Thus spake Jason Thorpe ("JT> ") sometime Yesterday...

JT> > I have a question, though -- since we're migrating to 2.0, the major
JT> > bump represents, to me, a new starting ground, a new plateau, more
JT> > or less fresh and clean.  Do we intend to be backward compatible with
JT> > 1.x, or can we just declare a cut-off point and move forward with a bit
JT> > less baggage (such as, i.e., libc.so.12)?
JT>
JT> Not without screwing over basically the entire user-base.

Not to be offensive, here, but considering all the changes which have
happened since 1.4 up to now, that's kind of slowly happening anyway.  The
COMPAT_1* options seem pretty gratuitous, honestly, and maintaining
complete compatibility going backwards, well, how long will it be before
it is deemed to be too much of a stretch to maintain?

The implied attitude as with most software is "either keep up or lose
support", in spite of all the compat* stuff.  Obviously, this is more
or less reasonable, as unless people are willing to volunteer to try
to maintain full backward compatibility, some stuff is just going to
fall off the shelf.  The farther forward we go, and the farther back
we try to reach, the more resources it's going to consume, and the
more likely it is to be decided that it's just not worth the effort.

There must be something I'm missing here, or so I am hoping.

				--*greywolf;