Re: CVS commit: src/bin/hostname

[Thor Lancelot Simon <tls%panix.com@localhost>]

On Thu, Jul 25, 2013 at 08:23:26AM +0100, David Laight wrote:
> On Thu, Jul 25, 2013 at 06:30:54AM +0700, Robert Elz wrote:
> > 
> > The hostname was always intended to contain the globally unique textual name
> > of the host - someone said that it couldn't have always been a FQDN, as they
> > didn't exist when the hostname syscall was created, and that's correct.  
> > But 
> > back then hostnames (including before even ".arpa" was appended) were
> > registered and intended to be unique (it was the growing absurdity of
> > continuing that in a flat namespace that led to the development of the DNS).
> > 
> > Using a (trivially derived) nickname as the hostname instead of the actual
> > (unique) hostname is simply stupid.
> 
> There are many (well some) systems that don't have a FQDN, and where there
> isn't one that can be assigned.

Then on such a system, the hostname should (and can) be a short name.
What's the problem?  What's this have to do with tarting up hostname(1)
with a bunch of options from Linux?

Next we'll have Linux's jekyll and hyde ps(1).  Please, no.

Thor