[Dev] Misleading information in EOMA68 news

Christopher Waid chris at thinkpenguin.com
Thu Aug 25 08:42:42 GMT 2016


I'm going to reiterate that it is not my position that the wording 'free 
software friendly' is perfect. Changing this to compatible with free 
software operating systems may work. It may not. It may depend on the 
context. It may be challenging to integrate.

The laptops we sell currently @ ThinkPenguin are not RYF'd and shouldn't 
be RYF'd, but we are working on something better than LibreBoot in that 
it solves the free software problems in a more permanent long term way: 
EOMA68. X86 is dead and we do not need LibreBoot for non-X86 systems. 
The reason this issue hasn't been solved by us is because it's simply 
not possible given Intel's hostility and refusal to cooperate. Reverse 
engineering is a non-trivial task and the resulting code would not run 
on modern Intel systems due to digital signatures. We can do a lot more 
than what is feasible with LibreBoot, but it has taken years. Now that 
EOMA68 crowd funding campaign has succeeded though or is about to 
succeed we can do a 100% free software system (that is LibreBoot doesn't 
magically make a computer 100% free, there are other problematic 
components). We've got the source code for LCD/Keyboard controller 
firmware, bootloaders, CPU micro code, and similar for the EOMA68 laptop 
housing and Libre Tea Computer Card. That's huge. And there are more 
significant developments coming including the release of schematics and 
higher end CPUs.

I'm not sure if it is clear so I'm going to state for the record that 
I'm the CEO of ThinkPenguin and ThinkPenguin sponsored Luke's work on 
EOMA68. I'm a good person to respond to the criticisms about pretty much 
anything related to TP, EOMA68, and the comments below. We've pulled off 
getting code for chips in the past and have a solid track record 
including release of other RYF'd products. More than anyone else 
actually. We were in competition for being the first company to release 
an RYF product and our efforts to free hardware pre-date the RYF 
program.

On 2016-08-24 03:31 PM, Tiberiu-Cezar Tehnoetic wrote:
> On 24.08.2016 20:18, Christopher Waid wrote:
>> When this text was written LibreBoot didn't exist
> 
> I beg to differ.

The quote was 2008/2009. If there was something else referenced I didn't 
see it.

> The text we refer to is from summer 2014 after
> SouthEast GNU/Linux Fest 2014 took place on June 20-22, 2014.
> 
> https://www.thinkpenguin.com/gnu-linux/short-interview-christopher-waid-about-thinkpenguin-linux-action-show

The systems are compatible with Trisquel and Parabola GNU/Linux-libre 
and not dependent on any OS-loaded proprietary bits. This is not to say 
they are completely free of proprietary bits. That's what we've been 
working toward, but it's a much bigger goal to hit than just freeing the 
BIOS.

> While FSF awarded the RYF certification to Gluglug for the X60 laptop
> preinstalled with deblobbed Coreboot in December 2013.

The RYF Lenovo laptops with LibreBoot from Mini Free are not entirely 
free even though they have RYF status from the FSF. It doesn't appear at 
a quick glance that it is being advertised as entirely free either.

> http://www.fsf.org/news/gluglug-x60-laptop-now-certified-to-respect-your-freedom
> 
> And the deblobbed Coreboot has been named Libreboot "during early 
> 2014".
> That is, before the summer conference and show edition.
> 
> 
> https://libreboot.org/docs/index.html#why
> 
> Knowing that, in the beginning of the GNU/Linux Action Show interview
> you still falsely stated that ThinkPenguin "makes sure that all the
> hardware in your catalog is 100% free software friendly" (57:20). That
> includes desktops and laptops.
> 
> You went on and said that "every bit of firmware on your laptops is 
> free
> software, except for the BIOS which is... outside". (57:26)

You have to watch the entire segment and keep it in context to 
understand what was being said. It was not that there weren't any 
non-free bits. It was that there weren't any OS-loaded non-free 
firmwares or drivers that would interfere with compatibility and 
support. Notice the word 'outside' in the quotes. That part might not 
have been very clear, but by BIOS I was really referring to anything not 
loaded by the OS, and it was clearer if you quoted the later bit:

"basically anything that is loaded by the operating system is free"

Which obviously excludes the BIOS, hard disk firmware, keyboard/LCD 
controller firmware, CPU micro code, etc. These bits are all stored on 
flash and not loaded by the OS (or need not be loaded by the OS as a 
result).

Humorously it also becomes more apparent when you take into account that 
I had mentioned Linux Mint alongside Trisquel. This was essentially a 
sales pitch to users of non-free distributions (ie most Linux Action 
Show viewers) to care about and understand some of the damage done by 
proprietary drivers/firmware.

> Fast forwarding to November 2015 when FSF announces the discounted
> ThinkPenguin products for their associate members.
> 
> https://www.fsf.org/blogs/community/fsf-members-now-get-5-off-thinkpenguin-free-software-friendly-devices
> 
> Quoting:
> 
> "ThinkPenguin sells free software-friendly hardware, including laptops,
> desktops, WiFi adapters (useful if your laptop's WiFi can't work with a
> free driver), printers and more."
> 
> I guess it's convenient to have FSF recommend their associate members
> your laptops and desktops with proprietary BIOS 1 month after their 
> 30th
> anniversary in October.

These were not my words and I'm not sure this is really incorrect. The 
laptops and desktops do support 100% free distributions like Trisquel 
and Parabola GNU/Linux-libre.

> Right after in October you were sending off Libreboot founder and lead
> developer to go work on free software for GPS devices, because in your
> opinion she "would have done better" . You went on and said the effort
> to have free BIOS for older laptops "would have been better spent
> elsewhere", because getting software freedom-respecting laptops "is a
> harder problem to solve and it is going to require massive amounts of
> money".

Yes- I did say something along those lines. It's taken out of context 
though. I didn't feel that LibreBoot was the best approach and suggested 
working on a different hardware effort. I still would like someone to 
produce a free software GPS navigation device.

To back up on why this was the wrong approach you have to look back to 
2009 when I began looking into CoreBoot and talking with CoreBoot 
developers about porting of CoreBoot to modern systems. This was mainly 
at LinuxCon 2009. The problem as it was explained to me is it's 
difficult enough to port CoreBoot from one system to another let alone 
to newer generation hardware. At the time things were going from being 
difficult to impossible because Intel was adding signature checking. 
Ultimately it wasn't going to work out long term so we didn't move on 
it, and it hasn't. There were other important projects we worked on 
instead including getting sources for modern USB N wifi chipsets and 
resurrecting LibreWRT.

If you look at what we've done since then it becomes obvious why that 
was the wrong approach too. Luke was working on a modular computing 
standard (EOMA68) and we teamed up with him to work on this. He was the 
only person working on anything that seemed remotely plausible for a 
solution to the problem. The reason we went this way was because it 
would reduce the cost long term of designing and manufacturing systems 
that we would be able to make utilizing only 100% free software. It also 
gives us an additional angle. Because we can easily switch CPU/SOCs  we 
can play prisoners dilemma with the companies designing the CPUs/SOCs. 
If one or a handful don't cooperate with us we can easily switch to a 
competing CPU. It's much less likely to fail because of financial issues 
(insurmountable costs to design a new model or inability to get 
cases/parts, etc), will scale, and won't hit a brick wall like how 
LibreBoot has.

> 
> https://trisquel.info/en/forum/asus-chromebook-c201-now-supported-libreboot-arm-cpu?page=1#comment-80943
> 
> It's also convenient to let know all FSF associate members that your
> devices are OK in respect to freedom:
> 
> "5% off free software-friendly devices from ThinkPenguin" --
> https://my.fsf.org/
> 
> But for you it must be as you say, just a "minor issue" you "can't
> afford waste time" on it. For the rest of us, including founder of the
> free software movement and the GNU project, it's not:

I'm having a hard time taking you seriously. Wake me when you've 
actually contributed something of significance and aren't just trying to 
undermine those working on solving these problems. For some of us it 
isn't about financial gain. We actually want to see 100% free hardware.


> 
> On 16.08.2016 22:52, Richard Stallman wrote:
>> On 16.08.2016 17:14, Tiberiu-Cezar Tehnoetic wrote:
>>> IMO, we should teach users to avoid this ambiguous term. Instead of
>>> "free software friendly", they should use the term "compatible with
>>> fully free operating systems" if the hardware is compatible with free
>>> distros endorsed by FSF.
>> 
>> I agree.  The FSF could post something about this.  I will suggest it
>> to the campaigns people.
>> 
>> In the long term, I hope that our endorsement, RYF, will set a
>> standard and that people will come to see other terms, without clear
>> and strict definitions as inadequate.



More information about the Dev mailing list