[1640] in Kerberos_V5_Development
Re: preliminary appdefaults patch
daemon@ATHENA.MIT.EDU (Sam Hartman)
Mon Aug 26 19:11:32 1996
To: "E. Jay Berkenbilt" <ejb@ql.org>
Cc: kenh@cmf.nrl.navy.mil, eichin@cygnus.com, krbdev@MIT.EDU
From: Sam Hartman <hartmans@MIT.EDU>
Date: 26 Aug 1996 19:10:57 -0400
In-Reply-To: "E. Jay Berkenbilt"'s message of Mon, 26 Aug 1996 17:32:20 -0400
>>>>> "E" == "E Jay Berkenbilt" <ejb@ql.org> writes:
E> Do kerberos clients have both long and short versions of
E> options, like most gnu commands? Then you could have kinit -f
E> or kinit --forwardable and kinit -N or kinit --not-forwardable
E> (or some such).
Nope, they currently do not. I don't think we want to roll a
copy of getopt into krb5util--the tree is too big already. Of course,
we've probably already got two or three copies of getopt floating
around in there, knowing Kerberos.
E> Also, just a word of warning: a lot of
E> DOS/Windows types out there do not have the appropriate
E> software installed in their brains to handle case-sensitive
E> options. Even in my short stint of working with PC users, I've
E> run into this several times. Too bad something like kinit +f
E> vs. kinit -f couldn't have been natural, but I'm not really
E> suggesting that since the use of - to introduce options is so
E> deeply ingrained (and doesn't require a shift key like +
E> does).... Just my $0.02...
Yes, + vs - was my first idea, and I discareded it because it was too strange. The warning is appreciated, but is too late: almost every Kerberos program is using upper case options for something already.
E> -- E. Jay Berkenbilt (ejb@ql.org) | Member, League for
E> Programming Freedom | lpf@uunet.uu.net, http://www.lpf.org
