> After some thought, I think this would be a GREAT case study to > demonstrate why Linux needs to have certain standards in place that are > COMMON across all distributions. The efforts that were spent on getting > FOSSology running on another distribution of Linux (from Debian to Redhat > / Centos) should help open some eyes that this really is a problem
As bobg pointed out, this is what the LSB is for, and we follow it as best we can. The problem is that a lot of the behavior isn't standard across the distributions and so the LSB can't define it yet. The LSB doesn't get to dictate standards, it's a "trailing edge" effort where _after_ things are mostly common across distors they can be added to the LSB. This means that to create a new standard for something you need to invent a solution to the problem, get the majority of distros to adopt it, properly specify it at the interface level, and then it can be added to the LSB. But I'm not even sure that lack of standardization is the biggest hurdle to getting fossology running on other distros, I think it's mostly the lack of packaged dependencies. The reason why it has been easy to support Debian is that everything was already _in_ Debian. Fedora is nearly as easy, but RHEL is proving to be difficult. But I think progress is being made and we should have "cookbooks" of where distro users need to get things from soon. -- Matt Taggart [email protected] _______________________________________________ fossology mailing list [email protected] http://fossology.org/mailman/listinfo/fossology
