Current state of funtoo, roadmap and future?

[nrc]

Yes, the goal is to be more current.  It's a work in progress.   There's an issue open for this.  The problem is cleaning up packages that claim dependence on python2.7.

https://bugs.funtoo.org/browse/FL-7503

 

[drobbins]

@stilzchenno actually, that is not true. python3.7 is the standard interpreter in 1.4-release and this has been the case for a long time. We just didn't participate in Gentoo's 'pythocalypse' where 2.7 was forcefully removed from the tree prematurely resulting in a lot of chaos. We still have py2.7 and compat ebuilds for packages that need it.

We now have a Support Matrix online, which is here: https://www.funtoo.org/Support_Matrix

I have also added Wolf Pack Philosophy: https://www.funtoo.org/Wolf_Pack_Philosophy

I have also updated the main wiki page with this text:

"Funtoo is similar to Gentoo, but take heed -- it is also different! Our distinctives all stem from our different approach to community. If you are new to Funtoo, please familiarize yourself with Funtoo's Wolf Pack Philosophy. It will help you to understand what Funtoo is all about. And the second half of the page will explain some key technical differences in comparison to Gentoo."

So hopefully this will help people to understand that Funtoo is different from Gentoo, and that the roadmap is defined by users who contribute and 'howl' on the bug tracker. We do not have a 'team' of people who are building funtoo for you. WE (all the users) are the team.

[funfool]

On 9/16/2021 at 5:44 PM, drobbins said:

Also, frankly, I think that distros are now treated as commodities where somehow it is expected that we are supposed to 'prove to users' that they should use us instead of some other distro. Really, this is the attitude these days. And I don't really like this attitude. Asking for a roadmap is a perfect example of this attitude -- 

I Love Drobbins. First time I prepared to meet him it was intense for me. I expected a stuffed shirt developer with a attitude. Daniel greeted me with a honest & sincere smile. I then learned that Daniel was truly a great person to know & have as a friend in this lifetime. I still feel this way today.

I think that a roadmap is a very important vision in ANY business adventure. Nobody should be shunned for asking for such a thing. A computer is very important to people today.

I also think that BDFL is a great Honorary position for a leader. Those with ideas need help to implement them. If a BDFL is free to change course at any time ... that is not a community distro. Or a wolfpack whatever.

I send my love to Joost What another great linux developer. I hope to see sabayon back up soon. .... I know Sabayon is dead but the project ls living on.

A roadmap to future linux .... have you considered making a educational linux for home schooling?  In todays market Home schooling is going to go crazy ... what a road map to have.

[nrc]

The problem with traditional "roadmaps" is that they assume that you know exactly where you're going and how you're going to get there.  The typical business roadmap also includes a schedule for getting there.   None of those things align with the objective of being a community distro.  We have a general vision of what we want the distro to be and how we want to accomplish that, but exactly how that plays out depends on what the community wants and how much they can contribute.  The ability to pursue an important opportunity or idea when it arises is more important than ticking boxes for comparison shoppers.

With limited resources it's more important to be focused on what they community wants and needs than to create roadmaps that dictate a course that may not be appropriate.  Wayland is a perfect example.  Everyone has to support it because everyone else supports it and you have to check the boxes in the distro competition.  Who cares?  Nobody, really.  Until it makes my desktop better without creating any fresh headaches, Xorg is fine thanks.

As for BDFL, every pack has a leader.  The pack spots opportunities and communicates them, but it's the pack leader that says go or no go.   There needs to be an executive authority.   We've seen that community distros go terribly wrong when they become distro by committee.  The squeaky wheel gets the grease and and corporate paid developers can get very squeaky.  There needs to be a single decision point that will put the needs of the distro first.  Only those who trust drobbins to do that should join the pack.

[seemant]

Hi All,

I'm seemant.  I worked with Daniel on Gentoo a long long time ago, and recently I've using Funtoo and participating in its development a little bit as well.  The roadmap discussion is interesting and this thread has evolved a lot.

One of the things that I'm, in fact, looking to do within the next couple of months is to create a sort of roadmap for Funtoo.

Please consider this a request for comments and input:  in which areas would you like to see Funtoo improve and evolve? How can we make Funtoo better for everyone?

 

 

[drobbins]

Thanks, @seemant. Everyone, Seemant is going to be helping to build community efforts within Funtoo. I am so very glad to be working together with him again! If you are a long-time Gentoo user you will know that Seemant was my second-in-command, basically running the people, community and development coordination aspect of Gentoo during its golden era, and much of Gentoo's tremendous growth and success during this time was attributable to his insight and involvement in steering the project as it grew.

While Zogg started this thread with (in my opinion) overly trolling and negative opinions about Funtoo -- which me and some others found triggering -- there was some useful critical feedback mixed in, which I definitely reflected on and it would frankly be a mistake to ignore. I have tried to address much of the basic conceptual issues he raised (like differences from Gentoo) via the Wolf Pack Philosophy and Support Matrix (linked from main page of www.funtoo.org). But clearly there is still more work to do.

I've asked Seemant to specifically engage with our user community and help to collect feedback and build a roadmap for the project. So please know that Seemant is here to collect your ideas and feedback about what you are hoping to see in Funtoo. Please do not hesitate to reach out to him, either here in the forums or by chatting with him on Discord or Telegram.

Seemant will be collecting the ideas and working with me to incorporate them with my technical plans so we can have a public roadmap for the project, to help others get a sense for what is coming and also provide ways for others to get involved and influence the future.

Seemant, thanks for joining us and welcome 🙂

[dutch-master]

I'm not sure this is within the realms of the Funtoo dev's, but better conflict resolution for emerge would be a great start. One of the things putting me off Funtoo (and straight back to Devuan) was the issue that portage drops out the moment it encounters a conflict (like multiple packages for a slot, etc). For all its flaws, perceived or otherwise, apt has really good conflict and dependency resolution mechanisms build in. Now, I'm aware there's a different design philosophy behind emerge/portage against apt, but dropping a user back to the cli with a cryptic error message stating a conflict has occurred and masking a package may resolve it and effectively leaving a system in limbo while updating/upgrading is not a great way to experience Funtoo for less experienced users.

 

FYI: I've been using Linux as my daily OS since 2005 and experienced the "Dependency Hell" with Red Hat 8+9 and Suse (pre-Novell days), then I found Debian (Woody) and apt was such a relief. Stuck with Debian until systemd made it impossible (really-old Stable, IIRC Jessie) then eventually switched to Funtoo (2015) before finding Devuan a few years later (2018). Tried Funtoo again earlier this year, but still got issues with emerge/portage so now running Devuan once more. Thing is, I'm not smart enough to be a coder. If I were, I'd be a Funtoo dev for years. But as it is, my contributions are (severely) restricted to signalling something went wrong and that's pretty much it.

 

HTH!

[drobbins]

@dutch-master I totally agree. Ironically, I started Enoch (which became Gentoo) when I got frustrated by dpkg in Debian 1.3. Portage has since 'evolved' into something that has a lot of problems that dpkg originally had -- of being very good at telling you why it can't do something but not very good at resolving the problem it's complaining about, or providing a coherent error message when there is a problem.

Software tends to evolve this way -- it becomes more rigid over time. The cause is that most engineers will try to make the software more 'correct' and eliminate problems, but they often do this without advancing the architecture and just adding more internal checks. This actually can significantly reduce the usability of the software. This approach becomes much worse when the error messages displayed are not comprehensible by mortals.

Despite Portage's strengths, it does suffer from this way more than I'd like. We do have some technical work in progress that should offer some relief for this to some degree.

Edited December 7, 2021 by drobbins

[dutch-master]

@drobbinsThx for the feedback. I'll await future developments, I haven't given up on Funtoo just yet! (which is why I'm still here 😉 )

[seemant]

Thank you for your very kind words, Daniel.  It's great to be back working together again.

 

@zoggI think you brought up some great points, and I appreciate your unfiltered perspective.  I'd love to have a more in depth conversation with you, if you're open to it -- especially on Telegram or Discord or another more real-tme medium?

@nrcI appreciate your enthusiasm and positivity for Funtoo, and am looking forward to building the community with you and all on these forums (and the other media)

Edited December 7, 2021 by seemant

[nrc]

Welcome @seemant.  Happy to see you join the project.  

I think the challenge is communicating direction and progress in a way that is more fluid and responsive to development opportunities and user needs than the traditional "roadmap."   As we've seen in this thread, we can't expect casual users to get an good view of the direction and activity of the project by gleaning it from available information scattered across sources.

How do we communicate to users the value of the Funtoo approach over the mainstream roadmaps and checkboxes approach?  How do we stop trying to be all things to all people without coming off as arrogant, dismissive, or incomplete when something a user is asking for falls outside our focus?  This will arise more and more as the beard is trimmed.  

 

 

 

 

 

[morphmex]

I found Zogg's post disrespectful.  Daniel Robbins is the creator of Gentoo and now of Funtoo.  He is definitely an out of class person an Elon Musk of Open Source.  We need more respect for people who use their free time, which today is increasingly scarce for projects open to all.  I found this post offensive, specious and useless.  Open source and GNU Linux offers many operating systems that can offer what everyone is looking for.  You can use the derivatives debian arch, gentoo, fedora or Rocky linux etc .... I have been following Daniel for a long time because I think he has brilliant ideas but he must respect his times.  For my part, I would like to thank Daniel for what he does, for his altruism and his desire to share his culture, intelligence and genius with those who deserve it.  Thanks Daniel and best wishes for a Happy New Year to you and your family.

Edited January 1, 2022 by morphmex

[drobbins]

23 hours ago, morphmex said:

I found Zogg's post disrespectful.  Daniel Robbins is the creator of Gentoo and now of Funtoo.  He is definitely an out of class person an Elon Musk of Open Source.  We need more respect for people who use their free time, which today is increasingly scarce for projects open to all.  I found this post offensive, specious and useless.  Open source and GNU Linux offers many operating systems that can offer what everyone is looking for.  You can use the derivatives debian arch, gentoo, fedora or Rocky linux etc .... I have been following Daniel for a long time because I think he has brilliant ideas but he must respect his times.  For my part, I would like to thank Daniel for what he does, for his altruism and his desire to share his culture, intelligence and genius with those who deserve it.  Thanks Daniel and best wishes for a Happy New Year to you and your family.

Thank you, morphmex. Very kind words. I do think Zogg had some decent points but also with some acidity mixed in. Thank you for standing up for me, sometimes I don't realize when there is a bit too much acidity. It is definitely confusing when it is mixed with feedback that is attempting to be constructie. We are currently a small (but rapidly growing) project and I have tried to not just ignore Zogg's feedback, but I will gladly ignore the bad attitude I felt from him at times 🙂 As you say, there are a lot of projects to choose from and it is wrong to have a sense of entitlement from a volunteer software project.