[R] reference for ginv

[Wacek Kusnierczyk]

?ginv provides 'Modern Applied Statistics with S' (MASS), 3rd, by
Venables and Ripley as the sole reference.
I happen to have this book (4th ed) on loan from our library, and as far
as I can see, ginv is mentioned there twice, and it is *used*, not
*explained* in any way.  (It is used on p. 148 in the 4th edition.)

ginv does not appear in the index of MASS.  ginv is an implementation of
the Moore-Penrose generalized inverse, but I can't find any reference to
an appropriate publication from Moore and Penrose in MASS either.

Admittedly, MASS is inadequate as a reference for ginv, and it is
desirable that the reference be corrected.  Is this an isolated
incident, or a general practice?  (Perhaps the appropriate reference has
been removed in MASS 4th?)

vQ