I'm sorry, but it *IS* actually a clever thing to
intentionally break the interface whenever you make
substantial changes to the inner gears of whatever programs
you maintain.
One could argue that changes to the API should be kept to a
minimum (and I tend to agree with that), but if you do change
the semantics of a software component, then you've got to
break your own software to avoid having to chase unexpected
bugs due to different semantics, whenever other components try
to talk with the modified module.
It's plain and simple and the core Linux maintainers know this
very well -- and this may be the reason (note that I'm not
actually saying that these were *the* reason) why we've had a
major breakage of file systems: the semantics or functionality
of the FS seem to have changed. So, there are now only a few
file systems working (NFS, ext2, AFAIK).
IMO, it is better to have software that does NOT work for a
known reason (e.g., API change) than to have software that
DOES work for an unknown reason.
[]s, Roger...
--
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
Rogerio Brito - rbrito@iname.com - http://www.ime.usp.br/~rbrito/
(still an) Ugrad. Comp. Science student - "Windows? Linux and X!"
Nectar homepage: http://www.linux.ime.usp.br/~rbrito/opeth/
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/