dist(*, "euclidean") [was "dist function suggestion"]

Martin Maechler Martin Maechler <maechler@stat.math.ethz.ch>
Wed, 20 Jan 1999 14:22:45 +0100

>>>>> "BDR" == Prof Brian Ripley <ripley@stats.ox.ac.uk> writes:

    >> Date: Wed, 20 Jan 1999 12:03:38 +0000 (GMT)
    >> From: Jonathan Rougier <J.C.Rougier@durham.ac.uk>
    >> To: Friedrich Leisch <Friedrich.Leisch@ci.tuwien.ac.at>
    >> cc: r-devel@stat.math.ethz.ch
    >> Subject: Re: dist function suggestion
    >> On Wed, 20 Jan 1999, Friedrich Leisch wrote:
    >> > dist is contained in the mva package which comes with (base) R. it
    >> > provides `"euclidian"', `"maximum"', `"manhattan"', `"canberra"' and
    >> > `"binary"' distances both in matrix and triangular form.
    >> Thanks Fritz, but could I still make a general plea for a dist() that
    >> takes one OR two matrices?  I often need to find dist(X, Y) as a
    >> rectangular matrix (eg covariance calculations), and it is very
    >> inefficient to use dist(rbind(X, Y))  and then discard most of it. 

    BDR> You will need to call it something else: dist is a clone of an S
    BDR> function, and dist(X, "manhattan") is well-established usage.

one could still imagine an extra Y argument such that
	dist(X, Y=myY, method="euclidean")
and	dist(X, "euclidean", Y=myY)
would work
one could even make it such that
	dist(X, myY)
and	dist(X, "euclidean")
would work.  However, the extra hack for 	dist(X, Y)
may not be worth the pain.

    BDR> I pointed out to R-core recently that "euclidian" will cause
    BDR> problems for those who know how to spell it .... (-ean in my
    BDR> dictionaries, and in S).

Yes, thank you;  for me, I've considered this a bug report.
Hopefully nobody has currently code relying on the wrong spelling...


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