core R package in universe nags me to install proprietary software at every start-up

Bug #433799 reported by Nathaniel Smith on 2009-09-21
This bug affects 8 people
Affects Status Importance Assigned to Milestone
r-base (Ubuntu)
Nominated for Karmic by ChrisDesjardins

Bug Description

Since upgrading to Karmic alpha, whenever I start R (r-base-core 2.9.2-1ubuntu2) I get the following text:
R version 2.9.2 (2009-08-24)
Copyright (C) 2009 The R Foundation for Statistical Computing
ISBN 3-900051-07-0

R is free software and comes with ABSOLUTELY NO WARRANTY.
You are welcome to redistribute it under certain conditions.
Type 'license()' or 'licence()' for distribution details.

R is a collaborative project with many contributors.
Type 'contributors()' for more information and
'citation()' on how to cite R or R packages in publications.

This is REvolution R version 2.0.1: the optimized distribution of R from R
Evolution Computing
REvolution R enhancements Copyright (C) REvolution Computing, Inc.

Checking for REvolution MKL: REvolution R enhancements not installed. For
improved performance and other extensions install with: apt-get install re

I.e, upstream's 9 lines of GPL boilerplate with succint information on usage and pointers to more information, and then Ubuntu has apparently added 6 lines of somewhat-garbled advertising for a proprietary product.

There are a few things wrong with this:
1) It's way too long, and in the wrong place -- if you're including enhancements from REvolution Inc. then by all means give them credit, but make it a single copyright line next to the existing copyright line. Using their trademarked name 6 times, their slogan, and putting it all at the bottom is wholly inappropriate. They should not be more prominent than the actual, y'know, R core team!

2) The added text is fluffy and not at all information dense -- for instance, compare the amount of information that the upstream text gives you about licensing to what we learn from the text at the bottom. Running 'apt-get install revolution-r', as suggested, will install some utterly proprietary EULAed gunk on my system, but there is no indication of this whatsoever to let me make an informed decision. (I only know because I installed it, noticed it was pulling from multiverse, went WTF and checked the copying file.) Nor do I know *why* I would want REvolution R. (Apparently I get both "performance" *and* "enhancements"? Lucky me?)

2.5) Also it's *garbled* on an 80 character terminal, which hardly seems ideal.

3) And worst of all, taking a piece of free software and added nagging for one company's products is completely uncool. I'm not going to start ranting about precious software fluids -- I don't disagree with Ubuntu's approach to handling the NVidia kernel module, for instance -- but for me this is way over the line.

Frankly, I feel betrayed. It's like some part of Canonical transmogrified into Joerg Schilling or something. Don't make Joerg Schilling my image of Canonical, please?

As a practical matter, I can also imagine this creating a nice internet flamefest/PR obnoxiousness ("it's like adding advertisements for MS Word to openoffice!" yada yada, you can imagine what people would say, esp. since AFAICT there is no public information anywhere on what you are trying to do or what your relationship with REvolution Inc. is). I doubt anyone wants this.

- If I install R from universe, please do not tell me anything about proprietary software on every startup. If I want it then I can find it myself. (Well, or could if you had information about it anywhere... maybe my google-fu is just weak.)

- Anywhere that you do tell people about proprietary software, please make it clear that that is what you are doing.


bbolker (bolker) wrote :

+1 ... (not a Karmic user yet -- actually, I'm still back at intrepid -- but this is an important issue which has started to come up on the R mailing lists)

ChrisDesjardins (cddesjardins) wrote :

I agree. This should be removed especially as it's misinforming the user that they are running a modified version of R when in fact you have to install that revolution-r package. Let us decide what R packages we want installed. Please remove this.

ChrisDesjardins (cddesjardins) wrote :

Also for those annoyed by this you can comment out that message in /etc/R/ ... but really this seems like such a simple bug to fix and hopefully it will be fixed before Karmic is released.

Fred (fredrooks) wrote :

The r-base-core in universe is now adware. This is unacceptable. If this bug isn't fixed soon, I will be forced to replace all my Ubuntu installations with another distribution, as I could not conscientiously tolerate the use of free software as a vector of advertisements for proprietary software.

Andy Choens (andy-choens) wrote :

I think it is Canonical's right to enter into agreements with other companies to distribute their software (regardless of the license). That being said I think it is equally important to respect the end-users. I don't like the adware and I especially don't like the fact that the extensions are not clearly labeled as proprietary.

If this product provides a useful service to someone and they are willing to work within the constraints of the license, I think they should use it. That being said, end-users should be told up front when they are installing non-free software. This message need not be derogatory, but it should make it clear that the license is proprietary.

I would also like to offer a couple of suggestions here.

1) I think something like the App Center is a better place to position advertisements for related products. A well designed App Center SHOULD help me find new Free and non-Free software that is related to or similar other applications that I am interested in or have already installed. For example, if I searched for R in the App Center, I think it is reasonable to get a list of base packages that include PSPP, Gretl, and Revolution R, and a few dozen other R extensions. If I am interested in R, it is reasonable to assume I may be interested in these other products as well. As long as the App Center clearly displays the licenes (GLP, Apache, LGPL, Proprietary, etc.), end users can make a well-informed decsion regarding what software to use.

2) I always thought Java's approach to installation was a good one. The old non-free Java installation made you accept or decline the EULA. This way I was given the opportunity to review the license and decide for myself if I wanted to use it or not. I think this is a good way to install other non-free software such as the Microsoft core fonts, Flash, etc. Make it easy, but also give end-users the opportunity to see the license before installing the software when the license is not a Free Software license.

3) Remove or at least minimize the advertisement for the stuff in R.

From my view (correct me if I'm wrong), I think that string is mostly
making users aware of the fact that developers put work into getting
those two packages to work together:

I don't think we can jump to the conclusion that there was a financial
reason for that "advertisement" (I can't find record of it). However,
it is intrusive and does ask users to install non-free software.

r-base is in universe, we can write a patch to tone done the language
(or even remove it). Those developers that wrote it are quite good at
what they do, so I believe it is there for a reason - so perhaps we
should submit a patch with and ask for an explanation as to why the
text is there.

This file:

Nathaniel Smith (njs) wrote :

No-one is objecting to their right to publicize their work -- all the comments here are objecting to the *form* this publicity took. There are a wide range of ways to provide information from users, from putting up a web-page to adding misleading advertisements to GPL'ed software, and they are currently doing the latter. This is completely uncool.

AFAICT, your comment is the first to bring up potential financial reasons for this decision? (Obviously Canonical and REvolution R have negotiated some sort of deal, though, or Canonical couldn't be distributing revolution-r at all; if they'd like to describe the terms of this deal then that would be a wonderful way to pre-empt any speculation, but so far they seem to have decided not to do that.)

Please do submit a patch, if you have universe rights; perhaps it will produce something useful. But we asked for an explanation a month and a half ago with this bug report and the result has so far been silence :-(.

Ondra Pelech (ondra-pelech) wrote :

i'm just a common user, but I don't like ubuntu becoming an adware...
ubuntu developers made an claim about their philosophical views on the free software [1] and how this term fits into the ubuntu's philosophy, and I would appreciate really much to stay it the same way, thank you very much!
I mean it, I would like to express my thanks to all ubuntu contributors and developers, but I find this direction that ubuntu is heading (at least according to what I see in the R package) very disappointing.


Scott Howard (showard314) wrote :

Here's the simple debdiff patch to fix this. I don't think this is SRU-able in Karmic, but at least we can try for lucid. Subscribing MOTU-sponsors to take a look at this as well.

Changed in r-base (Ubuntu):
status: New → Triaged
importance: Undecided → Wishlist
Nathaniel Smith (njs) wrote :

Thanks for the patch. As it stands, I don't think it really fixes the problem -- it just makes the ad more vague. (It now simply prints "REvolution R not installed." at every startup.) This is still clearly an ad -- there are thousands of things I don't have installed, including many R extensions, and no-one feels the need to remind me of them every day! But now it's an even less useful ad. This seems like the worst of both worlds.

As for whether it's SRU-able... I dunno. Either the Ubuntu community agrees that this is a betrayal of their philosophy and users, in which case it seems SRU-able if anything is, or else they don't, in which case there is no bug.

I apologize if words like "betrayal" in the previous paragraph sound like hyperbole, but with all respect, Scott, I don't think you realize the extent to which this bug is playing with fire. It's a small thing by itself, but to many people it seems to threaten a fundamental shift in Ubuntu's philosophy. In Debian, this would be an automatic release-blocker and... I don't even know, get the developer who put it in kicked out of the project or something. Importance -> Wishlist is... perhaps not sending the right message.

Scott Howard (showard314) wrote :

This .diff is simpler: if you don't have r-base, don't display anything. If you already have r-base, display the copyright for r-base, and everything else the developers wanted. I'm proposing a merge, which will directly alert the maintainers of r-base in Ubuntu.

Benjamin Drung (bdrung) wrote :

The current version of r-base in lucid is 2.10.0-1. Please check, if this version is still affected. In this case, please update the debdiff and resubscribe ubuntu-universe-sponsors.

Nathaniel Smith (njs) wrote :

According to this bug, lucid has (temporarily) dropped all revolution R support:

However, lucid still contains revolution R packages, so I guess either these packages are useless and broken or else the above bug is incorrect? Either way someone should probably file a bug, but I don't know what's actually going on well enough to do so properly.

hamish (hamish-b) wrote :

Fundamentally revolution R is treating CRAN-R like it is LGPL, but it is not LGPL.

This causes a major problem.

Treating CRAN-R as a library causes REvolution R to be a derivative work under GPL term # 0:

"[...] The "Program", below,
refers to any such program or work, and a "work based on the Program"
means either the Program or any derivative work under copyright law:
that is to say, a work containing the Program or a portion of it,
either verbatim or with modifications and/or translated into another
language. [...]"

key words: "based on", "containing the program" and "verbatim"

Nothing ambiguous about it.

A test is if their program can effectively run without the aid of the GPL program. If it can not, it is a derived work in the eyes of the license.

This can not be considered a conglomerate distribution under the OS exemption, which reads:

"However, as a
special exception, the source code distributed need not include
anything that is normally distributed (in either source or binary
form) with the major components (compiler, kernel, and so on) of the
operating system on which the executable runs, unless that component
itself accompanies the executable."

certainly the r-base package should not be peddling REvolution's wares in the form of the startup message:
"REvolution R enhancements not installed. For improved performance and other extensions: apt-get install revolution-r"

Even if it were just a non-free package that would be inappropriate. but for a significantly darker-than-grey GPL violation it is even worse. Ubuntu (nor anyone else) cat not distribute this package without an upstream relicense of R (which ain't gunna happen) or REvolution Computing (ball's in their court).

Perhaps it is bug-creep to this bug report about the advertisement, but overall "wishlist" is inappropriate for the Importance setting.

sorry, but you have to respect the upstream authors license if you want to redistribute and that's really all there is to say.


Scott Howard (showard314) wrote :

Lucid does not have the advertisement:

showard@s-desktop:~$ dpkg -p r-base | grep 'Version'
Version: 2.10.1-2

showard@s-desktop:~$ R

R version 2.10.1 (2009-12-14)
Copyright (C) 2009 The R Foundation for Statistical Computing
ISBN 3-900051-07-0

R is free software and comes with ABSOLUTELY NO WARRANTY.
You are welcome to redistribute it under certain conditions.
Type 'license()' or 'licence()' for distribution details.

  Natural language support but running in an English locale

R is a collaborative project with many contributors.
Type 'contributors()' for more information and
'citation()' on how to cite R or R packages in publications.

Type 'demo()' for some demos, 'help()' for on-line help, or
'help.start()' for an HTML browser interface to help.
Type 'q()' to quit R.

To post a comment you must log in.
This report contains Public information  Edit
Everyone can see this information.

Other bug subscribers