Fear and Uncertainty - Trisquel 7.0

The other thing I was thinking of was that if, in the future, Trisquel will return to Debian as its base - perhaps gNewSense could amalgamate with Trisquel - or at least share development resources, since both distros would be sharing a common base, both technologically and politically.

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA256

This news about Ubuntu 14.04 is indeed troubling.

To be honest, I'm not sure what the solution is - but I think it's pretty clear that it will become harder and harder to provide a free distro based on Ubuntu, as it becomes less and less free.

Gnome 3 is definitely the way to go for the default DE, IMO. I currently use it with the nouveau 3D driver on my nvidia card, but have used it successfully with an intel chip in the past. For users who have neither of these, perhaps software rendering of the gnome shell is the way to go?

The Debian question is interesting. Debian is certainly a superb distro, but basing on it presents a new set of issues to address. Chiefly, the extremely slow release schedule for stable. Whilst this is a positive thing for users who require an absolutely rock-solid stable distro, it can be a big disadvantage in the desktop space, where users generally desire the latest and greatest features. One of the advantages we have from being based on Ubuntu is that they base on Debain Testing, and do QA for us, essentailly providing an accelerated pathway for new packages and features to become relatively stable. We would loose that by basing on Debian Stable.

We could base on Debian Testing, which would bring us new features more quickly, but that would place a huge burden on the Trisquel release team to do a huge amount of testing, QA and patching, prior to each release, essentially doing the job that Ubuntu currently do - a job which I sense they simply do not have the resources to do.

Despite having an understanding of the predicament (I think), I don't actually have a good idea for a solution at this time - does anyone else have any suggestions for a good distro on which to base future versions of Trisquel? We'll need something current but also not too dependant on non-free software, so that the effort required to liberate it is not too great.

Of course, I don't speak on behalf of the Trisquel project, and am not associated with them in any way, except for the occasional donation. Their view may be that it's perfectly appropriate to continue basing on Ubuntu in the future.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.17 (GNU/Linux)

iF4EAREIAAYFAlHG7MkACgkQgijxUCZnvluBVwEAhl8pngslW+ueXXTKVNti1UKr
CzMDJdDhdr2QJLEZ4YIA/1nZoWG3P5n91dk8AMnD7oCW0QKH8GWi5byf91wjHun2
=Ph3k
-----END PGP SIGNATURE-----

I think a Debian stable + an optional backported Linux-libre kernel & updated mesa driver stack could work well - kind of what the Ubuntu LTS point releases are doing.

Sure, the programs might be a bit dated, but you can always add program specific PPAs.

An interesting (although impractical option for Trisquel) would to have a Debian sid based 100% libre distro. Now that would be cool. Effectively rolling release, fresh, but not bleeding edge software, and a guarantee of 100% free software. I suppose Parabola already fits that model, but from what I've heard anecdotally, Debian sid is actually more stable than arch and has a richer repository of programs.

Still, I think the Debian stable + optional updated core components could be a nice way to go.

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA256

You might be on to something there. But IMO there would have to be a few more backported packages than just the kernel and some drivers. A web browser, for example.

PPAs are great, but we don't want users relying on them too much, because they often contain non-free software, or software that wouldn't otherwise be allowed in a distro like Trisquel.

Actually, Debian is already pretty much free. No binary blobs in the kernel, no non-free software in the default repos. The main reason given by the fsf for not endorsing it is that:

"Debian also provides a repository of nonfree software. According to the project, this software is “not part of the Debian system,” but the repository is hosted on many of the project's main servers, and people can readily learn about these nonfree packages by browsing Debian's online package database."

My reading of this is that someone could create a free distro by simply mirroring the free repos from the Debian servers, and re-branding with a different name.

If Trisquel took this approach, there really would be minimal work for them to do. It would be limited to back-porting anything they feel was outdated in Debian Stable.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.17 (GNU/Linux)

iF4EAREIAAYFAlHG9bcACgkQgijxUCZnvluYYgD/TLB294F/sT7e88r5EN7vI/pL
v3ugrhZpiQkyXm5HitIBAKMCLJaKg94gkZFOXJAjV8pvl81fqjPiPRr9H2d37zPT
=uDF+
-----END PGP SIGNATURE-----

Trisquel already deviates from Ubuntu's choice of a default DE. I doubt this change will cause many problems.

Also, isn't Mir free software?

On 23/06/13 21:07, ahj wrote:
> So Ubuntu 14.04 will have Unity 8.0 and Mir as their default DE and
> display server. What will this mean for Trisquel?

I'm guessing X11 will still be packaged. Otherwise Kubuntu et al. are in
a lot of trouble.

> I'm assuming Ruben will want to keep GNOME 3 as the default DE - how
> is he going to implement 2D support?

GNOME Fallback is still being maintained at the moment, AFAIK. I recall
it being released with 3.8.

> Will Trisquel 7.0 have to be based on Xubuntu or any other *buntu
> derivitive that supports Xorg?

The GNOMEbuntu might be one to look at as well.

On 24/06/13 00:12, onpon4 wrote
> Also, isn't Mir free software?

Yes, it is GPLv3.

Andrew.

I run MATE from the official PPA in Trisquel 6 and it has worked fine for me. The default implementation is rough around the edges and I decided to purge it out and replace with MATE.

It works well and the low system resources are why it is my preferred DE.

What about XFCE4 ?

In the effort to get free software to the widest amount of people possible, Trisquel SHOULD focus on one of the lighter desktop environments like LXDE or XFCE not only because their requirements are low but also because the system is not likely to change its functionality in any extreme way any time soon. Right now, Trisquel is using what is essentially a modified Gnome 3 when it could just as easily use an unmodified XFCE to get to the same result.

If I had any say, I'd remove any link Trisquel has to Debian and instead have it be based on Fedora. It doesn't take much to make Fedora free software (Freed-Ora is available) and the repositories generally contain faster-running programs which are also at a more recent version than what Ubuntu or Trisquel currently offers.

My understandment of project health is as following:

1.| The creator - Rubén - has his very own life difficulties in Galicia region of Spain. Perhaps his brother Santiago, who commited a lot to what Trisquel represents now, too.

2.| Progress made on Sugar environment, additional Dextrose tools and OLPC hardware enablement (perhaps including reverse-engineering?!) during his dayjob makes him more fulfilled overall (unquestionably), because the next generation of children should be aware of Software, Societal & Digital Freedom in regards with mesh (ad-hook) networking as well.

3.| I agree on hackers' resource waste concerning gNewSense vs. Debian project, the only logic in this is that the Debian GNU/Linux guys do not officialy support and pre-build packages for MIPS computing architecture (thou am curious about whole MIPS know-how in hands of Imagination Technologies corp. and blobbed pre-requisities for their upcoming Warrior chips; still waiting for more proper revisions of China designed Longsoon 3A / 3B units inside of worldwide available SoC-s.

4.| Toutatis with LTS point releases will be sufficient to my needs for the foreseeable future, although users of nVidia graphic cores along with Nouveau KMS drivers are dying and longing for the most current software stack, either because of important regressions / bug fixes, added features or enhanced stability and drawing responsivenes (do not get me wrong guys, Intel driver is just way more mature and time tested).

5.| Myself am no expert in CLA-s and license exploitation, but I expect more lock-in ahead with Ubuntu upstream since more and more Microsofties (senior staff) get introduced and involved from within - it is like cancer you know or mole rush (Barnes & Noble? Ruined, check. Skype? Ruined, check. Nokia? Ruined, check. Dell? Ruined, check. Hewlett-Packard? Ruined, check. And of course Novell/SuSE.. Double-check. Oh damn, anyone mentioned Yahoo, yet??) I was talking about introduction of Mir. In this case I am on par with devs employed by Digia, Intel, Samsung and others who strongly advocate and support Wayland / temporary XWayland.

6.| Focus on light-weight DE / WM-s is not worth it. If necessary, then I would go for Razor-qt or Openbox && dwm. Forget the LXDE, Xfce, MATE, Cinnamon, forking issue altogether. GNOME and KDE is for the mass being introduced in here, switching from Windows and Macintosh.

7.| FSF and Debian Project Leader should finish off the remaining philosophical incompatibility asap, because there is no chance about Red Hat giving up non-free firmware and tainted kernel in comminity led Fedora (Speaking of this, note that many of their employees use Windows on their boxes daily and just from inside of virtual machines they test, develop and maintain range of products - this was told to me by a guy who worked in their network lab for a longer period before - actually he was mine idol before swapping his honourable aims for easier to get greed).

8.| I am saddened by lengthy legal review processes inside AMD/ATI management and need for binary files in order to make use of acceleration and playback decoders. This can be triumphed only by unfair practices pushed against adoption of salvation standards like WebRTC, SPDY, WebGL, HTML5 and libre codecs e.g. Opus, Daala, VP9/WebM.

9.| There needs to be a social movement against these f**kers and establishment support by creating multiple political-activism groups in similar manner to Pirate Parties. Also, consumers should pressure the manufacturers to release specs and get rid of global-scale trade secrets as enforced by treaties like TPP agressively exported from U.S. elsewhere.

10.| Dear Trisquel community, I appeal on you, stop trolling and pushing camp dividing agenda. One schizophrenic "Linux" news portal and (in)famous test suite.. such opinion skewing is handy to big market players and marketers.

Footnote: I would like to thank Chris for state-of-the-art ThinkPenguin ultrabook, it is for those who have disposable annual income. Also, congratulations on getting second Respects Your Freedom sticker! Well-done.

https://www.fsf.org/news/a-second-fsf-certified-device-from-thinkpenguin-long-range-usb-wifi-adapter-with-atheros-chip

Post Scriptum: In case the project leadership sees no hope in new Trisquel releases based on Ubuntu, community should re-transform into Triskel with Kubuntu grassroots.

http://www.h-online.com/open/news/item/Kubuntu-plans-to-steer-clear-of-Mir-and-switch-to-Wayland-1897942.html

At the current time Qt is not accessible on GNU/Linux, we need to fix that