Sending publickey subsystem to last call

[Sam Hartman <hartmans-ietf%mit.edu@localhost>]

Hi.

I found one minor problem in the newest version of the pbulickey
subsystem draft.  The text added to deal with spurious garbage is
inadequate because it is optional for client implementations.  To
guarantee interoperability this needs to be mandatory.

+   Implementations MAY use the first 15 bytes of the version packet as a
+   "magic cookie" to avoid processing spurious output from the user's
+   shell (as described in section 6.5 of [4]).  These bytes will always
+   be:
+
+   0x00 0x00 0x00 0x0F 0x00 0x00 0x00 0x07 0x76 0x65 0x72 0x73 0x69 0x6F
+   0x6E
+
+4.  Public-Key Subsystem Operations



Please s/MAY/MUST/.  Don't submit a new version now; I'll let you know
if/when I want a new version.  If this is the only change, I'll handle
it in an rfc-editor note in the protocol action message.