[personal profile] mjg59
I bumped into Mark Shuttleworth today at Linuxcon and we had a brief conversation about Canonical's IP policy. The short summary:
  • Canonical assert that the act of compilation creates copyright over the binaries, and you may not redistribute those binaries unless (a) the license prevents Canonical from restricting redistribution (eg, the GPL), or (b) you follow the terms of their IP policy. This means that, no matter what Dustin's blogpost says, Canonical's position is that you must ask for permission before distributing any custom container images that contain Ubuntu binaries, even if you use no Ubuntu trademarks in the process. Doing so without their permission is an infringement of their copyright.
  • Canonical have no intention of clarifying their policy, because Canonical benefit from companies being legally uncertain as to whether they have permission to do something or not.
  • Mark justifies maintaining this uncertainty by drawing an analogy between it and the perceived uncertainties that exist around certain aspects of the GPL. I disagree with this analogy pretty strongly. One of the main reasons for the creation of GPLv3 was to deal with some more ambiguous aspects of GPLv2 (such as what actually happened after license termination and how patents interacted with the GPL). The FSF publish a large FAQ intended to provide further clarity. The major ambiguity is in what a derivative work actually is, which is something the FSF can't answer absolutely (that's going to be up to courts) but will give its opinion on when asked. The uncertainties in Canonical's IP policy aren't a result of a lack of legal clarity - they're a result of Canonical's refusal to answer questions.

The even shorter summary: Canonical won't clarify their IP policy because they believe they can make more money if they don't.

Why do I keep talking about this? Because Canonical are deliberately making it difficult to create derivative works, and that's one of the core tenets of the definition of free software. Their IP policy is fundamentally incompatible with our community norms, and that's something we should care about rather than ignoring.

Date: 2015-08-18 08:12 pm (UTC)
From: [personal profile] jewelfox
I personally feel like it's less of a problem that they aren't conforming, and more of a problem that they are being jerks about this. >_>

Canonical IP policy

Date: 2015-08-18 08:46 pm (UTC)
From: [identity profile] flosslinuxblog.blogspot.co.uk
This is beyond ironic - given that you were one of the people suggesting that the stealthworks project which became Warty Warthog (04:10) was, essentially, "Debian for desktops" and not much more and Mark was fine and could readily be trusted.

Mark is a decent bloke in many ways but ...how are the mighty fallen.

At least you've been honest and consistent, as you ever are. I've been assured that I'm mistaken in believing your take on this but through all the vicissitudes, ups, downs changes in employment you're still mjg59, Ph.D

Binaries are mechanically generated

Date: 2015-08-18 09:32 pm (UTC)
From: (Anonymous)
Are mechanically generated works such as the output of a compiler (excluding any handwritten parts) even subject to copyright?

Re: Binaries are mechanically generated

Date: 2015-08-18 11:19 pm (UTC)
From: (Anonymous)
Isn't there sufficient creative effort anywhere but in Ubuntu's development process that makes this moot? In particular, GCC developers, Debian developers who do some work that then is taken verbatim (not a single Ubuntu patch on it), etc.?

This is fucking ridiculous.

Re: Binaries are mechanically generated

Date: 2015-08-19 11:47 pm (UTC)
From: (Anonymous)
"act of compilation" as in running GCC over source code, or as in putting together a bunch of software to make it work as a whole?

The first probably doesn't get you any copyright (it's mechanical), but the second most likely does (it's creative), and if the final end result is binaries, the creator gets copyright over those.

Re: Binaries are mechanically generated

Date: 2015-12-26 11:01 pm (UTC)
From: [identity profile] tepples [launchpad.net]
More likely "act of compilation" refers to choosing the particular compiler flags for each source code file. These flags come from the makefiles and/or the environment variables passed to Make. And GNU standard practice for "small supporting files" is to put them under an all-permissive license (see "Information for Maintainers of GNU Software: License Notices for Other Files"). So it appears Canonical is asserting copyright on its CFLAGS and its changes to packages' makefiles.

Date: 2015-08-18 10:00 pm (UTC)
From: (Anonymous)
Dustin's blog post starts with "I am speaking for my employer, Canonical".
So when you say that Canonical refuses to answer questions, I think you are wrong regarding, at least, containers.
Canonical, through the voice of Dustin, explicitly said that building Ubuntu-based container is not a copyright violation.

But I think your concerns remain valid for other things (custom ISO for example).

Date: 2015-08-19 03:56 am (UTC)
From: (Anonymous)
Canonical, through the voice of Dustin, explicitly said that building Ubuntu-based container is not a copyright violation.

No, that is not what he said. He placed all sorts of restrictions on it. Here, let me quote one for you:

if you’re going to redistribute Ubuntu to third parties who are trusting both you and Ubuntu to get it right, come and talk to Canonical

If you want to distribute a container to a third party, you must ask Canonical for permission.

Ironic indeed...

Date: 2015-08-18 11:26 pm (UTC)
From: (Anonymous)
... as I distinctly remember Mark Shuttleworth criticizing Red Hat for closing down their binaries, and treating binary code differently than the free source code. Let me see - ah, here it is: http://www.markshuttleworth.com/archives/77

That was in 2007. Although the post is a bit confused on the issue, it appears to be saying that Canonical will offer their products free as in free speech *and* free as in free beer, both the source code *and* the binary code, to completely level the playing field for everyone, and build a base that everyone can build on and improve freely. The bit about source code vs binary code is particularly apparent from Mark's answers to Greg DeKoenigsberg and Spencer Johnson in the comments:

"Applications, as you know, don’t run on source RPM’s. Hardware vendors don’t certify source RPM’s. Users don’t install source RPM’s. Really – how many vendors (hardware or software) certify CentOS? As far as I’m aware, none of any consequence. You know that’s true. So let’s talk about the real meat – the binaries that make up RHEL. As you are well aware, these are a closely controlled and licensed under terms very similar to those of any traditional proprietary software."

"When you say “all of Red Hat’s bits are free”, you mean “all of Red Hat’s published source code is free”. But applications do not run on source code, they run on binaries, and Red Hat’s BINARY bits are not free, are they?"

Re: Ironic indeed...

Date: 2015-08-19 07:45 am (UTC)
From: (Anonymous)
No doubt there's a disconnect between Mark/Canonical's stance now vs 8 years ago. But I'm curious how Canonical's (current) stance on redistribution differs much than RedHat's here? AIUI, you can't even get access to RHEL's certified binaries/beer without paying for a license. Is it simply Canonical's ambiguity on the matter the source of mjg59 and others' frustration?

I don't necessarily agree with Canonical's shift but IMHO I'm not surprised. It seems RedHat's been operating under this model since the beginning, effectively prohibiting binary derivatives and forcing complete archive rebuilds like CentOS, but without similar criticism. What am I missing?

Re: Ironic indeed...

Date: 2015-08-19 12:58 pm (UTC)
From: (Anonymous)
Having to pay for RHEL services does not mean that the commercial binaries they distribute are (copyright) restricted. If there is a restriction, it centres around the use of trademarks (at least from a casual read of their licence terms).

Canonical on the other hand appear to be trying to restrict the use of their software by applying copyright law to their binaries. They explicitly say in their "intellectual property policy" that you can only redistribute Ubuntu where there has been no modification to it (then later that modified versions must not only remove trademarks, but must be recompiled).

Re: Ironic indeed...

Date: 2015-11-20 07:43 am (UTC)
From: (Anonymous)
Red Hat operate a subscription model. You can get the RH binaries now, e.g. "docker pull rhel7" works, afaics. What RH sell you is a subscription so "dnf update" functions.


Date: 2015-08-19 12:48 am (UTC)
From: (Anonymous)
We just need to get CloudInit working properly on Debian, then knock the Ubuntu dust off our boots, and walk away.


(And I was standing right there when you were chatting with Mark, and your summary of that conversation matches my own observation.

Date: 2015-08-19 01:47 am (UTC)
From: (Anonymous)
Did you ask him on the trademark issue? You claimed earlier that compilation of Ubuntu source packages possibly violated the Ubuntu trademark policy (because the word 'Ubuntu' is used in the build file). Did you receive clarification on this?

fart fart fart fart fart fart fart fart

Date: 2015-08-19 08:11 am (UTC)
From: (Anonymous)
fart fart fart fart fart fart fart fart fart fart fart fart fart fart fart fart fart fart fart fart fart fart fart fart

Re: fart fart fart fart fart fart fart fart

Date: 2015-08-19 04:40 pm (UTC)
From: (Anonymous)
Intelligent comment there.

Re: fart fart fart fart fart fart fart fart

Date: 2015-08-19 06:21 pm (UTC)
From: (Anonymous)
Mr. Garrett reserves the right to change particularly odious comments into repetitions of the word "fart". I can't remember when this started, but it has been going on for a long time now.

Re: fart fart fart fart fart fart fart fart

Date: 2015-10-07 06:51 pm (UTC)
From: (Anonymous)
Or just comments that he disagrees with. Who will know?

In doubt, derive from Debian!

Date: 2015-08-19 04:53 pm (UTC)
From: (Anonymous)
Debian is technically very similar to Ubuntu and makes it more than clear that derivatives are welcome, whether based on Debian binaries or self-compiled.

Date: 2015-08-20 03:51 pm (UTC)
brainwane: several colorful scribbles in the vague shape of a jellyfish (jellyfish)
From: [personal profile] brainwane
It's amazing how much my gall rises when I read:

Canonical have no intention of clarifying their policy, because Canonical benefit from companies being legally uncertain as to whether they have permission to do something or not.

That is so counter to our community's values (mutual aid, transparency, disambiguation) that I cannot fully articulate how skullbendingly wrong it is.


Matthew Garrett

About Matthew

Power management, mobile and firmware developer on Linux. Security developer at Google. Ex-biologist. @mjg59 on Twitter. Content here should not be interpreted as the opinion of my employer.

Expand Cut Tags

No cut tags