[R] GPL or not (was "(no subject)")

Wed Nov 10 11:35:21 CET 2021

>>>>> Arnaud FELD 
>>>>>     on Tue, 9 Nov 2021 23:03:40 +0100 writes:

    > Thanks for your answer. That is helpful even if that is disappointing
    > for me ;) I can see now that, as R (base and recommended packages) has
    > had contributions from a lot of people, it's not like a package, you
    > just can't have a punctual "okay do what you want". I'll have to stick
    > with a slow window call, I guess.

    > I was right to ask anyway, and thanks for your advices.

Rather you should think long and hard if and why you  really
really want to stick with a non-GPL licence for you package.
That is the real problem and a shame in my eyes!

R would almost surely never have grown big if the original authors,
Ross Ihaka and Robert Gentleman  hadn't decided to adopt the GNU Public 
Licence (GPL)  very early ... and Ross & Robert said it was to a
large extent because I had insisted that this was important and
would potentially bring quite a few highly qualified people (all
from academia at the time) to adopt and contribute to R.

So, *not* using  GPL  for your package feels slightly as a slap
in the face of all the people who made R what it is today.

and just for those it is not obvious:
You *could* easily copy part of R code into your package if
you switched to use GPL -- of course still attributing these
parts a copyrighted by the R Core team.
Quite a few package maintainers have done so, e.g.,
testthat, actuar, pryr, vetr, DPQ, round, ... {I count about 80
CRAN packages}

Martin

--
Martin Maechler
ETH Zurich  and  R Core team

    > Le mar. 9 nov. 2021 à 22:51, Duncan Murdoch <murdoch.duncan using gmail.com> a écrit :
    >> 
    >> On 09/11/2021 3:52 p.m., Jeff Newmiller wrote:
    >> > I understand that, but the reason he gives for copying code from the stats package is to change the license. That decision seems like something that requires a very direct communication with/permission from the maintainer.
    >> 
    >> The stats package doesn't give a valid maintainer() address.  It gives a
    >> vague reference to the r-project.org website, but that website doesn't
    >> give a contact address.
    >> 
    >> In fact, the authorship of R is very widely distributed among at least
    >> dozens of contributors.  Identifying and contacting them all would be a
    >> nightmare but probably isn't necessary.  However, identifying and
    >> contacting the necessary ones (the ones who hold copyright on the parts
    >> he wants to copy) would be very difficult.  Not contacting the relevant
    >> ones would mean Arnaud's work would be potentially in violation of their
    >> copyright.
    >> 
    >> Even if he did manage to contact all the copyright holders, some of them
    >> would probably not agree to more permissive licenses.  For example, I am
    >> a copyright holder on some parts of the source, and I wouldn't agree to
    >> relicensing.
    >> 
    >> Duncan Murdoch
    >> 
    >> 
    >> 
    >> >
    >> > On November 9, 2021 12:21:18 PM PST, Duncan Murdoch <murdoch.duncan using gmail.com> wrote:
    >> >> On 09/11/2021 2:45 p.m., Jeff Newmiller wrote:
    >> >>> This question isn't a "how to do package development" question... this is about a specific package so you should send email to the package developer identified by the maintainer() function.
    >> >>
    >> >> I think Arnaud is the package developer; his question is whether he can
    >> >> copy R source into his package.
    >> >>
    >> >> Duncan Murdoch
    >> >>
    >> >>>
    >> >>> I can't foresee this request finding a positive response from R Core, but email seems the most correct approach.
    >> >>>
    >> >>> On November 9, 2021 11:34:12 AM PST, Bert Gunter <bgunter.4567 using gmail.com> wrote:
    >> >>>> Questions about package development should be posted to
    >> >>>> R-package-devel (**not R-devel**).
    >> >>>> See https://www.r-project.org/mail.html  for details.
    >> >>>>
    >> >>>> (I am not sure that they get into legal weeds there, but it seems like
    >> >>>> the right place to try).
    >> >>>>
    >> >>>> Bert Gunter
    >> >>>>
    >> >>>> "The trouble with having an open mind is that people keep coming along
    >> >>>> and sticking things into it."
    >> >>>> -- Opus (aka Berkeley Breathed in his "Bloom County" comic strip )
    >> >>>>
    >> >>>> On Tue, Nov 9, 2021 at 11:17 AM Arnaud FELD <arnaud.feldmann using gmail.com> wrote:
    >> >>>>>
    >> >>>>> Hi,
    >> >>>>>
    >> >>>>> I use that mailing list because I’ve tried to send a message to the
    >> >>>>> r-core address and received :
    >> >>>>>
    >> >>>>>> "Non-members are typically *NOT* allowed to post messages to this
    >> >>>>>> private developers' list. Please use an appropriate mailing list (from
    >> >>>>>> http://www.r-project.org/mail.html). For R packages, use
    >> >>>>>> maintainer("<pkg>") in R (and if that is R-core using .., use the R-help
    >> >>>>>> address).
    >> >>>>>
    >> >>>>>    I would like to borrow most of the code of stats :: window within my
    >> >>>>> package disaggR, which has an uncompatible license as it is MIT not
    >> >>>>> GPL2. window has some extra overhead because of the calling of time()
    >> >>>>> at the beginning of the function (that is only used for extend =
    >> >>>>> FALSE). Copying the function and modifying it would allow me a bit of
    >> >>>>> optimization, but I need some agreement as my package is MIT.
    >> >>>>> Is it possible ?
    >> >>>>>
    >> >>>>> Thanks,
    >> >>>>>
    >> >>>>> Arnaud FELDMANN
    >> >>>>>
    >> >>>>> ______________________________________________
    >> >>>>> R-help using r-project.org mailing list -- To UNSUBSCRIBE and more, see
    >> >>>>> https://stat.ethz.ch/mailman/listinfo/r-help
    >> >>>>> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
    >> >>>>> and provide commented, minimal, self-contained, reproducible code.
    >> >>>>
    >> >>>> ______________________________________________
    >> >>>> R-help using r-project.org mailing list -- To UNSUBSCRIBE and more, see
    >> >>>> https://stat.ethz.ch/mailman/listinfo/r-help
    >> >>>> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
    >> >>>> and provide commented, minimal, self-contained, reproducible code.
    >> >>>
    >> >>
    >> >
    >> 

    > ______________________________________________
    > R-help using r-project.org mailing list -- To UNSUBSCRIBE and more, see
    > https://stat.ethz.ch/mailman/listinfo/r-help
    > PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
    > and provide commented, minimal, self-contained, reproducible code.