R Style Guide -- Was Post-hoc tests in MASS using glm.nb
Thanks Martin, Your points are, of course, well taken. Nevertheless, I still think it might be useful to put a link or links to one or more style guides in the FAQ with a comment to the effect that these are various recommended ways to help write better, more readable code. Something like: ------------------------------ Q: What are best practices for R coding style? A: There is no simple answer to this question, as programming style is legitimately a personal choice and may depend on the nature of the programming task. However useful guidelines and alternatives can be found at <links to one or more style guides>. ------------------------------- One important point: The links need to be reasonably stable, and this could be problematic. However, I am happy to defer to R Core and experienced R programmers like yourself on these matters. Cheers to all, Bert On Thu, May 19, 2011 at 3:26 AM, Martin Maechler
<maechler at stat.math.ethz.ch> wrote:
? ?BertG> Thanks Bill. Do you and others think that a link to
? ?BertG> this guide (or another)should be included in the
? ?BertG> Posting Guide and/or R FAQ?
Hmm, that guide is Google's work, and is probably quite good if
you have a group of R programmers in the same company,
but e.g., has not been published in collaboration with the R core team,
and actually somewhat differs from our own .. much less formal
and not officially laid down styles
{yes: "s".. but we still have a few parts we agree on...}
2nd problem with any style guide: "Base R" already comes
with several thousand of functions, classes and other objects, which
have grown from more than 20 years of S, S+ and then R history,
and most things cannot feasibly be changed now (or could, say 5
years ago), for back compatibility reasons.
Further,... more philosophically:
For many of us, programming (R or otherwise) is considered
a creative activity to quite some extent, and creativity can be
crushed by too rigid rules. ?I'd state that cultural history
shows that human culture implicitly follows many rules, but it
is (almost) only interesting because the are enough exceptions
to those rules.
Martin Maechler
? ? ? @ ETH Zurich and R Core Team since its inception
? ? ? but speaking entirely for myself ..
"Men by nature long to get on to the ultimate truths, and will often be impatient with elementary studies or fight shy of them. If it were possible to reach the ultimate truths without the elementary studies usually prefixed to them, these would not be preparatory studies but superfluous diversions." -- Maimonides (1135-1204) Bert Gunter Genentech Nonclinical Biostatistics