[Rcpp-devel] Don't believe a word he's saying (Was: Dependence on GNU make because of $(shell))

Dominick Samperi djsamperi at gmail.com
Wed Nov 17 03:09:37 CET 2010


See my reply to Romain in the original thread.

I have nothing more to say about most of this,
except for the last bit about GNU make. This
comment was for the benefit of the R community.
It appears that it is not generally known that
in GNU make there is an important difference
between ':=' and '=', and the former should be used
in most cases because it does not cause undesirable
side-effects.

Dominick

On Tue, Nov 16, 2010 at 4:15 PM, Dirk Eddelbuettel <edd at debian.org> wrote:

>
> Samperi started to publicly accuse me on mailing lists a good year ago and
> I
> have tried successfully not to engage for a long enough time-- as it is not
> worth debating with clowns.  This however is to good to pass up, and too
> full
> of misinformation.
>
> On 16 November 2010 at 15:38, Dominick Samperi wrote:
> | On Tue, Nov 16, 2010 at 12:36 PM, Dirk Eddelbuettel <edd at debian.org>
> wrote:
> | > It's been about two years since I started to maintain and extend the
> Rcpp
> | > package which its previous maintainer had left dead and rotting in the
> | > digital archives for two years already at that point.
> |
> | Correction here. That package that was left "dead and rotting" on CRAN
> | was updated by its author (that would be me) in November of 2009 to
>
> Precisely.  November 2009 -- I was talking about November 2008.
>
> So that bings "dead and rotting" to about three full years (from the last
> November 2006 release of RcppTemplate til this three year hiatus was
> briefly
> interrupted in November 2009) before the package was once again withdrawn
> by
> Sameri, and left to rot which it has since.
>
> Withdrawn with strong words, by the way, and in direct violations of its
> GPL
> copyright terms.  But who cares about these such for details. Certainly not
> Samperi.  [ For the record, we do in fact care about the GPL. ]
>
> | include object-mapping features that were later reimplemented in what
> | is now known as the Rcpp package. The syntax used for these features
> | in Rcpp is Rcpp::as and Rcpp:wrap.
>
> This directly implies that Romain and I read Samperi's code for
> implementing
> our package. Let me be very plain and clear hear: We have not.
>
> Folks are free to believe who they will: Samperi (one random package on
> CRAN,
> no users beside himself) or us (well over a dozen packages between Romain
> and
> myself, with fairly widespread usage by other packages).  I'll rest the
> counting case here.
>
> | Shortly after I released this work the following comment was added to
> | the README file for Rcpp (version 0.8.3):
> |
> | "As of November 2009, Dominick has re-commenced maintenance of
> RcppTemplate
> | with new releases on CRAN.  This may provide a good opportunity to import
> | some new ideas into Rcpp as well, time permitting."
>
> That as a required update to the fact that README previosly said that
> RcppTemplate was witdrawn; it actually wasn't for a few days so I wrote
> something to fact that RcppTemplate was back (however briefly that turned
> out
> to be the case). The comment refered to the fact that Samperi replaced a
> rough first implementation of a data frame in C++ (still in Rcpp's classic
> API as we believe in maintaining interfaces -- rather than letting them
> rot)
> with something that looked better.  But thanks to Romain we have something
> even better anyway.
>
> As for the Make documentation hint: that's funny. Samperi was the one
> yelling
> at Uwe for suspecting that he purposely broke his builds (please...)  It
> only
> ever broke on Windoze anyway, and as I said -- the point is moot because of
> LinkingTo.  Which Samperi's silly little cxxPack still doesn't use.
>
> I guess Samperi is too busy flaming people and telling lies rather thank
> coding. Each to their own.
>
> Dirk
>
> --
> Dirk Eddelbuettel | edd at debian.org | http://dirk.eddelbuettel.com
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.r-forge.r-project.org/pipermail/rcpp-devel/attachments/20101116/65de98d1/attachment.htm>


More information about the Rcpp-devel mailing list