[Rd] Rcpp: Clarifying the meaning of GPL?

From: Dominick Samperi <djsamperi_at_earthlink.net>
Date: Tue, 22 Dec 2009 13:14:15 -0500

I wrote the Rcpp library and the RcppTemplate package to make it easier for developers to contribute packages to the R community. In addition to providing detailed documentation on package creation it provides a clean object mapping between R anc C++ that helps developers to implement packages that benefit from the performance of C++ and the flexibility of R.

The package named 'Rcpp' was forked from my work and is being developed independently, in spite of many protests from me. A diff of Rcpp_0.6.6 and RcppTemplate_5.3 (written several years ago), both available at CRAN, will show that Rcpp added a few cut-and-paste changes. (The latest release of Rcpp has been split up and reorganized so that it would be difficult to find the differences now.)

More importantly, while GPL gives developers the right to make changes (without the permission of the original contributor) it explicitly states that these changes should not leave misleading impressions about the original developer.

Unfortunately, GPL does not spell out what it means to be misleading. I think using the same name ('Rcpp') as a library still being developed by the original author, and listing yourself as a copyright holder on source code alongside the original author without that person's permission counts as misleading, but that is my opinion.

I am posting this message seeking the opinion of others in the R community. Perhaps by sharing ideas we can "self-organize" and find an interpretation of GPL that benefits all R users, and all package contributors as well.

A minimal resolution of this issue would be to simply rename 'Rcpp' to something like 'RInside', or to something else that is not misleading.


R-devel_at_r-project.org mailing list

https://stat.ethz.ch/mailman/listinfo/r-devel Received on Tue 22 Dec 2009 - 18:19:10 GMT

This archive was generated by hypermail 2.2.0 : Wed 23 Dec 2009 - 05:41:12 GMT