Re: R-beta: R-0.50-a4 library problem (survival4)

Thomas Lumley (
Wed, 24 Sep 1997 09:02:56 -0700 (PDT)

Date: Wed, 24 Sep 1997 09:02:56 -0700 (PDT)
From: Thomas Lumley <>
To: U-E59264-Osman Buyukisik <>
Subject: Re: R-beta: R-0.50-a4  library problem (survival4)
In-Reply-To: <>

On Wed, 24 Sep 1997, U-E59264-Osman Buyukisik wrote:

> I have built R-0.50-a4 on HP9000/C160-hpux10.2 with gcc (cc for
> libs). When I try 
> library(survival4)
> Error in : invalid "pos" argument
> And it does not load. Other libs seem to work. I tested integrate (has
> f2c translated c), polynomial.

I'm not sure that r-help is the best place for this problem: the theory is
that r-devel is for people who want to be exposed to the full horrors of
prerelease versions of R. In any case, there have been two questions about
survival4, so I am giving a full description of the problem.

In R-0.50-a4 model.frame() has been made into a generic function and has
been improved in various ways.  This breaks the gee and survival4
packages. I have not sent the revised versions to CRAN as they don't work
with R-0.49, the current "stable" release, and they won't work with the
next release R-0.60, which includes major reorganisations of the package
system. Also, the S gee() package is being rewritten by its author, Vince
Carey, and I am hoping to port the new version soon.

A probably working, full version of survival4, including the expected
survival functions, is available from my homepage. This will work with
R-0.50-a4 *only*. 
The survival4 library is at
and the US mortality tables are in the separate library

Thomas Lumley
Biostatistics		: "Never attribute to malice what  :
Uni of Washington	:  can be adequately explained by  :
Box 357232		:  incompetence" - Hanlon's Razor  :
Seattle WA 98195-7232	:				   :

r-help mailing list -- Read
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: