Slashdot Mirror
Evolution Of Debian Package Dependencies Resemble Predator-Prey Relationships
An anonymous reader writes "Scientists have performed an ecological analysis of software packages in the Debian GNU/Linux distribution over time; they found that dependencies can be successfully modeled as a predator-prey relationship."
Forgive my adjacent moment to rant since we are discussing Debian:
The whole move to Gnome 3.x and now 3.2.x on Debian is a time bomb, especially with gnome-shell, gnome-session and gnome-desktop-data and their clusterf*** of two versions which break some packages between the 3.x and 3.2.x branches. Quality Assurance has gone downhill considerably over the past 5 years within Debian. I'm on my 11th year of Sid/Experimental daily consumption and the amount of times now I've seen kernel crashes, Xorg crashes and broken applications isn't decreasing, but increasing. I hold Linus and his underlings responsible for the 3.x/3.1 crap and 2.6.32+ junk, as well as Xorg and it's abortion known as XWindows but the package breaking and massive growth in piddly little packages with gir1.2- and circular dependencies in various projects really begins to grate on the nerves.
Phrases that include, ``It's free, the source is open now pitch in and fix it...'' are a joke. The tens of billions having been poured into the Linux Community, at large, and the ownership of these packages like they are your first born child turns my stomach. I like having two platforms to do scientific work on, [Linux and OS X] but I plan on spending more time on FreeBSD and less on Linux with the way Linux continues to move forward. Wake me up when Linus has a stable ABI and the ``binary blobs taints the kernel crap'' ends and perhaps Linux might gain 2% on the desktop.