Tracy Pierce writes:

>Ah, the BREAK key... Just how does (or did) BREAK work?  This is left over
>from Teletype, right?  Anyway I've never been able to associate a character
>with it.

BREAK isn't a character, but a line state. A serial link has two states,
called "mark" and "space". The line idles at "mark" (1), and a zero bit
is defined as a certain period of "space". The period the inverse of the
baud rate (which, on a serial line, is the same as the bit rate). So a
0-bit at 300 baud is defined as 3.33 ms of "space". A full character,
including one start and one stop bit, is ten times that, or 33.3 ms. (The
start bit is a 0, which is how the receiver gets in sync with the
transmitter. The stop bit is a 1, which guarantees that the start bit
will always be a 1-to-0 transition.)

If the line stays in the "space" condition for more than one character
time, that's a framing error, and if it's long enough -- the usual
standard is 1/2 second -- that's considered a BREAK.

So a BREAK isn't a character at all, which is why you can't associate a
character with it. On a network, the BREAK has to be signaled by the
protocol. Telnet defines control codes -- pairs of characters that begin
with a 255. The sequence 255 243 is the Telnet BREAK implementation, so
when the host sees this byte sequence, it sends a BREAK signal to
whatever program is using the telnet connection.

NS/VT is transaction-oriented, so when you press the BREAK key on an
NS/VT terminal, the terminal sends a BREAK transaction to the host, just
as it sends CHARACTERS TYPED transactions and so on.

-- Bruce


--------------------------------------------------------------------------
Bruce Toback    Tel: (602) 996-8601| My candle burns at both ends;
OPT, Inc.            (800) 858-4507| It will not last the night;
11801 N. Tatum Blvd. Ste. 142      | But ah, my foes, and oh, my friends -
Phoenix AZ 85028                   | It gives a lovely light.
[log in to unmask]                   |     -- Edna St. Vincent Millay