"Respects Your Freedom" certification?


On the contrary, I think the current situation potentially causes confusion - there's no easy way to check the Freeness status of what you're downloading, only an optional "license" field that doesn't always have an informative value.
I do support promoting free software and agree with your first post, but in practice once people added the 'non-free' repo (what everybody will do anyway) it is all one big repository again. 
 
Well this is an Interesting debate. :)

In terms of hardware I'd say that if making P2 with 'free' hardware does not make it too expensive / under powered / difficult to reliably source components etc. then this is an entirely worthy goal. One that is highly worthy of serious exploration from the hardware team.

However in terms of software then I strongly disagree. Sure have a toggle on the repo website / PND Manager so you can only see / browse free software, if that is what floats your boat, but denying people the right to use / upload software that is not free strikes me as a bit like saying 'I agree with free speech but only as long as you say the sort of thing I'm comfortable with'.

Also, in terms of getting official certification is it really so draconian that any links on the official hardware site to non free software would invalidate the certification even if the hardware itself was free? If so, then screw certification, but try and make the hardware free anyway.
 
Any 'Freedom Certification' that has a gag order restricting free speech as a pre-requisite for qualification is, IMHO, absolutely not a direction to go in.

I have no qualms in compensating people reasonably for their efforts if they're providing and supporting a product I'm using - whether that is hardware, software or media.

I'm not a big fan of DRM restrictions either.  Ensuring that we have open/free license to make full use of the hardware we're paying for - that is a good thing.  This business of proprietary drivers for purchased hardware is for the birds.  The hardware maker is compensated by selling the hardware.  Access to drivers and hardware specs to make drivers shouldn't be a question.

When it comes to software, if someone spends their valuable time to create something and chooses to charge for it - that is their option.  Yes, someone could write a driver for the hardware and charge for it - but the specs on the hardware need to be available to do so.

The big thing that makes the FSF a joke is their willingness to throttle free speech (regarding paid software) in a militant manner.  It's like, "Anarchists for Greater Order" and such.  They're their own joke.
 
For the certification, I think it suffices to make sure that the firmware that comes pre-installed is Free Software, as well as any software that is not pre-installed but somehow "recommended" by the hardware vendor. There is no problem to have the ability to install proprietary software. No "gag order".
 
Last edited by a moderator:
There isn't even a requirement that free and non-free software needs to be separated, only product software needs to be free. Product software includes "all software that the seller includes in the product, or provides with the product, or recommends for use in conjunction with the product, or steers users towards installation in the product." Applications pre-installed, recommended for use, or steers towards installing are all that the requirement includes.

They then go on to say "a general-purpose facility for installing other programs, with which the choice of programs to install comes directly from the user, is not considered to steer users toward anything in particular. However, the product suggests using the facility to install certain programs, that is steering users towards those programs."

PNDStore (and PNDManager, and the repo in general) are simply general-purpose facilities for installing other programs, the choice of what is entirely up to the user. OPT makes no suggestions as to what needs to be installed, they are not steering the user towards anything, it's entirely user driven. So long as OPT doesn't label DraStic as a suggested application they should be fine.
 
For the certification, I think it suffices to make sure that the firmware that comes pre-installed is Free Software, as well as any software that is not pre-installed but somehow "recommended" by the hardware vendor. There is no problem to have the ability to install proprietary software. No "gag order".
If it requires references to paid software to be removed from the manufacturer's web site, that is a gag order.

if it requires paid software to be removed from the manufacturer's application store, that is a gag order.

Freedom includes speech, commerce, travel and thought.  Impinging on any of those and claiming to promote freedom is simple hipocrisy.

How do you interpret it otherwise?
 
As far as I'm concerned, what should be important is if the hardware can be used entirely freely, and if there's proper support infrastructure for using it this way. So that means all vital driver support should be open and hardware documentation available, and there should be distros, repo options, and generally clear and easy means of getting a totally free system. Note that I do think you can still have closed blobs for things like GPU as optional extras, because in mobile systems GPU drivers are not required to get basic functionality out of the device and often they're pretty usable without it (Pandora certainly is, minus the programs you don't get to use)

I find the idea of requiring that the hardware vendor avoids any recommendation of anything proprietary to be going too far. It's too ambiguous what could be construed as recommendation. And I think this goes outside of the pragmatic needs of the user and focuses too much on furthering an ideological agenda by demanding social awareness is promoted. While I'm sure some will disagree this feels too much like getting permission to stock someone's products only if you don't stock their competitor. I think that the platform's capabilities should be judged and not what the developers think or feel.
 
Last edited by a moderator:
A gag order is something someone or some entity imposes on someone else in court. The FSF is not doing that afaik.

Here we are talking about a voluntary self-imposed censorship to not ship or advertize non-free software for use with the hardware, which the FSF "rewards" by endorsing the hardware. That is something entirely different.

It's a bit like an anti-slavery activist who declares "I don't have any slaves and I don't want to have any either, because I would prefer everyone to be free". The pro-slavery reply (not that I want to associate Grench to that position ;) , just trying to make a point here) would in that case be: "You're a hypocrite and you're impinging on my freedom, because you want to take away my freedom to have slaves and my slaves' freedom to be slaves."
 
Last edited by a moderator:
As for what is a "recommendation": there obviously is a grey zone, but the examples given by the FSF shed some light on what is meant.

A sticker on the device or packaging saying "Works with Windows" would not be allowed, a note on the website saying that it is able to run Windows (amongst others) would be OK.

I don't think it would even be a problem to refer to the possibility of using non-free software like DraStic on the official website. As long as it's clear that it is up to you to install or not install it.

So if the Pandora successor gets advertized as a general-purpose computer that can do many things, including running free and non-free emulators, that is fine. If however it gets advertized as a Nintendo DS emulator, and the only way to get the advertized functionality out of the device is by installing non-free software, then the situation would be different and the FSF would not certify it.
 
Last edited by a moderator:
Hmm. All of my software would be removed then, I suppose? Actually, no - xmountains would be ok.

PandaBAS, even though GPL'd, contains libbass.so, and source is not available for that. Spheres of Chaos would be a no-go, the author doesn't want source distributing. Ditto for Forget-me-not.

Not in favour here, and I don't see the benefit for a community of fanatics that numbers at best 15 people globally.

D.
 
Not in favour here, and I don't see the benefit for a community of fanatics that numbers at best 15 people globally.
There is a trend to distrust closed source software and big companies, certainly in Europe. Stories about security, PRISM, NSA etc is undermining the faith of costumers in software in general. If they are right or wrong is not important, I am talking about perception.   

A certificate from this 'group of fanatics' might be the best PR possible when P2 is launched. 

You can compare it with biological food. 10 years ago a few fanatics and now you can find it everywhere. 

 

If certification is possible I would definitely try to obtain it, even if it is means sacrificing some comfort or performance. 

 

Edit: typo
 
Last edited by a moderator:
Hmm. All of my software would be removed then, I suppose?
No, absolutely nothing would need to be removed. The repo is perfectly fine as it is. The requirement is simply that OPT doesn't require or recommend any closed software, having a general purpose application source (which may contain either open or closed or both) is entirely acceptable.As _wb_ says, if they ran advertisements that stated it could emulate a DS but requires a closed source application in order to do so, that would be violating their restriction. They can advertise it as being able to run Java: even though the SUN java is closed source there is an open source alternative, so no problem there. If they advertised it as being able to run PandaBAS then they may have problems.

The fact that a user can, at their own discretion, install anything they want, including closed source applications, from a central repository has no bearing on certification.
 
As for what is a "recommendation": there obviously is a grey zone, but the examples given by the FSF shed some light on what is meant.

A sticker on the device or packaging saying "Works with Windows" would not be allowed, a note on the website saying that it is able to run Windows (amongst others) would be OK.

I don't think it would even be a problem to refer to the possibility of using non-free software like DraStic on the official website. As long as it's clear that it is up to you to install or not install it.

So if the Pandora successor gets advertized as a general-purpose computer that can do many things, including running free and non-free emulators, that is fine. If however it gets advertized as a Nintendo DS emulator, and the only way to get the advertized functionality out of the device is by installing non-free software, then the situation would be different and the FSF would not certify it.
But wouldn't that be hipocritical ? In one hand I am holding the certificate and in the other I have a officially supported software repository with half of the software in it beeing not "spiritually" compatible with that certificate. With the certificate probably beeing interesting only to a certain group of people, they may come to the same conclusion, or not ?
 
Then here's a question - is it okay to advertise that it can run emulator X, which is under a free software license, by showing it play game Y, which is proprietary?
 
It's OK to advertise that it can run non-free software or free software with non-free data or free software emulating non-free games or non-free software emulating non-free games or whatever. It's not OK to advertise that it is somehow better to run non-free software on it than free software, to require non-free firmware to be able to run any software at all, or to ship non-free software with it. As long as it is possible to fully use the device (i.e. all of the hardware that is on the tech specs, including wifi and GPU) using 100% Free Software, and the "default" that gets shipped with newly sold units is 100% Free, it is OK.

So no, ZXDunny, none of your software would have to be removed, because I assume that none of your software would be shipped with new units.
 
But wouldn't that be hipocritical ? In one hand I am holding the certificate and in the other I have a officially supported software repository with half of the software in it beeing not "spiritually" compatible with that certificate. With the certificate probably beeing interesting only to a certain group of people, they may come to the same conclusion, or not ?
The importance of the certification is that the device can function as advertised using entirely "free" software. The fact that it can also run non-free software actually makes it even more "free". The hypocrisy would be in saying that it is not allowed to run non-free software as that would be a restriction on the rights of the users. As long as it doesn't NEED non-free software in order to function as intended (with some exceptions, even) then it passes their certification.
 
But wouldn't that be hipocritical ? In one hand I am holding the certificate and in the other I have a officially supported software repository with half of the software in it beeing not "spiritually" compatible with that certificate. With the certificate probably beeing interesting only to a certain group of people, they may come to the same conclusion, or not ?
The importance of the certification is that the device can function as advertised using entirely "free" software. The fact that it can also run non-free software actually makes it even more "free". The hypocrisy would be in saying that it is not allowed to run non-free software as that would be a restriction on the rights of the users. As long as it doesn't NEED non-free software in order to function as intended (with some exceptions, even) then it passes their certification.
exactly
 
From the above, the software side of things sounds fine.

My questions now would be :

What hardware could be used that would make certification possible? Edit : i.e. has suitable open drivers.

How much of a limit would it put on manufacturing a successful device? (in terms of cost, availability, difficulty of sourcing components)

Are there any current phones / handhelds / computers (not chips or dev boards) that meet the necessary standard to achieve certification?

If the answer to the last question is no, then as mentioned by b_o_b there is the potential for pretty decent media coverage (in the IT world) with P2 as the only available free handheld / computer, as well as the all round good karma that such a device would bring. 
 
Last edited by a moderator:
isn't proprietary software (drivers) also "free software"? I think the important part for a base system is to be open source. So everybody can look at the code, validateing it, improving it etc.

let's just try to make it 100% FOSS (if possible) and put a 100% free and open-source software sticker on the box and advertisement photos - that should get the message across!

I very much like _wb_'s Lima driver idea!!! ; )

Edit: changed open source to FOSS/free and open-source
 
Last edited by a moderator:
Well, at the moment no full computers are listed on the FSF site, which doesn't mean that nothing exists yet that would qualify, but at least it means we have a good chance to be the first to get certified.

I suspect that most big hardware companies cannot get certified because they want or need to put Windows stickers on their units and ship them with Windows. The manufacturers of Android devices probably don't want to ship a fully Free version of Android and want to put in their proprietary stuff too. Even the Raspberry Pi relies on binary blobs and recommends/ships (a.o.) the proprietary RISC OS.

To answer your other questions:

Possible SoCs that could probably be made to work with 100% Free Software using existing reverse-engineered FOSS drivers:  Exynos 4412, 5220, 5420 (they use Mali GPUs, for which RE is well under way or nearly finished), Snapdragon 600 or 800 (uses Adreno), Rockchip RK3188 (Mali), HiSilicon K3V3 (also Mali).

Besides the GPU (which of course does restrict the choice of SoC), I don't there are other components which would cause trouble. I also don't think there would be any effect on cost or availability, except maybe if the vendor of the SoC demands that the device gets shipped with Android or Windows 8 or something, but that would be a problem for most of us anyway.
 
Back
Top