It's rather hard to believe also that the firms themselves don't have failsafes against huge mis-trades built into their software.
I guess it's conceivable that there might be instances when a firm might really want to dump 1000 contracts or 10000 or 100000 instantly, but it wouldn't take the world's greatest software genius to build such an option if really needed, while the main order entry system at least required a confirmation before executing orders in excess of some reasonable upper limit.
Maybe it's some macho thing - the traders get a thrill out of the idea that, at any moment, they could enter an order that would produce panic and financial carnage.
Or maybe it's just everyday stupidity.
As for the issue of exchange rules, I already stated earlier that I believe both the circumstances and the procedures for busting trades should be clarified explicitly. It also shouldn't take several hours to announce the decision. If a firm makes a mis-trade that it expects to get busted, it should be required to report it immediately, and the exchange should announce any decision to bust trades very soon thereafter. Facilitating this process would require that the conditions, rationales, and precise procedures for busting trades, and for assigning levels at which trades may be busted, would be spelled out clearly - something which would serve traders and exchanges alike.
The events on Friday leave the impression that first an "uh-oh" squad is assembled, that some time passes during which the cricumstances of the "event" are verified, and that some obscure process of negotiation occurs during which the potential impacts of various approaches are then compared - all before the nature of the event is revealed to the outside world. You have to wonder who if anyone outside the exchange is allowed input on the decision, what conflicts of interest arise, and whose a's are being c'd and in what order of priority.
Does anyone know exactly how and why the bust levels on YM were chosen this time around, for instance? If anyone can provide or direct me to a full explanation, I'd be very interested.
It seems to me that even the CME rules that PG linked don't go far enough in these regards.