God Ginrai
Godmaster
Wasn't it pointed out earlier in this topic that this is the exact reason why Debian doesn't have this certification?
-God Ginrai
"Respects Your Freedom" certification?
- Thread starter _wb_
- Start date
-God Ginrai
sebt3
homebrew player (P. & C.)
It's not the same certification. RYF certification is solely about hardware, while the certification the debian project still fail to reach is about software distribution (what would be the point of an hardware certificate for a project like debian ?).
God Ginrai
Godmaster
From _wb_'s initial post, I was under the impression that for this hardware certification that you still needed to meet these software standards.
-God Ginrai
Last edited by a moderator:
Debian does not satisfy the requirements for an FSF endorsement as "100% Free GNU/Linux Distro", but that is a different set of requirements than those for endorsement as FSF-certified "Respects Your Freedom" hardware.
God Ginrai
Godmaster
Well, if the repo doesn't have to change, then I am more open to the possibility of this certification. However, I'm not sure we will have much of a choice, honestly. EvilDragon has pointed out that it is very hard to find a good SoC in the quantities needed for the Pandora, which would make finding one with FOSS drivers even harder. :\
-God Ginrai
comradekingu
Glowing ember
God Ginrai:
There is hardly any SoC that doesnt have foss GPU drivers in development. The GWC-zero has free drivers shipping for their vivante stuff. They would have a huge bonus drive for adding a general purpose usable screen and free wifi.
How is the openpandora design proprietary to anything but making openpandoras? Do you really think someone will come along and make a knockoff from a community project out of china?
How many patents do we own again? Does china care at all? Are they not able to reverse engineer things anyway?
Changing the repo is a software change. Respecting peoples freedom and having a unison way of separating it, providing source packages make sense. It is benefit for the purpose of benefit. Explain to me how the current one is better before you side with not changing it.
Free drivers is also a software change, to the point of ability. And thats something thats important to have in the back of ones head when selecting drivers. If you do it wrong, you are shooting yourself in the foot down the road. It makes less sense to go with proprietary stuff, since it gives more freedom to have choices.
There is hardly any SoC that doesnt have foss GPU drivers in development. The GWC-zero has free drivers shipping for their vivante stuff. They would have a huge bonus drive for adding a general purpose usable screen and free wifi.
How is the openpandora design proprietary to anything but making openpandoras? Do you really think someone will come along and make a knockoff from a community project out of china?
How many patents do we own again? Does china care at all? Are they not able to reverse engineer things anyway?
Changing the repo is a software change. Respecting peoples freedom and having a unison way of separating it, providing source packages make sense. It is benefit for the purpose of benefit. Explain to me how the current one is better before you side with not changing it.
Free drivers is also a software change, to the point of ability. And thats something thats important to have in the back of ones head when selecting drivers. If you do it wrong, you are shooting yourself in the foot down the road. It makes less sense to go with proprietary stuff, since it gives more freedom to have choices.
Last edited by a moderator:
Grench
Forum Addict!
- Joined
- Oct 3, 2008
- Messages
- 6,629
Please name the SoCs that have FOSS drivers that work just as well as their non-FOSS counterpart.
The only modern SoC that I'm aware of that has a decent GPU built in that will likely have real FOSS driver support is the Z3770. What did you have in mind? It would be rather ironic if the most 'Free' SoC was from Intel, wouldn't it?
For everyone who isn't in the know - this "Changing the repo is a software change" should be written as RESTRICTING the repo to ban certain types of software in the name of 'Freedom'. Real freedom is respecting other's rights to do/use things that you don't approve of.
So we get 'more choice' by submitting to a quasi-religious mandate that restricts potential options to a small set of options that are acceptable to a 3rd party's dogma. Ummm. No.
Exophase
Nothing good will ever come of Exophase.
The author of the Lima driver (Mali-400) replacement says that the performance is better than it is in the official ones. This was just from testing one game, though. And I don't know if it's as compatible. More to the point, it doesn't support Mali-T6xx which is the current generation GPU from ARM. This is a problem you're going to find everywhere, the reverse engineer projects are always going to lack the hardware if a major generational change is made. Freedreno definitely looks like it's behind performance-wise, and I can't find any info on Grate (GeForce ULV) at all. Etnaviv seems less interesting to me all around just because Vivante's not in any awfully good SoCs right now. The details you can read in the blog are pretty cool though, much more thorough than what I'm reading for any of the other projects. Props for all the hard work.
But I guess there's not really much point talking about how well along the RE efforts for some GPU are if an SoC using that GPU doesn't first present itself as a viable or good choice. Adreno is probably irrelevant.
But I guess there's not really much point talking about how well along the RE efforts for some GPU are if an SoC using that GPU doesn't first present itself as a viable or good choice. Adreno is probably irrelevant.
Last edited by a moderator:
WizardStan
Mega GP Mania
- Joined
- May 24, 2008
- Messages
- 16,731
For the umpteenth time, this is not true. There is no restriction on what can be hosted in a repository whatsoever to be eligible for the RYF certification.The restrictions in software are in what OPT can ship on the device and what they can advertise as core features of the device. Applying for the RYF certification will not require any changes to the repo in any way. No bans, no shaming, no restrictions. It is a HARDWARE certification, not a software certification. The software requirements are only extended to what is required to make the hardware function. The Pandora is a GPU driver away from being able to apply for certification right now. If we were somehow able to ditch the binary blob that runs our graphics we would have a RYF device, complete with repo exactly as it is.
Exophase
Nothing good will ever come of Exophase.
Nonetheless, I think it'd be kind of lame to offer RYF hardware and not at least update the repo to allow filtering to remove "non-free" listings, maybe making that the default mode. It's going to be kind of jarring if the hardware promotes one message but the software infrastructure seems disinterested in it. I think if Pandora goes for RYF it should be because of an actual legitimate interest in what it represents and not just a sales grab. But that's just my opinion on it.
There is however nothing stopping anyone from dropping the GPU driver from a Pandora distro right now, it's not essential for the Pandora's function and probably a strong majority of apps don't use it. All of the basic 2D stuff is handled by the display controller which does have open support in the kernel.
Grench
Forum Addict!
- Joined
- Oct 3, 2008
- Messages
- 6,629
Anytime someone claims to be promoting freedom by stating their required 'restrictions' makes my skin crawl.
To be compliant then, the www.pandora.org web site would have to remove all of the emulated games shown in the animation/advertisment (not free even if the emulators are). Any way you slice the 'restrictions on advertising' it is a restriction on freedom of speech.
To be compliant then, the Pandora would need to ship without a link/repo that includes non-free software for which there is no 'free' version. There goes DS emulation.
'No bans' - how do you figure if you ban advertising non-free capabilities?
'No shaming' - Bull. That is the entire basis of operation for the FSF/RYF system - shame companies into doing something outside of their best interest.
'No restrictions' - You stated that TWO sentences after naming the restrictions.
SOOOOO not worth it.
It is a current trend for political and action organizations to name themselves using terminology that is the exact opposite of what they're really about. Calling something, 'Respects Your Freedom,' then requiring limiting free speech and choice - what a sham.
WizardStan
Mega GP Mania
- Joined
- May 24, 2008
- Messages
- 16,731
No, it wouldn't.
No, it wouldn't.
That has nothing to do with the repo.
That has nothing to do with the repo.
That has nothing to do with the repo!
The Pandora, as it is right now, with the repo unchanged and the website looking exactly like it does is an open source GPU driver away from qualifying for the RYF certification.
The "Respects Your Freedom" certification is EXACTLY WHAT IT SAYS. It respects YOUR (ie, the users) freedom. It is impossible to support two competing freedoms: if the manufacturer had the absolute freedom to do whatever they wanted, this would mean they have the freedom to restrict the user from doing something. If the user is to have full freedom to do what they want then it is necessary to restrict the manufacturer from putting restrictions on the user. These are a perfect dichotomy, This is a necessary fact, you cannot give one group absolute freedom without restricting the freedom of another. If you can think of a logically consistent way of ensuring that users have full freedom without even a single restriction on the manufacturer then I would love to hear it.
WizardStan is right. The point of RYF is that the hardware respects the freedom of the user. Not the "freedom" of the vendor to close stuff down.
User freedom means the vendor is restricted - vendor freedom means the user is restricted. DRM means Digital Rights Management from the point of the vendors, while it means Digital Restrictions Management from the point of view of users.
Of course the user is free to get himself "chained" by closed stuff on RYF hardware. But that's a choice the user makes - the important thing is that she gets a choice, i.e. it is possible to use the hardware without any chains.
User freedom means the vendor is restricted - vendor freedom means the user is restricted. DRM means Digital Rights Management from the point of the vendors, while it means Digital Restrictions Management from the point of view of users.
Of course the user is free to get himself "chained" by closed stuff on RYF hardware. But that's a choice the user makes - the important thing is that she gets a choice, i.e. it is possible to use the hardware without any chains.
Grench
Forum Addict!
- Joined
- Oct 3, 2008
- Messages
- 6,629
"It is impossible to support two competing freedoms"
That could be the mantra of religious fundamentalists and dictators everywhere.
I think you need to re-visit your, "No, it wouldn't" strings.
Advertising non-free software as a primary device purpose is verboten. Those games being shown in the emulators, on the device's home page, are non-free software. Emulation of old game consoles is a primary purpose of this device. Although RYF would allow you to show the emulator's welcome/pick screens, actually showing games in action on the device would be against their rules.
Even if YOU don't like non-free software, the act of banning the advertising non-free capabilities and software is a direct restriction of freedom of speech. To me, that makes the FSF/RYF simply evil in practice, regardless of their intent or name.
I've said it before - if we can get a FOSS driver device with the same performance characteristics for the same money, fine. It's a fine thing to strive for. Supporting and bowing to a totalitarian standards organization - regardless of their putting 'Free' and 'Freedom' in their names is not worth the effort.
Grench
Forum Addict!
- Joined
- Oct 3, 2008
- Messages
- 6,629
It is perfectly possible for a device to be run by FOSS drivers and OS - and still advertise the non-free software without impinging one iota on the freedom of the end user. Both the end user and the vendor can have their freedom and eat it too so to speak.
The RYF and FSF folks cross the line though by demanding that the vendor curtail their freedom of speech. To me that is about 10 steps too far over the line. It is overly onerous to the device manufacturer and grants a 3rd party organization unpaid power over the functions of a business.
The Pandora successor hardware could meet RYF requirements in every way - and I'd still be opposed to applying for certification because of the restrictions that it places on how the hardware can be advertised, what can be shipped on it and what types of uses can be promoted. Loosing those freedoms is not worth the piece of paper or stamp on the box.
WizardStan
Mega GP Mania
- Joined
- May 24, 2008
- Messages
- 16,731
That's still a restriction. You aren't telling the company they can't advertise non-free software but you are still restricting them to a FOSS operating system. You're not disproving my point, you're arguing something completely different, where the acceptable line between respecting the users freedom and the companies freedom is.You are misinterpreting the intention of the advertising restriction. It isn't to prevent the company from advertising non-free software, it is to prevent them from making the non-free software a core feature. If you sell an MP3 player and the MP3 package is closed source, it is a restriction on the users freedoms, they cannot modify that MP3 package, and therefore the hardware does not Respect Your Freedom. If you sell a music player, however, so long as at least one of the formats it supports for any given purpose is user serviceable, it Respects Your Freedom. You can even advertise it as a music player, capable of playing MP3s after downloading the proprietary, closed source MP3 package, and it still Respects Your Freedom because the MP3 playback is an extension and not the core feature itself, provided the OGG format (which serves the same purpose as the MP3 format) is user serviceable.
The restriction on advertising does not mean they cannot say "can play DS games", it is to prevent them from saying "is a DS emulator". The criteria specifically says "we don't object to clear factual statements informing the user that the product also works with specific proprietary operating systems", and combined with their talk about general purpose installation software which also specifically mentions the rights of the user to download non-free software, it is not in any way a stretch to see that they do not intend to restrict factual statements of capability of the device, even when those capabilities extend into closed source and non-free software. It is a distinction between advertising what the device IS vs what the device CAN DO.
The Pandora is advertised as, I don't know, handheld pocket gaming computer, whatever. There's going to be certain things about what that means IT IS. All the things that IT IS must be based on "free" software: music and video player, able to run user written code, browse the internet, play some games, bunch of other stuff, who knows, we could write a giant list of all the tings that it means TO BE a handheld pocket gaming computer if you really wanted but I don't think that's necessary. What it IS NOT is a DS emulator. It IS NOT an SNES emulator. It IS NOT a Windows machine. It IS NOT even an Android machine. It CAN run Drastic. It CAN run SNES9x. It CAN run Windows through QEMU. And it can run Android. These are factual statements and there is no restriction on any advertisement that says so. They can say it can run Drastic, they can say it can run Windows, they can say it can play SNES ROMs, so long as these are advertised as things it can do and not confused with things that it is meant to be then it is fully respecting the users "freedom".
The CAN say that it runs Drastic: it is the users choice to go for this secondary feature or not. They CANNOT say that it is a DS emulator: if the user is to get the advertised functionality they have no choice but to download non-free software.
IT. IS. VOLUNTARY. I don't understand why you're running this around in circles. You do not lock yourself into some contract by which there is no escape if you apply for certification, that is not what "restriction" means in this case. It is only a restriction if you want to maintain the certification. If EvilDragon discovers that everything he has done is completely RYF compliant down to the advertising he can receive the certification. By doing so, maybe he discovers that the certification really did help sales, and in future he does make specific decisions about how to advertise based around this I don't see any problem with this: he isn't making these changes because he CANNOT do otherwise, he is making the changes because he wants to keep the certification, because it turns out that in this hypothetical scenario the certification is a good thing.If it comes to light that the P2 can run SteamOS, he can then advertise, factually "Can run SteamOS" without risking the certification. If he thinks it is worthwhile to go a step further and actually advertise the P2 AS a SteamBox(*) then there is absolutely nothing to prevent him from dropping the certification and running the advertisement that it IS a SteamBox. It is his choice to discard the certification fully aware that he can no longer take advantage of whatever benefits he was getting from it, but that is his right. Your flat refusal on the basis of what is required to maintain certification is completely nonsensical. If there was a binding contract and an actually verifiable loss (beyond the loss of benefits) then you would be right and we wouldn't be having this argument (again), but there isn't and you aren't.
(*) Even this may not be a problem now that we've actually heard the announcement on what SteamOS is going to be, but for the moment and until we hear otherwise let's assume that the SteamOS license isn't FOSS compatible.
Last edited by a moderator:
Grench
Forum Addict!
- Joined
- Oct 3, 2008
- Messages
- 6,629
Step right up! Give up some of your rights to free speech and we'll give you a 'Freedom' button! If you later decide to speak out of line, we'll take the button back and tell everyone how naughty you were and how they should boycott your product.
Entirely not worth it.
Entirely not worth it.
WizardStan
Mega GP Mania
- Joined
- May 24, 2008
- Messages
- 16,731
You have absolutely no idea how absurd you are. It's actually quite amazing that you still don't get it. I don't know what else I can possibly say that would convince you how wrong you are on your reasoning.
comradekingu
Glowing ember
The absolute freedom to both parties, seller and purchaser, is diamentrally opposite. When we aquire hardware we are at the limits of the vendor, when we extend that freedom we get (hopefully) (or restrict it, alter through extending it), to the user, we are providing freedom, that the user couldnt have without us giving up the freedom to restrict that user through nonfree drivers. If the user so wishes, one can make or install a proprietary driver at will. Proprietary solutions, by extention, provide less freedom since they dont respect the freedoms of free software. And that is still from the users point of view. See posts by WizardStan 1 and 2 plus wb 1
By your logic the most free is when the vendor pulls out all the stops to really divert from free as in freedom software.
I dont think firmware is part of the spec (from my understanding) unless it has the ability to spy on the user. Will have to check that.
Why is running a romdump with free software emulators not permitted in whatever context you are referring to , i fail to see exactly how.
Even if you make that point, (and you dont) it isnt detrimental to freedom of speech, it would be in tune with whatever the certification is. Because that isnt mandatory, nor permanent. And that is down to, choice.
What old game consoles dont have FOSS games? I cant think of any. Its demos and those games that really exhibit the emulator performance for the older stuff. Nothing you say in terms of RYF cert. is really function specific, just as much as the pandora isnt a DS emulator device, (for obvious nintendo reasons) There are free DS emus btw, segwaying us into where your 8:47PM post is continues:
Again, wrong cert. You could use that to document how it works. I suppose it has other uses too, with homebrew DS programs/games, though they arent recognizable. It could possibly make sense to document wine with proprietary apps too since the foss stuff is mostly ported, and its easier to tell you arent cheating. Lets for the case of the arguement say you want to show reactOS in a world where no FOSS windows stuff exists, then you have to show non-free apps.
Hardware that Respects Your Freedom is actually a system that promotes hardware that respects your freedom, the two are not the same.
This is irony at its finest.
Could you clarify, or at least link to what you are referring to?
What is ironic about it, intel has to my knowledge free drivers across the board, except for the powerVR stuff.
The problem isnt what I write, its that you dont read it,
neither the post i was answering, (nor the context in reference to an understanding of how it is two different certifications.) You seem to not even be able to grasp that concept.
You failed to notice i was talking about driver availability (either through source, writing with documentation, or reverse engineering fully.) In any event, you were also wrong. Notice i had reasoning, you didnt.
If you quote something, or make a reference to something, make sure to have read and understood it.
You are shunning the thread topic in disregarding the distinction between two certifications, and continue to spread FUD.
Last edited by a moderator:
Similar threads
