--- On Thu, 11/24/11, Rob Weir <[email protected]> wrote: > > This is the surest and easiest way forward. If you want to > volunteer to > implement an alternative approach then feel free to make a > proposal for > what you want to do. >
Oh. I solved this issue already for PedroOffice. I will amuse myself letting you implement your "surest and easiest" proposal. Pedro. > > > >> > I'd be very careful before assuming being > discussed > >> here qualifies for that exception is being > discussed > >> here. I'm with Pedro on the prudence side.) > >> > > >> > > > > Now there are two issues that have got mixed up in > this > > thread: > > > > How we will support/use hunspell but likely also nss, > > mozilla, rhino and saxon. Officially We can use the > > binaries, and no one argues about that. I had doubts > > about an optional script (configure) that will > download > > the sources, patches them and produces such binaries. > > I saw this as a way to circumvent the restrictions on > > copylefted code but I have since then agreed that > this > > can be something reasonable as long as this support > is > > optional (it is) and the code doesn't touch SVN (it > > doesn't). On the long run I think there is also > consensus > > that we want replacements for these. > > > > The other issue is what to do about dictionaries. You > > have been implying we can carry MPL code and by > extension > > dictionaries in SVN and that is simply false. > > > > AFAICT there are no more infringing MPL headers or > code, > > in SVN, are there? (I recall I removed nssrenam.h > uglily > > but I did it) > > > >> These questions were already raised, discussed and > resolved > >> many weeks ago. > >> > > > > And these questions will continue to be raised, > discussed, > > and resolved many times until we graduate. > > > > Part of what I did in my previous job was prevent > errors > > from happening and that is what I do here now, but I > would > > really prefer to spend more time on the code. > > > > Pedro. > > >
