Re: R-beta: Re: S Compatibility

Ross Ihaka (ihaka@stat.auckland.ac.nz)
Thu, 1 May 1997 08:56:04 +1200 (NZST)


Date: Thu, 1 May 1997 08:56:04 +1200 (NZST)
From: Ross Ihaka <ihaka@stat.auckland.ac.nz>
Message-Id: <199704302056.IAA10227@stat1.stat.auckland.ac.nz>
To: r-help@stat.math.ethz.ch
Subject: Re:  R-beta: Re: S Compatibility

Martyn Plummer writes:

> How important is it to avoid being sued, or less facetiously, what is
> the legal status of R? If I were Mathsoft I would be less than pleased
> at the development of R and would try to stop it if I could. I have
> been wondering for some time if this is possible. But I am not a lawyer
> and the issue seems very unclear to me.

To me too (Robert spent a couple of semesters in Law School).
So far they've just tried to hire us :-).

> I suppose the argument that the S language (as opposed to the S-PLUS
> product) is public domain. Is this true? Surely S is a commercial 
> product developed by Lucent Technologies which is licensed exclusively
> to Mathsoft? Arguably (and I'm being devil's advocate here) the S
> language, and not just the source code, is the intellectual property of
> Lucent.

My understanding is that languages cannot be protected but implementations
can.  We would also have to attacked in a New Zealand court and I doubt
that they would be particularly sympathetic to Mathsoft.

John Chambers has told us that he is sympathetic and supportive of R
development and has often discussed internal details of S with us.
At the last Interface meeting he was asked whether he might make Sv4
"free".  He didn't completely rule it out, but I think the recent
Mathsoft/Lucent agreement does.

I think that John faces a real conflict.  He is at heart an academic
researcher, but his work environment has become increasingly commercial
and has driven him in a direction he is not entirely comfortable with.
Its a real tragedy for the field.  In the good old days we could get
source code and port to whatever platforms we were using.  Under those
conditions I was quite happy to contribute ideas and software (I
pointed out to Rick Becker that NA|T should return T rather than NA for
example).  Now we have to pay very large sums of money and can't even
get versions for common platforms.  With Mathsoft using Windows/NT as
their development platform how long can expect to any Unix versions to
be available?

> I know this isn't the first freeware project which is "not unlike"
> an existing commercial product.  But in other cases the commercial product
> is either a brand name which can be licensed to anyone whose software
> passes a validation test suite (eg Linux/UNIX or GNUstep/OpenStep) or else
> the freeware version is used with authorization from the commercial
> developer (eg Mesa/OpenGL). R does not fit into either category.

> Perhaps I'm worrying over nothing, but it would be nice to have
> some reassurance.

If its not R it'll be something similar.  The free software genie is not
going back into the bottle any time soon.  I had high hopes for Dylan
and am still closely watching what's going on in the Scheme world.
If there was a decent portable compiler available (Stalin and Rscheme
come close) I be tempted to switch to it as a base.

	Ross
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request@stat.math.ethz.ch
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=