Submitted by “In 2007, OSAlert ran an article about OPIUM, showing how to cast apt-get installation problems (choosing which of several possible dependencies to install) as a set of pseudo-boolean constraints which could then be solved mathematically to give the optimal solution. We have recently adapted this technique to Zero Install, addressing some problems experienced by the Sugar environment (One Laptop Per Child) and allowing better integration with distribution packages.”
Doesn’t suse the first to integrate the SAT solver in yast and zypper ?
http://en.wikipedia.org/wiki/ZYpp#SAT_solver_integration
Maybe I am wrong, but the difference is ~6 months.
As I have said, suse is the first to integrate it :
http://git.opensuse.org/autodocs/satsolver/HEAD/historypage.html
According to http://git.opensuse.org/autodocs/satsolver/HEAD/historypage.html:
“SUSE’s hack week at the end of June 2007 turned out to be a turning point for the solver. Googling for solver algorithms I stumbled over some note saying that some people are trying to use SAT algorithms to improve solving on debian.”
So, I guess the SUSE work was based on the OPIUM work. I wonder which other systems have built on the same ideas?
(I write here due it seems original author of the post read here and comments’ on the original website are locked).
Any plan to use PackageKit while trying to solve dependencies with the local package manager?
The Sugar branch of Zero Install has experimental PackageKit support and the plan is to merge that to the main trunk soon:
http://thread.gmane.org/gmane.comp.file-systems.zero-install.devel/…