beagleboardblack fully free software?

41 risposte [Ultimo contenuto]
tonlee
Offline
Iscritto: 09/08/2014

I had an freedombox irc chat. Freedombox claimed, that the beagleboard black is all free software and that it could get fsf approval.
Is that correct?

Pyraman
Offline
Iscritto: 06/05/2014

Shortened quote from FSF website:
"The BeagleBoard (various versions) uses the TI OMAP family of SoCs... The graphics accelerator (GPU) and the video decoding hardware for formats such as MPEG-2 are nonfunctional, because they require nonfree blobs to be installed into them. The workaround for these flaws is to do these jobs on the CPU with free software."
So, maybe a BeagleBoard Black could potentially get a FSF approval, e.g. if these jobs will be transferred to CPU by default, but this could greatly decrease a performance of this already-not-so-fast board

suitsmeveryfine
Offline
Iscritto: 08/15/2014

I might add that the BBB normally comes with Debian GNU/Linux and with non-free repositories enabled by default (for no apparent reason).

tonlee
Offline
Iscritto: 09/08/2014

If freedombox is on it it does not use the gpu? Then you can have a freedombox and only free software?
What is performance compared to raspberry pi 2? Freedombox spoke about a 64bit version around the corner.

lloydsmart

I am a member!

Offline
Iscritto: 12/22/2012

You can use freedombox on it without using the GPU. Because of this, it is possible to use this device in 100% freedom.

The hardware itself, however, is unlikely to get FSF approval, because of the non-free GPU. Also, the fact that it comes with the non-free Debian repos enabled by default, which is bad.

I'd like them to make a GPU-less version of this, for applications like freedombox and other headless servers. Alternatively, they could release GPU drivers as free software. That would be the best option.

GNUser
Offline
Iscritto: 07/17/2013

There is not single computer in the entire world that works 100% with free software only. Even the Thinkpad that the FSF awards, those are full of proprietary code running inside. The fact that the FSF considers those "unimportant" means little. The fact is, we have no 100% free machines at this point in time. We need to go with what we have and do the best we can.
These boards can actually use less proprietary code than other machines. Allwinner boards for example are usable with free bootloader, and have experimental drivers for GPU. So, yeah, maybe we should accept these as being "ok for the time being". For me these boards respect my freedom more than the laptop I have in front of me.

suitsmeveryfine
Offline
Iscritto: 08/15/2014

I agree with you that we should choose machines that can run in freedom. Then we should also not recommend anyone to install non-free stuff to make the GPU or something else work better afterwards. I have both a BBB and an Allwinner A20 myself and I'm very happy with them (since I run them in complete freedom). There is a lot you can do with a computer without fancy graphics. Depending on the use case, machines like these are perfectly acceptable I'd say.

Regarding the distros that can be installed, I believe this issue will be solved very soon. Besides, getting Das U-Boot + Debian by default is much better than EFI + Windows or OSX.

GNUser
Offline
Iscritto: 07/17/2013

I already replied to you (see below) but another question: which image are you running in your AllWinne? Where did you get a libre kernel?

suitsmeveryfine
Offline
Iscritto: 08/15/2014

I'm currently using Debian 8 (with only free software) installed on a fast microSD card. Please look here for instructions. Debian's "main" kernel is fortunately only free software.

Next, however, I'll try to install Parabola by first installing Arch and then migrate to the former by following these instructions. Hopefully it won't take too long until the Parabola team can provide native installation instructions, but they sure can need some help.

GNUser
Offline
Iscritto: 07/17/2013

Oh, I see. You installed it yourself. I thought you were using a pre-built image. There are so many available I wish I could get one to work in freedom.
Actually, bananian (www.bananian.org) is the closest thing, Nico actually was kind enough to let me know which non-free packages need to be removed and other than that, he said one can use it with the main repo only and it all works well. I might give that one a shot.

I have tried making an image of mine, some time ago, but it didn't go all that well. Might try again at another time though :)

ADFENO
Offline
Iscritto: 12/31/2012

As a side note (and maybe even off-topic), last time I checked,
GNU+Linux-libre GuixSD had ARMHF builds for some packages, except for
GNU Linux-libre[1].

REFERENCES

[1] https://www.gnu.org/software/guix/package-list.html

SuperTramp83

I am a translator!

Offline
Iscritto: 10/31/2014

The thinkpad that the FSF awards is not "full of proprietary code" - it has two proprietary firmwares, one for the hard drive and one for the embedded controller.
Btw it wasn't me who downvoted your post. I prefer to use words to dissent and not this stupid and completely useless sissy voting system.
cheers

suitsmeveryfine
Offline
Iscritto: 08/15/2014

I just upvoted to brick his post back to neutral.

SuperTramp83

I am a translator!

Offline
Iscritto: 10/31/2014

I just upvoted because you upvoted. Your turn again.

Jodiendo
Offline
Iscritto: 01/09/2013

Super tramposo

I don't care if you down vote me! But rest assure I'm still your sissy and dissident guy..
Debian rocks!
Debian rocks!
and it will always rock you!

For those about to Rock with Debian! We salute you!!

GNUser
Offline
Iscritto: 07/17/2013

First thanks suitsmeveryfine for the upvote. Wish I could see who upvotes and downvotes.

Second, maybe I exaggerated when I said "full of blobs". But it's true that the FSF considers those as little importance (hence they awarded the RYF) when in reality those two firmwares alone can render full disk encryption useless (totally possible that your password is recorded somewhere inside the device, ready to be read by an adversary).

Third, I would like to ask for your attention/help here:
http://forum.armbian.com/index.php/topic/422-free-as-in-freedom-image/
As you can see, I am trying to convince Igor to release an image using only Free/Libre software. He asked for help with the kernel, a request that I consider appropriate. I asked jxself for help, don't know if he will actually do it or not, I know it's a lot of work but hope he realises the importance of this. Anyway, I was hoping some of you who do use / wish to use these boards in freedom, could maybe express that desire in the forum too, maybe help with the cleaning, or if someone can provide the kernel all the better.
I am fine with cleaning the images myself, but I don't know enough about the kernel to do it myself. My earlier attempts actually didn't go all that well.

SuperTramp83

I am a translator!

Offline
Iscritto: 10/31/2014

>when in reality those two firmwares alone can render full disk encryption useless (totally possible that your password is recorded somewhere inside the device, ready to be read by an adversary).

any link/proof? I heard that too or read it somewhere, but I'd like a link pointing to a website or video where someone actually proves that he/she has successfully managed to do so.

GNUser
Offline
Iscritto: 07/17/2013

The problem is not that they have done it already, the problem is that they COULD do it if they wanted.
I don't like sleeping with that uncertainty.
If you know what I mean.

As for the basis of why I think this way:

http://hemingway.softwarelivre.org/fisl16/high/40t/sala_40t-high-201507111600.ogv (best reference I have seen in this matter, clearly illustrates the idea)
Several talks by Jacob Appelbaum, he mentions the dangers of any piece of proprietary code running inside your machine.
I think I saw a Replicant video some months ago that also mentioned the dangers of any firmware running inside your machine. Can't find it now though.

Look, my laptop doesn't even support Coreboot, so I am not saying that I am in a better position. I am just saying that I would like to be in a better position! And I don't feel that the "RYF Award" actually puts me in that better position. Brings me close, but it is not enough.

SuperTramp83

I am a translator!

Offline
Iscritto: 10/31/2014

>http://hemingway.softwarelivre.org/fisl16/high/40t/sala_40t-high-201507111600.ogv

Hooly cow! I watched this today and I have to say: scary scary! Tx for the link Gnuser. If you have more like this post mate.

ADFENO
Offline
Iscritto: 12/31/2012

Wow! And to think that everyone started to link to videos of FISL 16
from Hemingway.SoftwareLivre.org after I gave hints of the server's
existence in a "professional" photography forum topic[1] (I used quotes
because I'm no professional, and besides, I don't even know how to set
bleedings and margins correctly, let alone measure paper weight, that
is, grams per square meter of paper ream). :D

Nice to see some interested people on free software events around the
world (although not all talks from FISL 16 are in favor of free
software, but instead, in favor of open source software, but anyway, my
point still holds somehow).

REFERENCES

[1]
https://trisquel.info/gl/forum/professional-photography-tools-free-software#comment-79981

GNUser
Offline
Iscritto: 07/17/2013

So, am I absolved from the whole "full of blobs" thing? Now you know what I meant and why I disagree with the RYF. I think "FSF best choice" is the correct definition. FSF should be humble enough to change.

SuperTramp83

I am a translator!

Offline
Iscritto: 10/31/2014

Well, the EC firmware being able to capture your keystrokes and the hard drive firmware potentially containing a full OS on its own are scary things and it may very well be that all PC implement those malicious features, so yeah, until we have a computer that is completely libre we can not assume that it respects our freedom so probably the RYF may as well be changed into RYFM (respects your freedom **maybe**), for we indeed don't know if it really respects your freedom..
Post more links like that one if you find some.

P.S: In that vid Garrett mentions Nouveau and the Nvidia GPUs as being able to run without any proprietary firmware but he says nothing about Intel GPUs. Is that because Intel GPU have proprietary firmware embedded in the chip just like most Atheros wifi chips that work with free software have proprietary firmware embedded in the rom so they work with free software but still their firmware is proprietary? Is this the case with Intel GPUs as well? It is not very clear to me.

suitsmeveryfine
Offline
Iscritto: 08/15/2014

As I answered above, the Parabola team is almost there it seems. Regarding kernels specifically, they've got both the mainline linux-libre and linux-libre with grsecurity that I'd like to try soon. I'll check out the forum and try to post something there when I find the time to do so, but again, no blobs are required if you install the OS onto a microSD card (for booting at least), something that works very nicely. If you wish, you can then have the OS and/or media files on a separate hard drive (either via SATA or USB interface, the latter requiring a powered USB hub).

jxself
Offline
Iscritto: 09/13/2010

"But it's true that the FSF considers those as little importance"

This is not true at all, sorry. It does represent a compromise, but that doesn't mean they aren't important. One does not mean the other. Being a temporary compromise only is covered in the latest article too: "We decided to ignore these programs in our certification criteria today, because otherwise no computer could comply, and because firmware that is not normally changed is ethically equivalent to circuits. So our certification criteria cover only the code that runs on the computer's main processor and is not in true read-only memory. When and as free software becomes possible for other levels of processing, we will require futer would could comply with said criteria, they seemed to have said something along the lines that "it is better to show the world a computer that is at least going in the direction we want to go in -- and encourage future work and development in that direction -- rather than first wait and hope for a perfect computer to arrive first -- and only then show the world the direction we should be moving in."ree software at those levels too."

So it is a temporary compromise only, and should not be interpreted to mean "unimportant."

Since no computer would could comply with said criteria, they seemed to have said something along the lines that "it is better to show the world a computer that is at least going in the direction we want to go in -- and encourage future work and development in that direction -- rather than show nothing at all and wait and hope for the perfect computer to somehow (maybe, possibly) arrive on its own."

The notion of the machines that currently have the RYF certification no longer being RYFed in the future is also covered in their criteria: "We want users to be able to upgrade and control the software at as many levels as possible. If and when free software becomes available for use on a certain secondary processor, we will expect certified products to adopt it within a reasonable period of time. This can be done in the next model of the product, if there is a new model within a reasonable period of time. IF THIS IS NOT DONE, WE WILL EVENTUALLY WITHDRAW THE CERTIFICATION." (Emphasis added.) So the RYF mark will always represent the best we have. As we have advances in freedom, devices that can't keep up will be dropped from the program so it's very possible that in the future people's X60 and X200 will no longer bear that mark because they are superseded. Please help the free world to get there instead of complaining about how we aren't already there because we don't have the perfect computer to show off. :)

ADFENO
Offline
Iscritto: 12/31/2012

As an addendum: Software that runs on read-only memory is ACCEPTABLE,
but not free software. It's acceptable because there's no way to
exercise the results of freedom 1 (to study and to adapt the software as
you wish) when using freedom 0 (to run the software and the adaptations).

In other words, what is freedom 1 good for in this case if the memory in
which the software is on is read-only?

"Ah! But I can study and adapt it! I can redistribute it and my
adaptations! So it's free software!". But can you use YOUR ADAPTATIONS
on the very same computer (freedom 0)?

quantumgravity
Offline
Iscritto: 04/22/2013

So basically if a manufacturer locks down the bios (make it 'read only'), the freedom issue would be solved?
Quite paradoxical.

ADFENO
Offline
Iscritto: 12/31/2012

Indeed, you're right to consider it to be paradoxical, and this is one
of the reasons that this justification is not really an end goal, that
is: This is why the Respects Your Freedom certification will probably
change when the time comes where there's a possibility to run free, and
writable, software on similar hardware. Besides, I would actually edit
my previous publication as to replace "acceptable" with "passable for
the time being (must check later on)", since this situation doesn't
qualify as a good use of non-free software for free software advancement[1].

REFERENCES

[1] https://www.gnu.org/philosophy/is-ever-good-use-nonfree-program.en.html

ADFENO
Offline
Iscritto: 12/31/2012

Unless, of course, you manage to extract the software from the read-only
memory and engage on the development of a free replacement, despite
knowing that you cannot test it without losing an entire electronic
board. :D

GNUser
Offline
Iscritto: 07/17/2013

Lol, didn't even think of that but... yeah, damn right you are!

GNUser
Offline
Iscritto: 07/17/2013

Well, I don't see that as acceptable. Or to rephrase that: I don't think it's acceptable enough.
I don't fully understand the need to award a "RYF", is it such a priority, to stamp a logo "RYF" in a product? On the other hand I consider unfair to remove it afterwards. The X200 is now a good product that has the RYF. If a 100% Libre computer came along (with no blobs at all) the X200 would still be a good product. Should it be "removed" when it never changed (had no proprietary bits added to it), just got "second place" because some other did better? I don't think that is the best way to deal with things.

Please note "The X200 is a good product". I am not against the work done by Libreboot. Actually I think it's one of the best projects we have right now in terms of freedom! What I consider misleading is the whole "RYF" thing. I think the FSF could totally recommend those computers while mentioning that they are not 100% libre but are the best choice for now, and still not need an "award". I think that is in some ways... smoke and mirrors if you will. It's like saying we have something that respects your freedom, but it's not perfect, until something better comes along we will say it does then we discard it. I know they are not being "rude" to these projects, but it's kinda the message I see in there.

What should be the difference between X200, Chromebook C201, Macbook 2,1? I know they have some technical differences (I for instance would hardly trust any hardware coming from Apple) but they all have LibreBoot support. And none of them come from a "nice to freedom" maker. Again, I think the FSF is right in recommending some machines, but I don't think the "RYF" will do any good in the long run.

Magic Banana

I am a member!

I am a translator!

Offline
Iscritto: 07/24/2010

It's like saying we have something that respects your freedom, but it's not perfect, until something better comes along we will say it does then we discard it. I know they are not being "rude" to these projects, but it's kinda the message I see in there.

It is not. The message is "do the best you can". Companies like MiniFree and ThinkPenguin love the RYF label. It is great advertisement. Today the best we can in terms of laptop is not perfect. Whenever a new piece o firmware can be freed (including by people selling RYF hardware and who want to supersede their competitors), every vendor will have to follow this freedom path to not lose the certification. In the end, the RYF program is a great incentive for increasingly freedom-respecting hardware.

GNUser
Offline
Iscritto: 07/17/2013

"It is not. The message is "do the best you can". Companies like MiniFree and ThinkPenguin love the RYF label. It is great advertisement. Today the best we can in terms of laptop is not perfect. Whenever a new piece o firmware can be freed (including by people selling RYF hardware and who want to supersede their competitors), every vendor will have to follow this freedom path to not lose the certification. In the end, the RYF program is a great incentive for increasingly freedom-respecting hardware."

I think you are being too optimistic there. Free Software advocates are not the majority. So sales will never force anyone to do that. Apple is known (and has been, for years now) for a lot of issues (making their boards on sweat-shops at china, insecure services that get hacked all the time, too hard to install any software in a non-jb iphone) and yet they sell like hell. Things are never gonna be like that.

But maybe I am wrong, who knows :) Would be nice if I was wrong.

Magic Banana

I am a member!

I am a translator!

Offline
Iscritto: 07/24/2010

So we should not do anything hardware-wise until we can change Apple's politics? Am I the one being too optimistic here?!

quantumgravity
Offline
Iscritto: 04/22/2013

"It is not. The message is "do the best you can". "

No, it's not. The message is: "This product respects your freedom".
And I think GNUser is rightfully criticizing it.
If there is such a need for an award (for advertisement etc.) why not call it the "FSF Best choice" award or something like that.

A product that "respects your freedom" now can't suddenly stop doing so without changing at all; just because a new "more free" product is available.

The RYF award is misleading and should be changed into something more honest.

Magic Banana

I am a member!

I am a translator!

Offline
Iscritto: 07/24/2010

Read the first paragraph on http://fsf.org/ryf (emphasis is mine):
The "Respects Your Freedom" computer hardware product certification program encourages the creation and sale of hardware that will do *as much as possible* to respect your freedom and your privacy, and will ensure that you have control over your device.

That is pretty clear/honest to me.

GNUser
Offline
Iscritto: 07/17/2013

Yes, but makes little sense to take that away when a better product comes along.
X200 won't be less free when that happens, so if it deserved the RYF why take it away? truth is, and you can see the link I posted above, these computers don't really respect our freedom, they are a step in the right direction. One i would like to take myself. However quantumgravitywas dead right: FSF BEST CHOICE instead of RYF would make much more sense.

SuperTramp83

I am a translator!

Offline
Iscritto: 10/31/2014

>it is better to show the world a computer that is at least going in the direction we want to go in -- and encourage future work and development in that direction -- rather than show nothing at all and wait and hope for the perfect computer to somehow (maybe, possibly) arrive on its own.

Indeed I think that is a good way and those laptops deserve the RYF approval. But I also think they should make it absolutely clear which proprietary firmware those laptop ship.

lembas
Offline
Iscritto: 05/13/2010

> Second, maybe I exaggerated when I said "full of blobs". But it's true that the FSF considers those as little importance (hence they awarded the RYF) when in reality those two firmwares alone can render full disk encryption useless (totally possible that your password is recorded somewhere inside the device, ready to be read by an adversary).

Please don't say stupid things like that. What do you suggest they do, treat all hardware as equal and make no suggestions? That would be counterproductive.

GNUser
Offline
Iscritto: 07/17/2013

I am not the one who downvote your post (who has been messing with the votes in this thread anyway??), but I don't think your reply was fair to me.
Did I suggest that they should treat everything as equal? Not really.
That is not to say that caution should be taken. I think (and you can read it above in another comment of mine) that they can and should make suggestions, but the "RYF Award" is not necessary here, nor is it being well applied.
On another hand, if a suggestion is possibly misleading, I would rather not have it, though I believe they are totally well intentioned in their suggestions :)

btw, I will upvote your post and put it back to 0 because like SuperTramp said, we should use words and not "votes" to discuss these things. Whoever downvoted your post shouldn't have.

lembas
Offline
Iscritto: 05/13/2010

Thanks for the clarification. Sorry for being rude.

(Nevermind the votes.)

Jodiendo
Offline
Iscritto: 01/09/2013

I don't care if you down vote me!
But rest assure I'm still your sissy and dissident guy..
Debian rocks!
Debian rocks!
and it will always rock you!

For those about to Rock with Debian! We salute you!!

SuperTramp83

I am a translator!

Offline
Iscritto: 10/31/2014

viejito: Debian rocks and you are a viejito loco. :)