Re: Your DISCUSS on draft-ietf-secsh-newmodes-05

[der Mouse <mouse%Rodents.Montreal.QC.CA@localhost>]

>> Can we just get confirmation: Is your intention that this should be
>> required for all SSH implementations, required for all
>> implementations that implement any mode defined by newmodes, or
>> required for all implementations that claim to implement newmodes as
>> a whole?

> In the near term, I believe that it should be REQUIRED for
> implementations of the newmodes document.

But what is an "implementation[] of the newmodes document"?

(1) Any ssh implementation after newmodes reaches RFC status?

(2) Any ssh implementation that supports any of the newmodes-specified
    ciphers?

(3) Any ssh implementation whose documentation claims support for any
    of the things specified in newmodes?

(4) Any ssy implementation whose documentation claims support for
    newmodes, without specifying exactly what portions of newmodes it
    supports?

(5) Some mix of the above?

(6) Something else?

This makes a difference to me as an implementor, and apparently to Ben
as well (though I suspect our reactions to "bad" meanings would be
different).  It's what I was trying to point out when I wrote that
newmodes qua newmodes was not a useful thing to "implement", that it is
really an umbrella under which to gather together a number of useful
things - a way to avoid a blizzard of 15 different drafts, one for each
cipher and each rekeying recommendation.

/~\ The ASCII				der Mouse
\ / Ribbon Campaign
 X  Against HTML	       mouse%rodents.montreal.qc.ca@localhost
/ \ Email!	     7D C8 61 52 5D E7 2D 39  4E F1 31 3E E8 B3 27 4B