[R-pkg-devel] R CMD check "no-nanny" option?

[Thomas Petzoldt]

Am 6/11/2015 um 4:55 PM schrieb Dirk Eddelbuettel:
>
> On 11 June 2015 at 09:15, Kevin Coombes wrote:
> | Grumpy Observation: In reaction to (1), I went to CRAN and randomly
> | clicked on a dozen packages to see how many of them follow this rule.
> | Half of them violate it explicitly, mostly just starting with "This
> | package...".  Most of the rest get around it by starting with a sentence
> | fragment that omits the subject (which is _implicitly_ "this package").
> | Only one of them started with a complete sentence that gave the
> | background of the problem the package was intended to solve.
>
> Also look at the Date: of those packages -- CRAN Policy changes have always
> been implemented on a rolling basis.   So new(er) packages conform to new(er)
> tests, updates are always never forced on already-uploaded packages.

Yes, that's true, but may also create headaches to me when regularly 
updating packages. I appreciate high standards and clear rules, given 
that their motivation is obvious or clearly explained.

Therefore, I do not completely understand yet the motivation of "the 
Description field in the DESCRIPTION should file should not start with 
'This package' or with the package name ..." ? Where is it explained?

Now, what is the goal, a relatively long abstract with a general problem 
description and then the solution, like in a scientific paper - or a 
short and efficient paragraph what the package does, especially as 
"Writing R Extensions" tells us:

"Description: A (one paragraph) description of what
the package does and why it may be useful."


I think that the "first sentence is a sentence fragment with implicit 
'this package'" - style is not so bad, cf.:

graphics: R functions for base graphics.

stats:  R statistical functions.

lme4:
Fit linear and generalized linear mixed-effects models.The models and 
their components are represented using S4 classes and methods. The core 
computational algorithms are implemented using the Eigen C++ library for 
numerical linear algebra and RcppEigen "glue".

zoo:
An S3 class with methods for totally ordered indexed
observations. It is particularly aimed at irregular time series
of numeric vectors/matrices and factors. zoo's key design goals
are independence of a particular index/date/time class and
consistency with ts and base R by providing methods to extend
standard generics.

ggplot2: An implementation of the grammar of graphics
in R. It combines the advantages of both base and
lattice graphics: conditioning and shared axes are
handled automatically, and you can still build up a
plot step by step from multiple data sources. It also
implements a sophisticated multidimensional
conditioning system and a consistent interface to map
data to aesthetic attributes. See http://ggplot2.org
for more information, documentation and examples.


I'm still searching for good examples ...

> The communication of such changes is not quite what I liked it to be.

I agree, so it is a big step forward that we now have this list.

> But
> if you care you can subscribe to the output stream of a cron job I have
> watching over this:
>
>    https://twitter.com/CRANPolicyWatch
> If you don't like twitter subscription you can still glance at that page.
> I should probably do the same for the SVN source of Writing R Extensions in
> R-devel...

Yes, that's what I +/- regularly did.

Many Thanks,

Thomas

-- 
Dr. Thomas Petzoldt
Technische Universitaet Dresden
Faculty of Environmental Sciences
Institute of Hydrobiology
01062 Dresden, Germany

E-Mail: thomas.petzoldt at tu-dresden.de
http://tu-dresden.de/Members/thomas.petzoldt