This post asks members of the R community, users and developers,<br>to comment on issues related to the GNU Public License<br>and R community policies more generally.<br><br>The GPL says very little about protecting the the rights of original<br>
contributors by not disseminating misleading information about them.<br>Indeed, for pragmatic reasons it effectively assumes that original authors<br>
have no rights regarding their GPL-ed software, and it implicitly leaves <br>it up to the community of developers and users to conduct themselves in a fair and <br>reasonable manner.<br><br>After discussing these matters with Richard Stallman I think<br>
we more-or-less agreed that a GPL "copyright" notice is nothing<br>more than a way to deputise people to serve as protectors of the<br>principles of the Free Software Foundation (FSF). It has nothing to<br>do with protecting the "rights" or the "ideas" of original<br>
contributors. There is no peer review, no requirement to<br>explain your contributions, and anybody can essentially<br>do as they please with the software provided they retain<br>the copyright/FSF deputy notice---of course, you can<br>
always work-around this last restriction by modifying the<br>implementation and placing it in a new file, because <br>nobody is checking (GPL doesn't require it).<br><br>The GPL is all about "freedom", not responsibility. It is entirely<br>
focused on "deregulation", not on the protection of intellectual<br>property or professional reputations. It serves the useful purpose<br>of making great software more widely available, but it does not<br>dictate how people should behave and should not be used<br>
as a moral compass. (See recent book titled<br>"You are not a gadget: a manifesto", a rejoinder to the<br>GNU manifesto.)<br><br>As a counterbalance I think the community of developers and <br>users need to play a more active role in the evolution of <br>
shared values and expectations. In this spirit I respectfully request <br>that the R community consider the following.<br><br>The author line of the latest release of the R package <br>Rcpp (0.8.9) was revised as follows:<br>
<br>From: "based on code written during 2005 and 2006 by Dominick Samperi"<br>
<br>To: "a
small portion of the code is based on code written during 2005
and 2006 by Dominick Samperi"<br><br>As it is highly unusual (and largely impossible) to quantify the relative<br>size of the the contribution made by each author of GPL'ed software, this has<br>effectively changed an acknowledgment into a disparaging remark. It<br>
is also misleading, because I am the original creator of the Rcpp library<br>and package (it was forked by Dirk Eddelbuettel and is now effectively<br>part of R core development). Incidentally, the README file for<br>Rcpp 0.6.7 shows that my contributions and influence were not<br>
confined to the period 2005-2006.<br><br>A look at the change history of Rcpp would quickly reveal that to be<br>fair other authors of Rcpp (and perhaps other R package authors)<br>should have their contributions qualified with "a small portion of the code",<br>
or "administered by", but this is precisely the kind of monitoring that<br>inspired Richard Stallman to say we must "chuck the masks" in the<br>GNU Manifesto.<br><br>It is obviously a great benefit for the R community to have Rcpp actively<br>
supported by the R core team. I am very grateful for this. What I do<br>have a problem with is the fact that my contributions are disparaged<br>by people who have benefited from my past work.<br><br>It seems to me that there are two possible resolutions. First, if my<br>
name is used in the Rcpp package it should be used to provide fair,<br>accurate, and courteous acknowledgement for my past contributions.<br>Second, if this is not possible, then my name should not be used at all.<br>If the second option is selected then the only place my name should<br>
appear is in the copyright ("deputy") notices.<br><br>Incidentally, the fact that the word "copyright" is profoundly misleading in<br>the context of GPL is not a new idea, and the word "copyleft" is<br>
sometimes used instead. But copyleft is not used in source files<br>because this would unlink GPL from the well-established legal<br>framework associated with "copyright", making it more difficult for<br>the FSF to enforce its principles (the critical link is provided by<br>
the copyright holders or "deputies").<br><br>A final clarification: authors of original works do retain a legal<br>copyright on their original work in the sense that they are free<br>to modify this work and release it as non-free software (or<br>
under a different free license), but this has no effect on the<br>version that was released under GPL. The latter version and<br>all of its progeny belong to the public (or to the FSF from<br>a legal point of view).<br><br>
Please feel free to express your opinion on these matters.<br><br>Thanks,<br>Dominick<br><br>