Following a very short discussion on IRC, here one idea for
after Gobo next release, sometime in the future:

* Enable people to change their FHS system to a Gobolinux system.

Of course this process would have to be assisted heavily by
scripts.

Also

- Warn that this is a procedure which is only for advanced users (in
a sense that they are able to solve their own issues) so that they
dont come and overflow Gobolinux with questions specific to
their environment (they already have LFS for this)

- Warn that this is a process WHICH OFTEN WILL NOT WORK OR
LEAD TO PROBLEMS. However, it is a process which should
work.


Some advantages of this:
- Goboscripts can be improved because more people would use it
- An ecosystem of Gobolinux--like OSes would be built, this in
effect being a lot more bigger an impact on the overall design
approach of Linux OSes right now. (Who knows, maybe other
distributions will offer their users the Goboway too)
- More people could be reached this way than with GoboOnly.
- People would learn more too, which is always good. Knowledge
is a great tool.


You might say, why would they not download Gobolinux and install
it? This question is a bit hard to answer. People often like to
explore. If something is offered to them, they wanna try it.
And following this way, they could get convinced that the
Goboway is a lot cleaner and logical than FHS builds.
Plus it would be a little bit of advertising for Goboway too,
even if it sounds like a lot of work ... :)

"Moreover, as you said why should peolpe try to convert their systems to the gobo way when they feel comfortable with what they already have, and without the certainty that their system won't become a mess (if not unusable?)."

But if everyone would be happy, they would never need to come to Gobolinux. :)
It is a very different philosophy though, i think the Gobolinux way is in essence a lot cleaner.

The unusable aspect for sure would have to be adjusted so that it becomes usable. I think this is a lot of work, true,
but I think fundamentally it should be doable.

"but I believe a better way is to let people know the rootless installation script, which combines the gobo
way with the security of a sandboxed installation."

Maybe :)
Personally though, I am more interested in design and beauty than security, (Security seems
to become so obsessed.... reading kerneltrap.org about the talks AppArmour vs ... i forgot,
selinux or something. And it adds complexity to the system, but not really a lot of beauty...)

"Btw, since July (when I first discovered Gobolinux) I've seen more people joining the community"
Yeah totally agreed. Almost 3 years ago there was a lot less "user-traffic" :)

"maybe this has also to do with the forum's opening?"
Not sure but could well be. Although, I noticed the increase of users on IRC even when there
was no forum available, so i think these two don't totally correlate to each other.

"Of course we're not at the level of other distros yet, but surely we're growing fast.
We're just 5 years old, remember? :D"

Sure but I think more it would be a nice long-term goal.

I think the whole recipe building and package needs more work so that it gets
a lot simpler and maybe automated too.