[Rd] Accuracy problems in summary()? (PR#4101)

Martin Maechler maechler at stat.math.ethz.ch
Tue Sep 9 16:09:34 MEST 2003

>>>>> "MM" == Martin Maechler <maechler at stat.math.ethz.ch>
>>>>>     on Mon, 8 Sep 2003 16:09:58 +0200 (MET DST) writes:

    MM> A bug report with a "?" is **wrong**, almost by
    MM> definition!  

I was wrong here.  As Grame has pointed out (privately), the "?"
can well be used to indicate that it's not clear  *where* the
bug is. 

    MM> (and your definitely is).

because  help(summary)  really tells you about the `digits ='
argument which is applied with some inconsistency in the
different summary() methods.
The reason for this has been S(-plus) compatibility that we
didn't want to drop too quickly.

    MM> Please ask questions on R-help.

or proposals about changing R behavior on R-devel.
I think there's a point here that could be discussed: Changing
current behavior while trying to remain S compatible as much as

    MM>   And this has become  almost a FAQ.
an exaggeration.
But people have been told quite a few times to use something
like  summary(*, digits = 7) if they wanted different behavior.

    MM> Note that sending bug-reports for non-bugs can be quite
    MM> counter-productive.  The waste of time it produces to R
    MM> developers will not improve the quality of development
    MM> nor answers...

    MM> (yes, I'm grouchy.....)

and hereby apologize to Graeme.

Martin Maechler <maechler at stat.math.ethz.ch>	http://stat.ethz.ch/~maechler/
Seminar fuer Statistik, ETH-Zentrum  LEO C16	Leonhardstr. 27
ETH (Federal Inst. Technology)	8092 Zurich	SWITZERLAND
phone: x-41-1-632-3408		fax: ...-1228			<><

More information about the R-devel mailing list