[R] "privileged slots",

[Tony Plate]

At Tuesday 03:44 AM 6/1/2004, Jari Oksanen wrote:
> > [snip]
>There are several other things that were fully documented and still were
>removed. One of the latest cases was print.coefmat which was abruptly
>made Defunct without warning or grace period: code written for 1.8*
>didn't work in 1.9.0 and if corrected for 1.9.0 it wouldn't work in
>pre-1.9.0. Anything can change in R without warning, and your code may
>be broken anytime. Just be prepared.

This is true of many software packages.  In our production environment we 
often (usually) run older versions of software, including statistical 
software, because of bugs or changed behaviors (or fears thereof) in new 
versions.  We usually run the latest versions in our test and 
non-production systems and only upgrade our production systems when two 
conditions are satisfied: (1) we need the features in the upgrade and (2) 
we are comfortable that the upgraded package will run reliably.  From what 
I can see, R is only distinguished from other software packages in these 
regards by the extreme speed with which bug fixes for the latest version 
are made available (in contrast, we're still waiting more than a year for 
fixes for bugs in some commercial software that were described as 
"critical" bugs by the vendor's support team) and the high level of respect 
accorded to users by the core developers (changes are debated and effects 
on existing software seem to be taken seriously).

One very helpful tool to deal with software updates is automated 
testing.  I highly recommend it.  R comes with a testing framework.

-- Tony Plate

>cheers, jari oksanen
>--
>Jari Oksanen <jarioksa at sun3.oulu.fi>
>
>______________________________________________
>R-help at stat.math.ethz.ch mailing list
>https://www.stat.math.ethz.ch/mailman/listinfo/r-help
>PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html