Elive 2.0: Where Debian meets Enlightenment

Posted Mar 18, 2010 4:22 UTC (Thu) by BenHutchings (subscriber, #37955)
Parent article: Elive 2.0: Where Debian meets Enlightenment

Under the hood lies Debian Lenny (5.0.3) with a Linux 2.6.30.9 kernel. ... Elive is pretty much a one-man show

So, already out of date and lacking any security support.

One of the consequences is that users can download the distribution for free, but they have to pay (Elive calls it a "donation") to install it to a hard drive.

I'm looking forward to seeing Debian receive its share of this 'donation'. After all, Debian and upstream developers have done 99.9% of the work involved in this distribution.

I wish people would work to improve existing distributions instead of spending the huge amount of time and effort needed to create a derivative which they can never support as well. It is a real waste.

Elive 2.0: Where Debian meets Enlightenment

Posted Mar 18, 2010 10:48 UTC (Thu) by eru (subscriber, #2753) [Link] (1 responses)

But maybe some very different idea, like this light-weight E17-based desktop, cannot be properly explored within the constraints of a mainstream distribution? For example, management tools in them might assume large libraries and daemons that you want to avoid. Not that I have too much experience in this, but I have recently been playing with the SUSE Studio for my edification (don't worry, I'm not inflicting my light XFCE-oriented spin of OpenSUSE on the world, except maybe to some family members...), and found that identifying and avoiding the inclusion of probably-unneeded stuff is pretty hard.

It also seems to me the Elive author already tried to avoid reinventing everything by basing it on Debian, instead of building a distribution from scratch. So I disagree about this being waste. It is an interesting experiment.

Elive 2.0: Where Debian meets Enlightenment

Posted Mar 18, 2010 14:21 UTC (Thu) by tajyrink (subscriber, #2750) [Link]

Debian has much more fine-grained dependency structure than openSUSE. I've also noticed in openSUSE that it's hard to avoid bringing hordes of unneeded packages simply because there haven't been enough development time put to dependencies to fine-grain them. In Debian, the situation is far more better probably thanks to it being developed by thousands of community people who have the time to spend on individual packages. And policies/processes enforce them to do that if they want to keep their packages top-notch. It is usually relatively easy to avoid masses of unneeded packages being installed in Debian, at least compared to some other distributions.

What is achieved by this kind of derivatives development is concentrating on the issues one wants to achieve instead of spending half of the time with learning and handling processes of the upstream distribution. In the end, though, the processes are what would make the effort lasting instead of slowly dying when/if the development in the derivative is ceased. It can be helped if the derivative has clear "diff" from the mother distribution that may help others to relatively cleanly understand what was done and how it can be incorporated elsewhere.

(yes, I like Debian...)