want?

        at some point we want to be able to specify collating
	sequences for sorting character vectors. Case in point: The
	Danish alphabet ends ..uvwxyz , but
        > sort(c("","",""))
        [1] "" "" ""
        (No such thing as a universal collating seq. E.g. Norwegian
	has the same three letters in a different order. So it should
	probably be via options(collate=..).

        The ~ notation could be used more consistently, in particular
	in t.test and friends.

   O__  ---- Peter Dalgaard             Blegdamsvej 3  
  c/ /'_ --- Dept. of Biostatistics     2200 Cph. N   
 (*) \(*) -- University of Copenhagen   Denmark      Ph: (+45) 35327918
~~~~~~~~~~ - (p.dalgaard@biostat.ku.dk)             FAX: (+45) 35327907
r-devel mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-devel-request@stat.math.ethz.ch