Quote: Samuel Verschelde wrote on Thu, 17 March 2011 09:14
> Well, that would be a real solution if we really wanted to flag those > packages > both as tainted and as non-free, as some people give more importance to > the > fact that it is tainted and others to the fact that it is non-free. Agreed. > For now, I would propose either to put that package in non-free, > explain to > users that non-free packages may be tainted too, and envision after > Mageia 1 > to add a new media if the current solution really doesn't work, and > maybe > require a meta-package from tainted OR put it in tainted, explain that > > tainted can contain non-free packages, and require a dummy package from > non- > free, as Anssi proposed (on a second thought, I think that second > option is > better). Why a temporary solution? The longer we postpone a proper solution the messier things will get. Also I really don't like the use of a meta or dummy package, that is even messier and confusing for the users. Since tainted+non-free packages will most likely have dependencies in tainted from a practical POV that would be the best place. A dedicated tainted+non-free repo would be the cleanest solution. Putting tainted+non-free in non-free is the worst solution both because of dependency issues and because it will be messy for mirror admins. So IMHO the choice is really between putting them in tainted and then describing tainted in the policy as being for ALL tainted packages (regardless if free or non-free) or else creating the dedicated tainted+non-free repo. > Can we reach a decision ? (add this question to the next packagers > meeting ?) TBH I don't thing IRC is suitable for decisions where people have to spend some time thinking about the consequences of various options. Email (i.e. here on the ML) seems better to me. > However, as the whole discussion seems to revolve around only one > practical > package, what would be even better would be convince and help upstream > to > solve the licensing issue (if that's feasible). This question was triggered by the first tainted+non-free package I came across (the 4th package I decided to work on). But so far I already found four tainted+non-free packages (even though two might be dropped if the FOSS replacement fully replaces them) and I don't think these will be the last ones. I haven't search for them, I only came acroos them as I wanted to package them up, and then discovered this repo issue. -- Mageia ML Forum Gateway: http://mageia.linuxtech.net/forum/