Hey jxself, can we get linux-generic-lts-raring and linux-generic-lts-saucy kernels?

14 replies [Last post]
t3g
t3g
Offline
Joined: 05/15/2011

The LTS enablement stack for Ubuntu 12.04.3 includes the Raring kernel/headers along with libdrm and xorg packages. The libdrm and xorg packages are in the Trisquel repos, but the Raring kernel (3.8) is still missing and it would be nice to have.

On the other end, the Ubuntu 12.04 repos also have the backported Saucy kernel (3.11) and headers available, but the libdrm and xorg packages have not been backported yet.

Would it be possible to get both of these into the Trisquel repos? I could understand going for the Raring packages first since they are more tested, but it would be nice to also have the option to use the 3.11 kernel to help extend the life of the Ubuntu 12.04 based Trisquel 6.

So what do you think Jason? Can you make this possible?

dadix
Offline
Joined: 07/01/2013

Here we have all the libre kernels we want:

http://linux-libre.fsfla.org/pub/linux-libre/freesh/pool/main/l/

t3g
t3g
Offline
Joined: 05/15/2011

We've had this discussion before where kernels matching the Ubuntu release may be the best for stability. Sure, you can always get the most bleeding edge kernel available, but the ones from Ubuntu are tested for that OS and all jxself or Ruben have to do is blacklist and remove specific non-free components.

jxself
Offline
Joined: 09/13/2010

I'd like to but I'm very busy. At the same time, anyone is free to work on modifying the appropriate deblob- script [1] so that it successfully deblobs the Ubuntu kernel (the deblob-check script from the same place can help in identifying areas that need investigation.) Just grab the Ubuntu kernel source & get busy. People that feel so inclined to work on it should feel free to share results so that it can get into Trisquel. Just let me know somehow; I can get it into the Helper git repository. Otherwise, it's whenever I get time... whenever that may be. Sorry; there needs to be two of me.

[1] http://www.fsfla.org/svn/fsfla/software/linux-libre/scripts/

t3g
t3g
Offline
Joined: 05/15/2011

You pretty much summarized my concerns with this project and most free software projects. Some of the most important people in the Trisquel project (yourself and Ruben) simply don't have the time to donate the resources to keep it afloat. You are faced with realities of a "real job" being your primary and only focus.

This also reinforces the proof that for a FLOSS project to be successful, it needs a commercial backing (like Canonical, Google, or Red Hat) or be in the top tier of popularity (jQuery, Linux kernel, Python, Firefox) to be sustained through personal and corporate donations in a non profit organization.

akirashinigami

I am a member!

I am a translator!

Offline
Joined: 02/25/2010

jxself gives you clear instructions on how you can help, and you criticize him and the entire project? How exactly is that helpful? What are you trying to accomplish?

t3g
t3g
Offline
Joined: 05/15/2011

The "oh you need this done? here's the source code so do it yourself" free software mentality is very common and thrown around a lot. It is like you owning a car and your transmission goes out and someone tells you that anything is possible if you learn how to fix it and do it yourself.

The reality is that people have various skillsets and they may not know how to do it properly or at all. In Jason's case, he has the skills and has established himself in the past as one of the best members of the Trisquel team. According to him, it is very simple to just take the code, modify it, and compile the kernel based on scripts that are available. If it is so quick and easy and anyone can do it, then why is it a struggle for him?

I'm not knocking him at all and I am sure he is a great guy, but simply telling someone to do the work themselves is not an easy task. Heck, if one of us did put in the time and effort to create this kernel, who is to say that it will get included in the repos or that Ruben will even respond at all? There is a clear heirarchy in this project with one man trying to do everything and everyone else is allowed to help when its convenient for him.

jxself
Offline
Joined: 09/13/2010

"here's the source code so do it yourself"

I believe that mischaracterizes the nature of my reply. Mine was intended in a helpful way, to encourage others to participate -- no more and no less. Characterizing my reply in this way makes it seem more as if I was throwing it over the wall and saying "do it yourself" and leaving at that, which is not the case. Hopefully it is possible to appreciate that there is a slight but important difference between these.

"and your transmission goes out and someone tells you that anything is possible if you learn how to fix it and do it yourself."

Someone probably could fix it on their own if they knew how. If not, perhaps a friend that knows about transmissions might be willing to help as an altruistic act. At the same time, if that friend were to say that they were busy and couldn't (or would get to it when they could), I'm not sure it makes sense to criticize them for not immediately jumping to provide immediate gratis help. That said, commercial mechanics would also be available.

"it is very simple to just take the code, modify it, and compile the kernel based on scripts that are available."

I didn't say it was that simple but only provided a high level overview of the work involved. Each of those steps involves further work and substeps.

"If it is so quick and easy and anyone can do it, then why is it a struggle for him?"

I believe I answered that in the original message - It is not a technical struggle but I am very busy right now with my own personal commitments in my private life. I will work to get to it when I am able to do so, but can't promise to do it any earlier than I can.

"if one of us did put in the time and effort to create this kernel, who is to say that it will get included in the repos or that Ruben will even respond at all?"

I believe I answered that in the original message with "Just let me know somehow; I can get it into the Helper git repository."

Ruben's availability is not the key issue.

trisq

I am a member!

Offline
Joined: 09/03/2013

"This also reinforces the proof that for a FLOSS project to be successful, it needs a commercial backing (like Canonical, Google, or Red Hat) or be in the top tier of popularity (jQuery, Linux kernel, Python, Firefox) to be sustained through personal and corporate donations in a non profit organization."

I wouldn't say Trisquel isn't successful.

Trisquel is what it is, and it's pretty darn good. Pehaps you didn't mean it that way but this "proof" you suggest is unconvincing to me.

t3g
t3g
Offline
Joined: 05/15/2011

Trisquel has had success in getting the approval of the FSF, but there are some issues with the sustainability. Ruben has stated that in order for Trisquel to get his full attention, there needs to be more donations per month so he can potentially quit his job and work on this full time.

Speaking of donations, has there been any talks of the FSF potentially throwing money his way? Trisquel is one of the rare GNU/Linux distros that follows their ideals without comprimise and the FSF hopefully sees value in that.

dadix
Offline
Joined: 07/01/2013

I want to help Trisquel but nobody has time to teach a non-programmer user like me to do things which is bad because from what I see the more than 90% of Trisquel users are like me. So I put this topic https://trisquel.info/en/forum/package-helpers-video-full-tutorial-request but nobody has responsed.

I also want to know how to deblob a kernel or a non free program. Well I know there are some scripts there, I saw them , but I don't know how to use them.

G4JC
Offline
Joined: 03/11/2012

The "oh you need this done? here's the source code so do it yourself" free software mentality is very common and thrown around a lot. It is like you owning a car and your transmission goes out and someone tells you that anything is possible if you learn how to fix it and do it yourself.

Actually that's great if you're mechanically inclined because it's cheaper to fix it yourself and you'll learn how to do it, perhaps from a Hayne's manual. Which is more or less like the source code/blueprint but the publication is not GNU FDL. (Which brings up another topic, we should have DIY car manuals in GNU FDL!) :P

Anyhow, I don't know extreme amounts of code either but I can usually figure my way around. Taking for example the latest deblob script:
http://www.fsfla.org/svn/fsfla/software/linux-libre/scripts/deblob-3.9
It seems to be written in bash and presumably is run on the source code of the kernel prior to a compilation, or custom build. One of the main reasons for me not attempting this is my internet is way to slow to download the entire kernel source.

There may also be some tutorials which will be similar to Trisquel found over at YouTube.
Hopefully that will help point others in the right direction. :)

t3g
t3g
Offline
Joined: 05/15/2011

Actually, if you want to bring over the lts-generic-raring kernel (3.8) cleaned up from Ubuntu 12.04, you should use http://www.fsfla.org/svn/fsfla/software/linux-libre/scripts/deblob-3.8 as your reference.

Getting the lts-generic-saucy kernel may have to come later as there is no deblob script for 3.11 on that site.

jxself
Offline
Joined: 09/13/2010

Yes, there is a deblob-3.11. It lives in the same place, at http://www.fsfla.org/svn/fsfla/software/linux-libre/scripts/deblob-3.11

t3g
t3g
Offline
Joined: 05/15/2011

My bad bro. The directory listing for 3.11 at http://www.fsfla.org/svn/fsfla/software/linux-libre/scripts/ was not listed after 3.9 and missed it.