Message-ID: <17305.36909.632949.308900@celebrian.ci.tuwien.ac.at>
Date: 2005-12-09T14:09:49Z
From: Friedrich Leisch
Subject: [RsR] Package title
In-Reply-To: <17305.34945.268465.777346@stat.math.ethz.ch>
>>>>> On Fri, 9 Dec 2005 14:37:05 +0100,
>>>>> Martin Maechler (MM) wrote:
>>>>> "Jean-Christophe" == Jean-Christophe BOUETTE <jcbouette at gmail.com>
>>>>> on Fri, 9 Dec 2005 13:47:41 +0100 writes:
Jean-Christophe> Hi everybody
Jean-Christophe> Ok, if a votation is going to take place I vote for
Jean-Christophe> robustat :-)
> you mean mean(c(robusta, robustats))
> or rather huber(c(robusta, robustats)) ;-) ?
> The reason for "*stats" was that there's already the basic
> "stats" (and "stats4") package...
Using google to evaluate vocabulary seems popular these days (see the
thread on Gibraltar on r-devel ;-), so let's see:
As someone (forgot who) has already mentioned in this thread, googling
gives 2.1 million hits for "robusta", 70 for "robustat" and 9 for
"robustats".
So together with the very good argument on "stats" and "stats4" by
Martin I'd strongly vote for "robustats". If we put a package of that
name on CRAN I expect it to be the Google top hit for its name within
a few weeks (if not days).
Just my 2c,
Fritz
PS: Of course we may also make the top hit for "robusta" ... googling
for the single letter "R" returns the R homepage as top link for a
couple of years now. But it is not really necessary to make searching
hard ... and searching for "robusta", e.g., in publication databases
may return a large number of biological articles for a long time.