[892] in Kerberos_V5_Development
Re: Proposed Kerberos V5 Password Changing Algorithm
daemon@ATHENA.MIT.EDU (John Gilmore)
Tue Mar 14 15:03:32 1995
To: "Theodore Ts'o" <tytso@MIT.EDU>
Cc: krbdev@MIT.EDU
In-Reply-To: Your message of "Mon, 13 Mar 1995 14:06:41 +0500."
<9503131906.AA19331@dcl.MIT.EDU>
Date: Tue, 14 Mar 1995 12:01:50 -0800
From: John Gilmore <gnu@cygnus.com>
> This command indicates that the client is willing to accept
> MIME-enhanced textual messages in place of the standard NETASCII textual
> messages.
>
> If this command is not sent, the server MUST send all textual messages
> using NETASCII, with <CR><LF> used as a line breaks, and line lengths no
> more than 80 characters. If this command is sent, and the server
> returns a status code of SUCCESS, the server MUST send textual messages
> as MIME-enhanced textual messages. The server may refuse to send MIME
> messages by sending a status code of CMD_UNKNOWN.
We're defining a new protocol. There's no point in making options in
it. If you think MIME is the way to go, let's make it standard.
Document it clearly so that "simple" clients can just check or skip
the MIME headers and printf the message. It's not clear to me what
a "complex" client would do -- fork off a copy of metamail to print
an error message? That sounds highly error-prone to me!
This part of the spec needs a LOT more definition.
John
