Best free software phone strategy

[Nico Rikken]

Dear Paul,

Thanks for making this public, that's a good practice. I've subscribed
to the mailinglist myself, as I recognize I lack insights on the current
Replicant developments.

On do, 2015-06-18 at 11:15 +0200, Paul Kocialkowski wrote:
> > Your post regarding Fairphone 2 [1] was linked in a forum thread on the
> > Fairphone 2's announcement. [2] I've been interested in getting free
> > software running on devices for a while and although I hope Fairphone
> > will succeed in offering freedom, although I find it hard to believe.
> > 
> > [1]
> > http://blog.replicant.us/2015/04/whats-happening-or-not-at-replicant/comment-page-1/
> > [2]
> > https://forum.fairphone.com/t/fairphone-2-announcement-discussion-on-new-fairphone-specs/6579/74
> 
> Good to see that it had some impact among the Fairphone community.

Although I'm weary that was in vain for this Fairphone version.

> > I hope you can answer some of my questions:
> > 
> > 1) Will the Qualcomm SoC with unlocked bootloader and without signature
> > checking be able to run current free software?
> 
> It allows to run a free system, but the bootloader itself is impossible
> to free, which I see as a fatal flaw when it comes to software freedom.

Ultimately yes, and as more free alternatives exist this is hardly
acceptable.

> > 2) Is Google still contributing back to the free software version of
> > Android, or have Google's versions drifted to far from the free software
> > base?
> 
> Google has been releasing the free versions of Android through AOSP
> based on their internal code, not the other way round. As long as Google
> keeps doing it, a free recent version of Android exists. Note that they
> stopped maintaining useful utilities as those get replaced by their
> proprietary counterparts.

Luckily the offering via F-Droid is decent, as I've been able to replace
most applications on my own Android phones with free software.

Is it feasible for Replicant (or any project) to keep up with Android
developments if no modifications can be pushed upstream?

> > 3) How would you compare Ubuntu Touch with Replicant OS? I consider the
> > plain GNU/Linux kernel in Ubuntu Touch to be a benefit as it prevents
> > segregation of development efforts. Replicant seems to be more free,
> > although I'm weary about driver blobs.
> 
> Ubuntu Touch uses all the same proprietary blobs as CyanogenMod, despite
> being based on GNU/Linux. This is also true for FirefoxOS (minus the
> GNU/Linux part). On the other hand, Replicant is a fully free system.

But would these OS's be able to take the free drivers from Replicant to
offer a free implementation for Replicant-supported devices?

Considering the number of 'missing' and 'missing without non-free
firmwares' regarding device support, usability can be questioned. Then
again I'd first have to experience the current state of development to
make and judgment.

> > 4) I guess getting drivers to work would be most important to forward
> > free software on devices. But as has happend with the Vivaldi Tablet
> > (Spark KDE Tablet) I'm weary that after the reverse-engineering efforts,
> > upstream electronic change requiring development to start all over
> > again. Is there a strategic guide for selecting targets based on say
> > manufacturers or architectures? (I might be able to use my electrical
> > engineering background for reverse-engineering if I find the time, but
> > I'd rather contribute for the long-run).
> 
> This is especially something to take in account when manufacturing a
> device. Currently, many chips good chips have been available in large
> quantities for a long time and widely-spread devices are never too hard
> to find. 
> 
> Of course, we should avoid working on chips that have a very limited
> time span, but we haven't had any problem with that before since
> Replicant mostly supports popular mainstream devices that were produced
> in millions.

Is there say a list of currently available (cellphone) chips offering
being supportive of free software? I get the sense that despite best
intentions, the Fairphone project is in the dark regarding hardware
issues related to free software.

Kind regards,
Nico Rikken

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 213 bytes
Desc: This is a digitally signed message part
URL: <http://lists.osuosl.org/pipermail/replicant/attachments/20150619/dff2718f/attachment-0001.asc>