@lxo
Today, there is no way to run a computer without non free firmware. The good and proper way to handle that would have been to accept that (as with the non free operating systems in the early 80s) and to fund/promote/encourage projects to produce free replacements.
Instead, the fsf chose to put their heads in the sand and pretend non free firmware doesn't exist when it's burned to ROM.
@mjg59
Today, there is no way to run a computer without non free firmware. The good and proper way to handle that would have been to accept that (as with the non free operating systems in the early 80s) and to fund/promote/encourage projects to produce free replacements.
Instead, the fsf chose to put their heads in the sand and pretend non free firmware doesn't exist when it's burned to ROM.
@mjg59
@lxo
When there was no free GNU system yet, most people believed that Emacs was a nice editor but there is no chance they'll ever succeed in writing a free os.
When there was a free compiler and a free libc, must people were like, this is a nice user space but nobody will ever make it a fully free os.
At every stage, the GNU project proved them wrong.
Why would the situation be different for non free firmware replacements?
@mjg59
When there was no free GNU system yet, most people believed that Emacs was a nice editor but there is no chance they'll ever succeed in writing a free os.
When there was a free compiler and a free libc, must people were like, this is a nice user space but nobody will ever make it a fully free os.
At every stage, the GNU project proved them wrong.
Why would the situation be different for non free firmware replacements?
@mjg59
@lxo
It would be one thing if you advocated against firmware that can't be changed without a signature by the hardware manufacturer, but that's not the case here. You're generalizing that all hardware requires signed firmware blobs.
I agree that verifying firmware signatures in hardware is evil and should be outlawed. But hardware that does no verification, or that verifies only a checksum, in hardware? That's perfectly fine.
@mjg59
It would be one thing if you advocated against firmware that can't be changed without a signature by the hardware manufacturer, but that's not the case here. You're generalizing that all hardware requires signed firmware blobs.
I agree that verifying firmware signatures in hardware is evil and should be outlawed. But hardware that does no verification, or that verifies only a checksum, in hardware? That's perfectly fine.
@mjg59
@lxo
And although I don't think it works for me, I can understand the argument for Linux-Libre. If there is no free firmware, and you prefer to keep your hardware unmodified, sure.
What I'm saying though is that forbidding any form of update, ever, of the firmware, and doing so in hardware, is wrong, because it makes it equally impossible to replace the non free firmware with a free one.
@mjg59
And although I don't think it works for me, I can understand the argument for Linux-Libre. If there is no free firmware, and you prefer to keep your hardware unmodified, sure.
What I'm saying though is that forbidding any form of update, ever, of the firmware, and doing so in hardware, is wrong, because it makes it equally impossible to replace the non free firmware with a free one.
@mjg59
- replies
- 0
- announces
- 0
- likes
- 0
@lxo
That's an opinion, not a fact, and one I very much disagree with.
Software is software. It doesn't matter whether the software is burned in ROM, it's still software.
To claim otherwise means you're fine with running non free software.
I'm not. I accept that it's not possible in today's world, but it's still not a good thing.
@mjg59
That's an opinion, not a fact, and one I very much disagree with.
Software is software. It doesn't matter whether the software is burned in ROM, it's still software.
To claim otherwise means you're fine with running non free software.
I'm not. I accept that it's not possible in today's world, but it's still not a good thing.
@mjg59
@lxo
Let me put it this way.
There are only very few cases in which firmware really needs to be embedded in the hardware and can't be put elsewhere.
The initial few opcodes of a CPU are a good example: those can't be outside the CPU because it still needs to initialise its components and so there is absolutely no possible way it can be outside the CPU die, and it's small enough that it doesnt matter if it's not updatable. @mjg59
Let me put it this way.
There are only very few cases in which firmware really needs to be embedded in the hardware and can't be put elsewhere.
The initial few opcodes of a CPU are a good example: those can't be outside the CPU because it still needs to initialise its components and so there is absolutely no possible way it can be outside the CPU die, and it's small enough that it doesnt matter if it's not updatable. @mjg59
That's software, but it can't really be free software (because you can't change it even if your wanted to)
Anything that can theoretically be updated or loaded from outside the die though is software. It doesn't matter whether it is "indistinguishable" from hardware, the fact is that it *is* software. And as it is software, the only ethical thing for it is to be free.
@mjg59 @lxo
Anything that can theoretically be updated or loaded from outside the die though is software. It doesn't matter whether it is "indistinguishable" from hardware, the fact is that it *is* software. And as it is software, the only ethical thing for it is to be free.
@mjg59 @lxo
I would have been fine with an ryf campaign that said, say, at least X amount of the embedded software in the device must be free software, and no device can be used that requires non free software if alternative for the same function that doesn't require it exists, and no device, whether with free software or not may require cryptographic signatures for changes unless the owner can control the keys.
@lxo
@mjg59
@lxo
@mjg59
Doing this incentivises the device manufacturers, who are in a better position than anyone to write free replacement firmware, to actually do so.
Instead, the fsf caved to the people who push non free embedded software and told them it's allowed, as long as the fsf can pretend the embedded software is not there.
I find this sad, and a betrayal of everything the free software movement stands for.
@lxo
@mjg59
Instead, the fsf caved to the people who push non free embedded software and told them it's allowed, as long as the fsf can pretend the embedded software is not there.
I find this sad, and a betrayal of everything the free software movement stands for.
@lxo
@mjg59
@lxo
I don't think it's a trend. Even so, you said, and I quote, "a piece of software that you don't stand a chance of modifying yourself, because it's digitally signed by the vendor so that you can't".
Not all non free firmware is like that. Yet the ryf campaign requires that no firmware be updatable as though it were.
So, when a replacement free firmware is built, a device that has the RYF badge will be less free than one that doesn't.
@mjg59
I don't think it's a trend. Even so, you said, and I quote, "a piece of software that you don't stand a chance of modifying yourself, because it's digitally signed by the vendor so that you can't".
Not all non free firmware is like that. Yet the ryf campaign requires that no firmware be updatable as though it were.
So, when a replacement free firmware is built, a device that has the RYF badge will be less free than one that doesn't.
@mjg59
@lxo
Also, no, I'm not just associated with Debian, I've been a Debian Developer for just over 25 years now, and have been a DPL candidate thrice. Trust me when I say that we don't welcome the non free blobs. Our strategy however is one of pragmatism: if people need to buy expensive hardware to run free software, that's a barrier to adoption. We keep the barrier low in hopes that it will convince some people.
@mjg59
Also, no, I'm not just associated with Debian, I've been a Debian Developer for just over 25 years now, and have been a DPL candidate thrice. Trust me when I say that we don't welcome the non free blobs. Our strategy however is one of pragmatism: if people need to buy expensive hardware to run free software, that's a barrier to adoption. We keep the barrier low in hopes that it will convince some people.
@mjg59
@lxo
This is not the strategy that the fsf chose to make, and that's fine. It also won't work every time, and that's also fine. But some people will be convinced, and become a member of the community, and that's a good thing.
I do think that having blobs is fine if there is no alternative, for very much the same reasons as why the GNU project started off by accepting non free kernels while replacements were being written.
@mjg59
This is not the strategy that the fsf chose to make, and that's fine. It also won't work every time, and that's also fine. But some people will be convinced, and become a member of the community, and that's a good thing.
I do think that having blobs is fine if there is no alternative, for very much the same reasons as why the GNU project started off by accepting non free kernels while replacements were being written.
@mjg59
@lxo
Oh come now. The ryf program is a hardware certification badge. Why would you create a hardware certification badge if you weren't trying to influence hardware manufacturers?
I vehemently disagree with you that your stance is consistent. If it's theoretically updatable, it's software. Artificially crippling your hardware so that you can't update it anymore it's like putting your fingers in your ears and going "La La La can't hear you"
@mjg59
Oh come now. The ryf program is a hardware certification badge. Why would you create a hardware certification badge if you weren't trying to influence hardware manufacturers?
I vehemently disagree with you that your stance is consistent. If it's theoretically updatable, it's software. Artificially crippling your hardware so that you can't update it anymore it's like putting your fingers in your ears and going "La La La can't hear you"
@mjg59
@lxo
It's modifiable, therefore it's software, therefore it should be free.
That's consistent and makes sense.
Yes that's difficult to reach today. The GNU project has accepted similar compromises in similar situations in the past, and a hardware certification program that encourages that could go a long way into fixing that.
But sure, tell me I need to go cripple the Debian kernel instead of admitting you were wrong.
@mjg59
It's modifiable, therefore it's software, therefore it should be free.
That's consistent and makes sense.
Yes that's difficult to reach today. The GNU project has accepted similar compromises in similar situations in the past, and a hardware certification program that encourages that could go a long way into fixing that.
But sure, tell me I need to go cripple the Debian kernel instead of admitting you were wrong.
@mjg59
@lxo
I'm not sure you're getting what I'm saying, so I'm going to give it one more try and then just give up.
I'm not going to insult your intelligence by doing a socratean dialogue here. I know you know that non free firmware is software, and that as it is not free, it would be better if it were made free.
@mjg59
I'm not sure you're getting what I'm saying, so I'm going to give it one more try and then just give up.
I'm not going to insult your intelligence by doing a socratean dialogue here. I know you know that non free firmware is software, and that as it is not free, it would be better if it were made free.
@mjg59
@lxo
You (the fsf, not you personally) came up with a rule that allows you to ignore the fact that it is there so you can live your live with computing infrastructure that is under your control for as much as possible. I get that.
But I know that you know, deep down, if you are honest with yourself, that the non free software is still there and that the rule is an illusion.
@mjg59
You (the fsf, not you personally) came up with a rule that allows you to ignore the fact that it is there so you can live your live with computing infrastructure that is under your control for as much as possible. I get that.
But I know that you know, deep down, if you are honest with yourself, that the non free software is still there and that the rule is an illusion.
@mjg59
@lxo
So your goal, as the fsf, should be to come up with a plan to eradicate that non free software. I know it will be an uphill battle, but the whole GNU project was an uphill battle and that never stopped you before. It shouldn't stop you now.
I already made a few suggestions as to how you could use the ryf program to make that happen. You dismissed some of those options for reasons that I agree make sense.
@mjg59
So your goal, as the fsf, should be to come up with a plan to eradicate that non free software. I know it will be an uphill battle, but the whole GNU project was an uphill battle and that never stopped you before. It shouldn't stop you now.
I already made a few suggestions as to how you could use the ryf program to make that happen. You dismissed some of those options for reasons that I agree make sense.
@mjg59
@lxo
But there are more things you could do. Here are some more suggestions:
Make the ryf program a multi tiered program, with the lower tier being the current situation and the higher tier not allowing non free firmware. Initially you won't have many submissions for this higher tier. That's fine. Some people will aspire to get there, and start working on free firmware. Even if they fail, they still may do stuff that improves the world.
@mjg59
But there are more things you could do. Here are some more suggestions:
Make the ryf program a multi tiered program, with the lower tier being the current situation and the higher tier not allowing non free firmware. Initially you won't have many submissions for this higher tier. That's fine. Some people will aspire to get there, and start working on free firmware. Even if they fail, they still may do stuff that improves the world.
@mjg59
@lxo
In this, have some allowance so that hardware manufacturers don't have to choose between losing the ryf badge or destroying stock when someone announces free firmware for something not in their device.
Introduce bonus points, or stars on the badge, or some such, for each piece of free firmware that's used by the device. Simple, but could be effective.
@mjg59
In this, have some allowance so that hardware manufacturers don't have to choose between losing the ryf badge or destroying stock when someone announces free firmware for something not in their device.
Introduce bonus points, or stars on the badge, or some such, for each piece of free firmware that's used by the device. Simple, but could be effective.
@mjg59
@lxo
I just came up with these during my breakfast this morning. I'm sure you can come up with more if you need to.
The point is to reward and encourage people to rid the world of free software. I know you want this. I want this.
Or, I don't know, your can keep your head in the sand and wait until the fsf is completely irrelevant because everyone knows they don't really care about free software.
🤷
@mjg59
I just came up with these during my breakfast this morning. I'm sure you can come up with more if you need to.
The point is to reward and encourage people to rid the world of free software. I know you want this. I want this.
Or, I don't know, your can keep your head in the sand and wait until the fsf is completely irrelevant because everyone knows they don't really care about free software.
🤷
@mjg59
@lxo
Great. I'm glad that you find my suggestions acceptable and that you can see the benefits in them.
If there's anything I can do to help with making that a reality... I can't promise to do much but I can try, and would be happy to be kept in the loop regardless. s/pleroma.debian.social/debian.org/ for my email.
Also, thanks for ignoring the accidentally missed 'non' in my last post. No we don't want to rid the world of free software 😲😂
@mjg59
Great. I'm glad that you find my suggestions acceptable and that you can see the benefits in them.
If there's anything I can do to help with making that a reality... I can't promise to do much but I can try, and would be happy to be kept in the loop regardless. s/pleroma.debian.social/debian.org/ for my email.
Also, thanks for ignoring the accidentally missed 'non' in my last post. No we don't want to rid the world of free software 😲😂
@mjg59