Code review comment for lp:~mir-team/mir/attestable-timestamps-server

For me, this is the remaining blocker:

> > > > Is libmircookie really something that will need to be released in sync
> > with
> > > > every Mir release? Or should it be a separate project?
> > >
> > > I dont have a clear answer for this one. I just imagine since the project
> is
> > > pretty small it makes it easier to keep it in lp:mir atm? Will have to
> poke
> > > raof about that one.
> >
> > A further thought, the header is dependent on MirCookie (introduced here
> into
> > include/common/mir_toolkit/common.h). That makes client code (compile time)
> > dependent on libmircommon-dev.
> >
> > As we're introducing this dependency on mircommon, could we just put the
> code
> > into libmircommon and not a separate library?
> >
> > Or should MirCookie be defined in libmircookie-dev and not libmircommon-dev?
>
> Well I dont see why it couldnt be defined in libmircookie-dev. I think *if* we
> keep libmircookie-dev we should move the MirCookie into it (unless Im missing
> something for why this isnt possible). If we move to libmircommon-dev then
> we'll keep it :).
>
> Need to talk with RAOF about it as well