Re: WG Chair Copyediting for draft-ietf-secsh-break-02

[Bill Sommerfeld <sommerfeld%east.sun.com@localhost>]

> "implementation-defined."  Hmmm.  What I am really trying to say here is
> that if the coder has no control over the break length, a break of any
> available length should be sent.  That is, if you only available command is
> to tell a chipset to "send a break," that is what you should do.  How best
> to word that?

I was thinking about it from an outside-the-system-box perspective.

The ssh(d) module may can't control it, but some other piece of the
system (serial chipset, device driver, serial i/o subsystem, etc.)
may be the entity which controls it in an
implementation-of-that-piece-of-the-system-defined manner.

That then becomes an implementation-defined property of the system as
a whole.

Anyone else got a wording suggestion?

						- Bill