Project

General

Profile

Purism Librem 5

Added by Abcd Efgh 4 months ago

Hello,

is the upcoming Purism Librem 5 (https://puri.sm/shop/librem-5/) interesting for Replicant?

The CPU will be an i.MX6/i.MX8, where we can separate the baseband modem from the main CPU, digging deeper and deeper to protect your privacy and isolate components for a strong security hardware stack.


Replies (8)

RE: Purism Librem 5 - Added by Fil Bergamo 4 months ago

I'm maybe wrong, but..
Purism's only interest to my eyes is into collecting bucks with found raising.
They claim to be building the "libre-est" hw available, but it has long revealed to be much less free than what we currently have already.
And much more expensive.
(see: found raising to build a "libre" computer from scratch.. with i5s and i7s.. in it)

Apart from that.. It could be interesting in a way..
But just like with fairphone, I believe that if their interest was really into making libre systems, the should've contacted the Replicant project to join efforts from the beginning, before the found-raising campaign.

To me, "purism" stands for "pure (and expensive) lies". Period.

I wouldn't bother too much about their proclaims..

RE: Purism Librem 5 - Added by Raúl Priego 4 months ago

I think that Librem5's developement of a comunication stack may be evaluate in order to improve Replicant skills.

RE: Purism Librem 5 - Added by Timothy Magee 2 months ago

Seeing that the Librem 5 is fully funded and will more than likely happen, I think this could be a great potential target. I understand the concerns of Fil Bergamo and I am concerned that they are using the "Purism" name but aren't even shipping with CoreBoot, not to mention LibreBoot. Though to be fair they have put significant effort into freeing their BIOS; they are currently working on a CoreBoot port and they have done a lot on ME Cleaner to clear up the nastiest binary blobs as they (hopefully) work toward a fully free BIOS.

However, even given those reservations, the Librem 5 should be way better for freedom than any currently supported Replicant phone.

It will have a free bootloader and will have support for accelerated graphics, gps, etc. with free software. Even if I thought Purism was all lies, if they make a phone with a free bootloader and free graphics firmware with better specs than the Galaxy s3 I would donate money toward purchasing one for anybody willing to try a port. (I currenlty lack the funds to purchase an entire device, hence donating funds toward purchasing and I lack technical know how to port though I could probably learn)

TL;DR I think that the Librem 5 would be an excellent choice for a possible port.

RE: Purism Librem 5 - Added by Fil Lupin 2 months ago

Fil Bergamo wrote:

They claim to be building the "libre-est" hw available, but it has long revealed to be much less free than what we currently have already.
And much more expensive.
(see: found raising to build a "libre" computer from scratch.. with i5s and i7s.. in it)

You seem to be reluctant especially because of i5 and i7 because of privacy issues, could you give us some references which could explain what would be bas with this?

But just like with fairphone, I believe that if their interest was really into making libre systems, the should've contacted the Replicant project to join efforts from the beginning, before the found-raising campaign.

To be really objective, I'm not sure the Replicant community has the forces for this kind of need. Currently, the community is only based on people which give free-time, that's great but we currently had difficulties to find answer to issues on the forum. These motivated but small forces can not be sufficient to help someone who project to spend a lot of money to launch a phone in a big scale.
Perhaps even these guys never heard of Replicant, but it would be a bit strange if privacy is one of their focus so I will not consider this reason.

Whatever, it will be interesting to contact them and talk about what could be done with replicant.

Since I only read a few of the forums threads and it seems some announces are sometimes not clear, it would be interesting to give a few links about some informations to understand who claims it (coreboot for instance).

RE: Purism Librem 5 - Added by Fil Bergamo 2 months ago

Fil Lupin wrote:

You seem to be reluctant especially because of i5 and i7 because of privacy issues, could you give us some references which could explain what would be bas with this?

It's basically because of Intel's ME and AMT (https://libreboot.org/faq.html#intel)
When Purism were foundraising for their first laptops, they claimed they were going to build up the first hardware that was totally free-software-friendly..
The ME affair was already well-known at the time, and Libreboot was already a reality.
They didn't even bother about the processor issues, and chose to put i5s and i7s in their laptops.
Those series of processors are still impossible to use without the ME part.
Purism was also claiming that they were working with the FSF on a RYF certification, which was clearly impossible from the beginning, given the above situation.
A very basic search online could reveal the state of the art in software-and-harware freedom.
I don't know anything about hardware myself, but still I have been able to find out such issues and to document myself.

It is then very strange for a company whose business is producing hardware, that claims to be focused on freedom, not to know such basic requirements.

To be really objective, I'm not sure the Replicant community has the forces for this kind of need. Currently, the community is only based on people which give free-time, that's great but we currently had difficulties to find answer to issues on the forum. These motivated but small forces can not be sufficient to help someone who project to spend a lot of money to launch a phone in a big scale.

That's the point. Spending a lot of money.
I could name a couple of people involved in the very inner core of Replicant that - I'm sure - would be very happy to get hired in order to provide their knowledge and skills to a company which is really concerned in producing free hardware.
It's all too easy to collect money from the free software community, promising to give back totally-free hardware, and then spend that money to produce a hardware platform that is perfectly equivalent to any other proprietary hardware, in respect to software-freedom friendliness.
I think that, instead, that money should be spent on hiring people that really know what is needed to produce free-software-friendly and freedom-respecting hardware, and that maybe also know how to write software for it.
Disregarding Replicant when claiming to build up a freedom-respecting phone is a clear sign of not wanting to do anything near freedom at all.

Perhaps even these guys never heard of Replicant, but it would be a bit strange if privacy is one of their focus so I will not consider this reason.

I would guess that a majority of regular non-technical people concerned with software freedom know about Replicant.
I strongly doubt that Purism never heard of it.
Again, let's suppose they really don't know about Replicant at all. Then it would only confirm that they don't care about free software enough to do a quick research on the state of the art.

Whatever, it will be interesting to contact them and talk about what could be done with replicant.

Yes, maybe.
Or maybe it could be a waste of the already very limited time we have..
Should that be the case (I think it is), I believe time is better spent into the project.

Since I only read a few of the forums threads and it seems some announces are sometimes not clear, it would be interesting to give a few links about some informations to understand who claims it (coreboot for instance).

Good suggestion.
I think we (as broaden free software community) need a clear Wiki-like page that explains why most of Purism's claims are wrong and misleading.
And we need someone that can spend time in putting it together..

Oh! if only Purism was really working WITH us... we wouldn't need to waste time in stating facts against their lies...

RE: Purism Librem 5 - Added by CyberLeo Kitsana 2 months ago

Can we stop with the insult flinging already? It's painfully obvious from the state of the art that a single all-or-nothing approach to libre hardware is not working fast enough. Idealism is important, but at the end of the day we all have work to finish, and the wider world doesn't really care how it gets done.

If Purism thinks they can make an iterative approach work, why not see what they can do? They've provided field updates that bring Coreboot and a neutered Management Engine to their already-deployed laptops, so they seem to be making good on their promises thus far.

It's notable to point out, by the way, that not even Bunnie's Novena is eligible for FSF's RYF, despite the free availability of everything from the board schematics on up.

RE: Purism Librem 5 - Added by Fil Bergamo 2 months ago

CyberLeo Kitsana wrote:

Can we stop with the insult flinging already?

First of all, I'm sorry if my previous words could be considered insulting.
That was not my intention.
I don't believe I insulted anyone, just stating my point of view, using no insulting words.
But, again, if you had that impression, I ask pardon. I didn't mean to insult anyone.

It's painfully obvious from the state of the art that a single all-or-nothing approach to libre hardware is not working fast enough. Idealism is important, but at the end of the day we all have work to finish, and the wider world doesn't really care how it gets done.

The point I was trying to make is exactly the opposite:
it is not a question of idealism (or "purism" as they speak it themselves).
It's the librem-makers that use such terminology, not me (see: <<creating an open utopia>> ...not free.. open)
From my point of view, idealism has nothing to do with the issue.

My point is instead very concrete:
we have reached a certain CONCRETE point in the long road to hw-and-sw freedom, which currently is represented by:
- computers with a totally-free boot software (read: Libreboot) running a totally-free operating system.
- computers that are on the way to be manufactured (or are already?), which are based on (almost?)totally-free hardware and can run totally-free sw, boot included (read: Talos)
- a free mobile operating system that runs on the main cpu of "smart phones" without requiring or suggesting any proprietary software (read: Replicant)
- non-free bootloaders that are needed on the above "smart phones" (clearly stated by Replicant's documentation)
- totally-non-free modem that resides on the same "smart phones" outside user's control (again, clearly stated by Replicant's documentation)

Is it enough?
No, not at all.
The point is that THIS is the borderline that needs to be expanded and advanced.
What Purism is doing is just claiming to be working on advancing the free-software state of the art, while they are in facts stepping backwards.
I don't criticize what they are doing.
I strongly criticize the misleading way they are trying to sell it.
They initially claimed they were working with the FSF for RYF certification on their first laptops, when this was EVIDENTLY impossible by the very hardware they were using.
They now claim they are about to ship an <<open utopia>> that is totally under the user's control, which is in fact not the case, given that they are going to ship proprietary firmwares and possibly proprietary drivers (their wording is so ambiguous that I can't tell for sure).

Replicant-driven systems are far from perfect, but we all know it and we all state it clearly in every possible context.
Nonetheless, we don't take that as an excuse.
Replicant is not taking shortcuts to make itself more appealing, because Replicant focus is actually on user's freedom.
While instead, Purism are only focused on selling things, regardless of freedom issues.

If Purism thinks they can make an iterative approach work, why not see what they can do? They've provided field updates that bring Coreboot and a neutered Management Engine to their already-deployed laptops, so they seem to be making good on their promises thus far.

Neutralize is a very biased word they use.
Actual neutralization of the ME could possibly mean only two things: either the complete (100%) removal of ME or its complete substitution with a free-software version.
Unfortunately, none of the above steps has been made.
Yet, instead of giving complete and accurate information about what is currently known about the ME, Purism tries to get along with the argument that a big enough part of it has been removed.
This is clearly trying to get around the issue: there is still a piece of powerful software running on their CPUs, that nobody can control or even study.
The sad truth is nobody knows what the remaining part of ME does.
As such, it would be honest to state that, and to put the emphasis on the issue, rather than trying to obfuscate it.
I strongly suggest to read the libreboot page about ME to get a better, and way less biased, understanding of it.

It's notable to point out, by the way, that not even Bunnie's Novena is eligible for FSF's RYF, despite the free availability of everything from the board schematics on up.

(sorry for the bad emphasis, but...) Now what???
Even Replicant itself is not eligible for RYF. We all know that.
The difference is that we are not claiming we could possibly in the future receive that certification.
And this is because Replicant is a honest project, that is not trying to sweep issues under the rug.
While it seems to me that Purism are not as much honest in this respect.
But this is just my opinion, which I've tried to support with facts.
It is still totally possible that I am wrong. But I currently see no concrete fact that can make me change my mind.

RE: Purism Librem 5 - Added by Fil Bergamo 2 months ago

I'd also like to point you all to a post by Vikings on trisquel's forum that describes what I think on the topic, better than I could possibly do by myself.
In particular, it expresses my same point of view on why compromising on freedom is never a step forward.
Freedom is the most important feature both in software and hardware.
No other feature can be worth losing freedom.
If we cannot agree on that basis, then it would be extremely difficult to agree on any other free-software-related topic.

    (1-8/8)