On Sun, 16 Apr 2000, matthew green wrote:

> i've implemented an lchflags(2) (that does not follow symlinks, and
> can thus operate on symlinks).  it appears to work just fine.  however
> the chflags(1) man page claims that symlinks don't have flags?
> 
> chflags(1) itself will need some updating, but does anyone object to
> this new system call?

Why do we want it? (*) I thought that one of the big decisions at Berkeley
when making symbolic links was that they wouldn't have their own
permissions, and that only the permissions on the referenced node would
matter. ?? I don't have all of the elements of the arguement with me, but
it seemed like things just got really messy if people tried to rely on
permissions on symbolic links..

Take care,

Bill

(*) This is meant as a real question, not some sort of off-handed
attack. :-)