> BTW the CNCP course was where I first learned the difference between BAUD > and bit rate. I guage my correspondents and listeners by their interest in this arcane corner of computer jargon: if they pay attention and understand the difference between baud and bits per second, they're technical; if they nod their heads ``un-huh'' a few times and don't get it, they're users. I explain the difference another way: remember boy-scout (or armed-forces?) semaphore signaling with flags? One flag posture per second is one baud (and a baud is a rate, like the knot: never ``knots per hour''), but the communication is clearly occuring at more than one bit per second, right? > Each technician had to be able to read baudot code right off of the paper > tape, and that was accomplished by about a week of code practice in class - > The instructor would key baudot messages audibly ala ham radio and each > student would hardcopy. This is the time of year I most miss paper tape. My father used to bring home tail ends of rolls from the phone company in Minneapolis, and we would weave and fold these many-colored remnants into Christmas tree decorations. > BTW the RS232 BREAK feature was a result of how the electromechanical units > used to work. When a space was detected (start bit) the serial to parallel > motor would advance whether the code was correct or not. Hence, if the > current loop was down it would be interpreted as a repeated sequence of 00H > (NULLs). The print carriage would go up and down but no striker would advance. > To this day most USARTS support BREAK generation. Much of our communications terminology hearkens back to telegraphy. A single wire and earth ground return, with a key in, say, New York, and a sounder in, say, Chicago, could carry news and stock prices one way: ``simplex''. Overnight, received messages could be recorded by an automatically advancing paper strip and a pen attached to the solenoid, hence: ``marks'' and ``spaces''. What should the idle state of this line be? Well, you probably want to know if Indians (or whoever) ever pulled down a telegraph pole, even if you weren't receiving a message at the time; besides, ``low-power sleep mode'' hadn't been invented yet, so the idle line state was ``mark'' and absence of current meant something was wrong. How about if Chicagoans think they have something to tell New Yorkers sometimes? Well, you could string a second wire, with key in Chicago and sounder in New York: this would be ``duplex'' connection. But you could also share the single wire and take turns in each direction, with a sounder in series with the key at each end, and a shorting switch across each key for receiving: this is ``half duplex'', so now the two-wire version needs a bigger name: ``full duplex''. So now suppose it's the morning of some particular day in 1871; both offices have their keys shorted to receive, so current is flowing and both sounders or recorders are marking. New York has a 4-hour news feed, so he opens his switch and starts keying. Both sounders are in series, so they both click away, and the transmitter has some confidence that he's keying into a good circuit. Now in Chicago, someone rushes up to the newsroom with some wild story about a cow and a fire and whatnot, and he doesn't want to wait for the 4-hour news feed to finish first. He can interrupt New York by opening his key switch, so that both sounders go silent. New York knows that this is either circuit trouble or else an important interruption, so he closes his switch for a little while to see if Chicago wants to key something. Chicago interrupted New York with a long ``space'', which we know as ``break''. > Anyway, I hope this is not too far off topic fot the list (I guess I'll find > out!). No, you just gave me an excuse to spew. > Regards, Dana Frank Raymond > dfr@icom.ca Likewise, Peter F. Klammer, Racom Systems Inc. PKlammer@ACM.Org 6080 Greenwood Plaza Boulevard (303)773-7411 Englewood, CO 80111 FAX:(303)771-4708