I have gone through the roadmaps and read other forums and it seems that RYF for all of purisms products will never be possible with companies like intel or amd or anything that isn’t pretty old unless they go with open CPUs which are still very early and overpriced for someone to actually buy. Will purism actually release an RYF product soon? Maybe the mobile phone? Got some doubt as heard proprietary blobs are still ran on a secondary chip. I don’t want to listen to promises or the goal of the company. Is it possible with the hardware available and the firmware needed to run that hardware with a well functioning OS for a truly freed mobile phone (that can connect to to the internet and handle GSM) and will purism drop a product that is actually freed as in more free than computers used by freedom lovers? Maybe a freed EC, a libreboot fork (not coreboot fork) and a fully functional computer. I know new CPUs are a no-go so can’t they just use old CPUs with some effort? Who will step up; purism or a different group?
As far as the Purism questions go , I would ask them.
This is no blocker for the RYF certification as blobs on secondary processors which are not intended to be upgraded by the user meets the exclusion criteria (https://ryf.fsf.org/about/criteria). I would bet the Librem 5 is the first product which could get the RYF. I don’t think the Intel notebooks could get it easily. Coreboot still has proprietary bits, libreboot is afaik not considered yet. Maybe we can hope to see some RISC-V notebooks in the next 5 years?
RISC-V notebooks. Guessing they are going to be quite pricey and more like webbooks than notebooks.
Edit: I am hoping they can achieve to lower the price of their products as I simply can’t be willing to pay so much for hardware that isn’t even RYF when I can just buy second hand computers for cheap like T400 for 60 GBP or an x200 for 70 EUR and libreboot myself.
‘If’ the L5 gets RYF, I will still not completely trust it as some things are still not freed and the FSF is not really communist enough to my liking but I would at least consider the device more. For the price and for what it is currently (a product without RYF), I think I will wait. I feel that something will eventually pop up about it when the FSF look at it so I’m not going to get my hopes up so if it does, it will be a pleasant suprise and if it doesn’t, I don’t feel dissapointed which is how I live most of my life. Would be pretty cool for a mobile phone to get RYF though :3
Edit1: I think it would be hilarious to see who releases a product first that can be truely free; purism or specific google chromebooks?
I don’t see anything that would block the RYF certification for the Librem 5. Purism has managed to get DisplayPort alt-mode working without proprietary blobs. The last thing to verify is whether Purism can get the cameras working without proprietary blobs, but I assume Purism chose image sensors that can work without them.
I assume that Purism will return to the Librem 13 and make a RYF 2-in-1 tablet/laptop based on the i.MX 8M Quad, but who knows when that will happen. I’m hoping that Purism will give the MNT Reform some competition for who can make the freest laptop.
Certain devices such as the current Librem 11 design will not meet RYF for the following reasons:
- ax201 WiFi firmware
- ME blob
- FSP blob
- Graphics microcontroller blob (though that one we might be able to remove, at cost of power consumption)
In fact, no devices from Intel or AMD in the last 20+ years will meet RYF. This is why the RYF devices you see are old Thinkpads, etc.
However…
Todd, @jonathon.hall and I had a productive conversation with Zoë Kooyman (FSF Executive Director) just now to clear the air on some past challenges and chart a course for our alignment moving forward. The meeting went very well and we were delighted that there was so much interest and overlap in our objectives. We are brushing off the dust on our efforts to have the Librem Key and Librem 5 devices RYF-certified, then will expand to additional options after we reach some of these milestones.
Good luck.
I’m glad to hear that the FSF is open to talking to companies that are actually trying to produce freer hardware. I really hope their new Librephone initiative is going to open their eyes to reality and they will stop promoting binary free/non-free metrics that help nobody actually move the ball forward.
What the FSF needs to do is promote the freest options available in each product category, rather than keep promoting its binary RYF certification that no modern manufacturer of a wireless device will ever be able to meet. The fact that the FSF does nothing to promote the Librem 5/Liberty and MNT Reform, which are the freest phones and laptops on the market today shows how broken the FSF’s approach has been over the years. It is ridiculous that the FSF looks at something like a bit of proprietary code to train the DDR4 RAM timing during boot, and decides on that basis that it won’t say a word to promote a free hardware project.
It made me tear out my hair that the FSF didn’t say a word to support Biktorgj’s efforts to reverse engineer the firmware on the PinePhone’s Quectel EG25-G LTE modem. The FSF should have been publicizing Biktorgj’s efforts and helping to organize donations or some kind of financing so he could keep working on it.
From what I heard on the FSF’s podcast about the Librephone, it appears that the FSF has finally started to realize that nothing is going to change if it doesn’t start financing efforts to work on free/open source firmware. However, the fact that they plan on spending 6 months studying the current situation is maddening. It is also maddening that the Librephone initiative is focused on AOSP phones, when the FSF should be promoting the Linux phones that are already on the market that are far closer to the FSF’s ideals than any AOSP phone using a Snapdragon, Tensor or Exynos processor will ever be. Good luck trying to reverse engineer the firmware on a mobile SoC like the Snapdragon. Qualcomm doesn’t even provide datasheets for its Snapdragon processors even for the Linaro engineers who signed NDAs. The Linaro engineers have to read Qualcomm’s source code to figure out how the Snapdragon works, whereas NXP provides 3000 pages of documentation on its i.MX 8M processors, and you can get most of that documentation without signing an NDA or being a purchaser of the i.MX 8M.
Despite my criticisms of FSF’s past actions, I actually do think the Librephone initiative is a positive step for the organization. It might come to nothing, but if the FSF starts funding free firmware development or directing people toward making donations for specific free firmware projects, that is something that should be cheered.
I fully agree with your points.
Zoë Kooyman was highly enthusiastic and supportive of working with Purism in every facet that was mentioned, so I give her and the FSF kudos that they are becoming easier for established companies to cooperate with.
In the FSF’s defense, I am told that the FSF did contact Purism and scheduled a meeting prior to the Librephone announcement, but the meeting date was bumped several times (by whom and for what reason, I am unaware) and it fell through the cracks. “Librephone” was not mentioned as an agenda item, perhaps to not put the cart before the horse, so I surmise that the partnership proposal fizzled out due to lack of urgency/engagement for whatever reason.
At least they gave the Librem 5 a mention here once upon a time: Ethical Tech | Giving Guide
… though not here: Products | RYF
They even could give a “closest to RYF” certificate. This would not weaken the RYF, but would also show “this product is the best we can have for that purpose”. So the ultimative goal is the same while honoring the best approach at the same time.
That is good to hear. I think that it might be worth Purism’s time to try to get included in FSF meetings/chats where the Librephone project gets discussed. It would be an opportunity to suggest expanding the project to cover both AOSP and Linux phones, since any FOSS driver/firmware development could help both operating systems. Purism might have the chance to direct the FSF’s attention toward WiFi/BT and cellular modems that can be used by Purism, Pine64, Mecha and Liberux, rather than just focusing on the components found in unified mobile SoC’s (Snapdragon, Tensor, Exynos, Helio/Dimensity or Tiger/Tanggula).
The only way I can see this happening is if the RYF criteria stay as they are but a section is added on the project’s website for non-compliant devices which a vendor willingly subjected to RFY certification process. That would require a description of why the candidate failed. It would allow the readers to make an educated decision. I imagine it to be somewhat similar to “Anti-Features” lists in F-Droid.
Out of curiosity though, is it the case that the FSF highlighted GNU systems that were truly free, and then people who feel like they should “move the ball forward” to modern times basically created that feeling in themselves by using nonfree proprietary software?
Because then if I were the FSF, it seems I would have to consider that how people at large feel on the issue might be the result of an intentional manipulation campaign. And if you already have freedom, why give up your freedom because other people already gave up theirs and created a society that pressures you to give up likewise? I mean just from a logical standpoint wouldn’t that be extremely, extremely emotionally hard to do because it surely wasn’t what you wanted and so a need to do it is an acknowledgement of not having absolute freedom (in the first place)?
Isn’t pinephone a psyop though, basically to convince people librem 5s dont work by providing a similar but cheaper dysfunctional one?
[Imo if I could buy a $5000 liberty phone with way better specs and equal trust, freedom, etc purisf focus, I probably would, so I’m obviously on one end of a spectrum of bias here, but still….]