After version 6, Trisquel is moving to LTS only releases.

The same decision was made during the March 5th meeting with a similar
reasoning. See http://trisquel.info/en/wiki/2013-03-05-meeting for
details.

To summarize the meeting:

* Yes, we will do LTS only.
* We will open a user repository so the community can submit their backports (newer software like gimp 2.8, etc)

I always liked the idea of limiting to LTS versions and do backports when possible. I think that this can benefit a documentation/translation project since all contributors can focus on the same version and have more time to work on it.

First off- this definitely has my full support.

I didn't have a chance to read the meeting notes. I wish I hadn't missed this meeting. I actually do have concerns I think we should look into despite being an advocate of this approach. In any case the meeting does sum up exactly what I've been saying.

I'm not sure all of the reasons this is moving forward although it sounds like Ubuntu's leap to a rolling release and the lack of funding are a big part of it. And I am attributing the lack of time/development resources to the lack of proper funding.

However- that said this is really a great move. It'll push things forward and hopefully fix a lot of the problems we currently have with GNU/Linux. The issues which I'm hoping will be fixed effect a lot more than just Trisquel. They negatively effect distributions across the board. This really could make Trisquel one of the best distributions out there (if it isn't already).

Now the trick will be to get everything right. Some of the packages might be tricky to back port and there could be issues introduced that might make this initiative backfire. Hopefully though my concerns are unfounded and things go smoothly.

I think we need to do a little more basic testing of the concept with Trisquel 4 and see if this approach can work. As long as there aren't major hurdles for backporting HPLIP or issues with the crazy way graphics are handled with X/the kernel/etc though I think this will work.

Like Ubuntu's LTS, Debian roughly releases a new version every two years. Could Trisquel come back to it (even though gNewSense wants to do the same)?

Debian is more stable, raises less freedom issues and is more in sync with the free software development community. I mean Ubuntu basically is the only player insisting on Upstart (the rest of the world jumped or plans to use systemd), Unity (although most Trisquel users seem to prefer GNOME Shell or GNOME Classic) and, soon Mir (whereas every other distribution plans to jump on the Wayland train). For the last project, it really looks like Ubuntu does not know what it is doing. For instance, the Wiki was originally criticizing Wayland with plain wrong statements.

---Trisquel 6 MAYBE gets released around the time of Ubuntu 13.04-----

ubuntu 13.04 Raring Ringtail will be Release 18 April 2013[wikipedia]. that mean again Trisquel 6 @ a huge delay. :-(

A wise decision, in my opinion.

I even thought of proposing just that, in the previous discussions that I saw being held here. But refrained myself from doing so, because, not only am I not part of the developing team, but also, because I'm still waiting for Trisquel 6 to be released (exactly because it is based on a LTS version of Ubuntu) I'm still not sure if I will adopt this distribution or not.

I've been a long time Ubuntu user, almost since that distribution appeared. And, through the years, I've repeatedly had so many bad experiences with non-LTS versions, that I don't recommend any of those to anyone any more. And will only use one if I really have to.

Non-LTS versions of Ubuntu are often buggy and, definitely, not as stable as the LTS versions... (And I think there's also no need for them. Since, 2 years is, to me, a decent time interval to wait for a new version of an OS.) To base a version of Trisquel in one of those non-LTS versions is to be inheriting all of those bugs and to be possibly wasting a lot of time trying to solve them, which can be spared and better used if you just concentrate yourselves on the more stable LTS versions.

I've always thought that it is best to do little, but well done, than to do much, and badly (or not as well) done.

An excellent decision.

I hope this alleviates the development pressures that Ruben must be feeling.

I wholeheartedly agree with this decision. LTS 4tW!

Regarding Debian - I can see advantages and disadvantages to moving over. The important ones, as have already been pointed out in this thread, are systemd vs upstart, Wayland vs Mir, etc.

Perhaps the most important advantage Debain would have as a base, though, is the fact that they also [I]try[/I] to be a free software distro, so everything is de-blobbed already and they'll throw up [I]far[/I] fewer freedom issues than Ubuntu will, thus alleviating the pressure on the Trisquel team.

Disadvantages are the obvious out-of-date packages in stable. Not sure how to address this to be honest.

Thoughts?

I was also thinking about how the upgrades from one LTS version to the next go. Traditionally, Ubuntu can be updated from one LTS to the next as soon as it is released if the user manually does a release upgrade. It is only with the .1 point release and higher (like 12.04.1) that the update manager and update manager core will give the user an option to update from one version to the next.

With that in mind, should future Trisquel LTS releases be based on the .1 releases of Ubuntu so when someone updates from like Trisquel 6 to 7, the update manager will ask the user to update the day of the release?

Great. I totally support migrating to this LTS-only release method.

Just a thought on a matter that appeared later: since there is already a 100% FaiF GNU/Linux distro based on Debian (gNewSense, despite it's not ready yet for most users), wouldn't migrating Trisquel's base to Debian let alone people desiring a 100% FaiF Ubuntu-derived/like/whatever GNU/Linux distro?