R.E's and Routines.
10. The theory of ordinary cribbing is very simple - it is the practice that is difficult - and there is not a great deal to add to the brief remarks in Chap. I Paras. 38 and 39.
11. Routines or "straight cribs" present the cribster with four problems (1) discovering the crib initially i.e. realizing that there is a regular message being sent in one or several standard forms (2) identification - i.e. the problem of picking out this message from a day's encyphered traffic (3) choice of form - the decision which form has most probably been used on the day in question (4) assessment of the final result - how likely is the crib to be right? There may be a choice of several different cribs and one of these has to be given preference on the machines; and if this fails on the best W.O's (Chap. VI, Para 23) should a second crib be run on the same W.O's or should the first crib be run on some of the less good W.O's?
12. (1) The discovery of cribs depends on a seasoned cribster looking through the whole of the traffic over a series of days. Cribs occasionally turn up which do not occur daily but at wider intervals - these will almost: certainly be missed by anyone except a good and experienced cribster; only he will see that the message is of a routine type and will search through other traffic trying to find another appearance of it. Inexperienced staff can keep track of cribs once they have been found but they cannot find them. (2) Identification depends on examination of external characteristics, such as time of origin, length, call sign and frequency. The chief difficulties arise when there is no obviously right candidate but several possible - one with better time, another with better length and so on. Here (as in all cribbing problems) experience is all important in developing the capacity for sound judgment. (3) Choice of form is decided by the non crashing test, which will eliminate a number of possibilities, and by examination of forms previously used; on the whole it is not difficult to see which is the best form in any particular case. (4) The assessment of the
-113-
Back to Cryptographic History of Work on the German Naval Enigma contents