RE: RE: [Lsb-cvslog] si/build/patches coreutils-i18n.patch, NONE, 1.1 coreutils-i18n-sort.patch, NONE, 1.1

>> Consesnsus has been that we would only put these in when the upstream
>> accepted them.

>If one of the primary purposes of the SI is to serve as a testbed to
>test LSB applications, doesn't this mean the SI needs to be completely
>compliant?

As with so many things, the si has multiple purposes.
It's also our example implementation, showing how
an conforming runtime could be built from upsteam
pieces.

But the app certification piece is probably more important.
In general, I'm starting to lean towards fixing, after
spending some time on the "leave it broken side".  The
view is changing because it looks like apps are actually
on the road to happening.

Bet I suspect that issues affecting internationalization of
/commands/ (not library interfaces) have pretty much zero
chance of affecting any of our miniscule number of conforming
applications. This is where the coreutils stuff falls...