> The license will need to a better job of protecting work donated to the commons than GPL2 seems to have done. I’m not willing to have any more of my work purloined by the likes of Revolution Analytics, so I’ll be looking for better protection from the license (and being a lot more careful about who I work with).
Not having used R or being more than passingly familiar with it, I'm wondering if anyone could shed some light on what this is about? I notice that on the Revolution Analytics website, they are selling an "enterprise" version of R which they claim has a number of advantages over the mainstream form of R.
How exactly is a proprietary form of R even able to exist if R's codebase is GPL'd?
Revolution Analytics has released some extensions that are proprietary, for instance their doSMP library that provides multicore support under Windows. They have also released several of their libraries into the "commons" -- doMC and foreach are the two that immediately come to mind.
I'm sort of torn on this issue because setting aside the terms under which RA releases their libraries, what they produce comes with great documentation, and tends to be pretty useful. Using foreach and doMC I was able to cut down a calculation that normally takes 8 hours to 3 hours on a 6 core machine.
On the other hand, I strongly believe that proprietary code should be avoided when doing scientific research because it inhibits peer review and makes it harder for others to replicate your work. As Warren DeLano said: "The only way to publish software in a scientifically robust manner is to share source code, and that means publishing via the internet in an open-access/open-source fashion."
And Robert Gentleman has always been one of the strongest proponents of reproducible research, which includes being able to create packages that can be shared and freely distributed.
Yes. CRAN R is licensed under the GNU General Public License (version 2). As permitted under that license, Revolution R is based on a modified version of CRAN R. The source code to that modified version (including a list of changes) is available for download when you download the binary version of Revolution R, as required by the GPL.
As an open-source company, Revolution Analytics respects the Free and Open Source Software philosophy and respects all free-software licenses. Revolution Analytics is also a direct supporter of the open-source R project: financially (as a benefactor of the R Foundation), organizationally (by sponsoring events and user group meetings), and technically (by contributing or modifications to CRAN R back to the community).
Hmm, that seems to be interpreting the GPL as closer to what the LGPL's intent is. I assume they're able to do this because their other tools (e.g. IDE) are careful not to actually link to R, but only call it via things like shared files and pipes?
RA claims to have adopted this "arm's length" approach, releasing their direct extensions to the community. They ask me for identifying information to let me download the software; don't know if that's against the rules. Critically, their enterprise product is pitched as a collection of separate tools: IDE, debugger, analysis tool, etc.
There are a few other ways that people have made GPL'd code proprietary: (1) providing a hosted service instead of distributing software, or (2) distributing only a hardware appliance that contains inextricable software. The latter is controversial with respect to GPL2; GPL3 more clearly prohibits it.
It's interesting that he saw that has a violation of the commons. I don't know the details, but it seems like a feature rather than a bug that businesses can be built around open-sourced code.
R has been really surprisingly shoddy about it's license enforcement on dynamicly-loadable extensions (esp. for an official GNU project). Crucially, they declared an extremely useful subset of their API as usable for non-GPL works (similar to how the two flavors of drivers in the linux kernel work) and people now drive trucks through it all the time. Early on this benefited R, though.
Not having used R or being more than passingly familiar with it, I'm wondering if anyone could shed some light on what this is about? I notice that on the Revolution Analytics website, they are selling an "enterprise" version of R which they claim has a number of advantages over the mainstream form of R.
How exactly is a proprietary form of R even able to exist if R's codebase is GPL'd?