--OgqxwSJOaUobr8KG
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

On Mon, Oct 30, 2006 at 01:05:06PM -0800, Doug Barton wrote:
> Simon L. Nielsen wrote:
>=20
> > Personally I think rm should do what you ask it to do - if you ask it
> > to overwrite a file which has multiple links, well... though luck. =20

>=20
> Sorry, I disagree. It's not always obvious to the user when a file has
> hard links, and I can't see any situation where the pre-recent-patch
> behavior (overwriting the actual file when using -P on a hard link) is
> the expected outcome.
>=20
> It's all well and good to say, "tough luck," but I don't think that's
> what our users expect.


How do you know what is "obvious to the user" and what "users expect"?
Have you done some surveys or researches? If not, than it is all relative
and happening inside _your_ head.

To speak in facts: How many users do you know that have complained about
rm(1)'s behavior before this thread was started (I mean the -P switch
when file has multiple links)? Or have lost their valuable data because
of "rm -P"-ing their files?

IMHO many problems arise when someone tries to please even the stupidest
user by writing a fool-proof software. To me the beauty of Unixes is
that they are _not_ fool-proof, e.g. your are holding a real gun, you
should be carefull not to point it to your head and pull the trigger.

--=20
Vasil Dimov
gro.DSBeerF@dv
%
Shaw's Principle:
Build a system that even a fool can use, and only a fool will
want to use it.

--OgqxwSJOaUobr8KG
Content-Type: application/pgp-signature
Content-Disposition: inline

-----BEGIN PGP SIGNATURE-----

iD8DBQFFRvq5Fw6SP/bBpCARAmorAJ4h8c4wvQuf7LAS9Ow/2uoAhCr/GwCeO5O+
n8vqD+Mw/WKDn2312MdKjpI=
=v1c4
-----END PGP SIGNATURE-----

--OgqxwSJOaUobr8KG--