Hi Dirk, You might have missed the point I made in my email which shows that some people from Debian (!) took exactly the position you believe does not exist and removed the binary version of afex from the Debian repositories. I link to the relevant Debian bug report again below which contains the following claims:
> I believe the r-cran-afex package violates the GPL. It is released under GPL3+, however it has GPL2-only dependencies (r-cran-stringr, r-cran-coin, r-crain-lsmeans), and it calls functions from these packages directly. > The nature of the interaction (function calls) means that r-cran-afex and its dependencies constitute a single work, and not two separate programs, as per the GPL faq: [...] See here: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=800891 While I also believe that this is likely a misinterpretation of the GPL, I was not in the mood to fight against this and simply changed my package's license. So while we might all be in agreement here, clearly not everyone sees it this way. Best, Henrik Am So., 14. Sept. 2025 um 13:38 Uhr schrieb Dirk Eddelbuettel < [email protected]>: > > Hi Ilmari, > > Am I reading this correctly that you imply that a relationship arising from > package A calling library(B), or having a more Imports or Depends on it, > would lead to same GPL "virality" as having actual object code from an > external package? > > The latter case (ie object code and linking) clearly is what the GPL has in > mind, where virality matters, and where it is commonly applied. The > established consensus, as best as I can tell, always has been is that the > former (an R package calling another R package) is not, and the virality > argument you are making does not apply. So I think you are fighting a > fighting that does not exist. > > Cheers, Dirk > > -- > dirk.eddelbuettel.com | @eddelbuettel | [email protected] > -- Dr. Henrik Singmann Associate Professor, Experimental Psychology University College London (UCL), UK http://singmann.org [[alternative HTML version deleted]] ______________________________________________ [email protected] mailing list https://stat.ethz.ch/mailman/listinfo/r-package-devel
