Talk:John Herivel

}}}}

Untitled
Good job. Very clear and easy to follow, for a complicated subject. --Rbraunwa 15:15, 27 January 2007 (UTC)
 * Thanks for the comments and taking the time to review. &mdash; Matt Crypto 19:55, 27 January 2007 (UTC)

First time am posting here, mistakes if any are due to ignorance, please overlook.

It is likely that there are some errors in the page, as far as I know, or as far as I can make out. Details mentioned below.

(1)The above quoted text seems to be wrong. As per my understanding, the procedure was :

(1.A) The instructions of the day specified which rotors were to be used (ie which 3 out of 5 rotors in existence)and in what order they were to be fitted. This was the 'Walzenlage'.

(1.B) The instructions of the day then specified which letter was to be seen 'on top' for each of the three rotors (ie specified the 'window letters'). This was the 'Ringstellung' and was specified, and was not at the discretion of the operator. Assume that the 'Ringstellung' was ZUY.

(1.C) Having fitted the specified rotors in the specified order, and then having turned the rotors so that the window read ZUY, the operator would select any three alphabets say 'SCN'. He would type out SCN on the keyboard, which would be enciphered as (say) 'TGH'. In the earlier part of the war he would encipher the same letters SCN again, which would now be enciphered as (say) 'PKA'. The letters TGHPKA would be transmitted before the actual message. They were the enciphered version of the setting the operator chose ie SCN, repeated once, and the enciphering was as per the specified settings. They were not sent in clear. Later in the war, the repetition was stopped as it was a weakness, so only the letters TGH would be sent. However, the enciphering was again as per the settings specified in the instructions. The repetition lead to the Polish Cipher bureau breaking the ENIGMA. For proof, see the sections "Key Setting" and "Rejewski's characteristics method" in the page http://en.wikipedia.org/wiki/Cryptanalysis_of_the_Enigma

(1.D) Having done this, he would change the 'window letters' (ie letters 'on top') to SCN, and type out the message.

(1.E) The full transmission would consist of : Firstly, the setting chosen by him and enciphered as per the instructions ie the first part would be TGHPKA (later in the war, just TGH), followed by the actual message, which was enciphered as per the order of rotors specified in the instructions, with the initial setting of the rotors being SCN, which was chosen by the operator.

(2)There is one more objection to the text in its present form (ie as quoted above). The text says, in effect, that the setting would be passed in the clear. If that was so, then the only secrecy remaining would be (firstly) which all rotors were in use ie which three out of the five and (secondly) in what order.

(2.A) So, three out of five rotors makes for 5*4*3 = 60 possible combinations. The order of rotors would make for a further 3*2*1 = 6 possible combinations, the total combinations required for a full deciphering would be 6*60 = 360 combinations, which is trivial compared to the reality.

(3) The ring settings could only be done after fitting the rotors into the ENIGMA structure. The rotor combination changed from day to day. even when there were three rotors only, their order used to change, so they had to be removed, and refitted in the specified order. The ring settings could NOT be adjusted before fitting the rotors into the ENIGMA. This is unlike what is said in the quoted text above, ie in effect that the rotors stayed in place unchanged for many days.

(4) The correct setting of letters HAD to be done, the correct letters HAD to be 'on top' unlike what is implied by the text quoted above (ie unlike what is implied by "adjusting the rotors inside the machine would likely leave the correct ring setting at the top, or near the top, of the rotors").

(5) Herivel's contribution was, as far as I can make out, was the fact that an operator in a hurry would choose letters close to the day's 'Ringstellung' eg in the example quoted at 1(B) and (C) above, the operator would end up selecting (say) AUY or maybe ZUZ or ZVZ, which are comparatively close to ZUY.

(6) Refer the text

In reality, the daily instructions specified which rotors were to be used (post introduction of rotors No 4 and 5), and the order in which the rotors were to be fitted. The quoted text above implies that the rotors stayed in place unchanged, which is wrong.

Since I have not registered, I do not have a name/id on Wikipedia, so, I sign this as "Lurker". 19:08, 03 February 2014 (UTC)
 * There are deficiencies, but what you say is not quite right. Re-reading Herivel's book: shows that he section needs improving. The daily settings included selection of the rotors, which were changed each day without repetition of rotor and position - so there were only 32, not 60 options for the code breakers, a fact discovered by Nigel Foreman. The ring setting (ringstellung) was the setting for each rotor of the letter (or number) against which the notch for advancing the next wheel was to be positioned, not the letters that were to appear in the windows. Herivel's brilliant insight was to realize that lazy or stressed operators might not rotate the rotors more than one position forwards or back after closing the lid, leaving the letters on display as close to the ring settings, and then use that as the indicator for the first message of the day. I will change the text when time permits unless it has already been done. --TedColes (talk) 13:16, 4 February 2014 (UTC)

(1) Thanks for taking note. That was quick. Yes, I had made some mistakes, I realize that now.

(2) A sizeable portion of what I know about the Enigma is from Vectorsite (http://www.vectorsite.net/ttcode_08.html#m1 ), and from Wikipedia. The amendment you have made makes things much more clear. Many thanks.

(3) That said, there are a few things I want to point out. These are mentioned below.

(4) Refer "The daily settings included selection of the rotors, which were changed each day without repetition of rotor and position - so there were only 32, not 60 options for the code breakers" -

(5.A) This holds true only if the previous day's settings are known. Or the next day's settings, in case the message is an old one. If only one of the two is known it is 32, as stated, and if both (ie the previous and next day's settings) are known, the combinations to be tested would reduce further.

(5.B) However, if both (ie the previous and next day's settings) are not known, the probability remains 60/24/6 (when five/four/three rotors exist). This assumes that the settings were constant over a day.

(6) Regarding the rule " The daily settings included selection of the rotors, which were changed each day without repetition of rotor and position - so there were only 32, not 60 options ..."

(6.A) It appears that the rule did not apply across the board to all networks. See http://en.wikipedia.org/wiki/Cryptanalysis_of_the_Enigma, section "Operating shortcomings", third paragraph, second bullet (ie eighth bullet in the section). It says :-

(6.B) "For machines where there was a choice of more rotors than there were slots for them, a rule on some networks stipulated that no rotor should be in the same slot in the scrambler as it had been for the immediately preceding configuration. This reduced the number of wheel orders that had to be tried.[89]". [89] = Bauer, Friedrich Ludwig (2002), Decrypted secrets: methods and maxims of cryptology (3rd rev. and updated ed.), New York: Springer, ISBN 978-3-540-42674-5

(7) Once again, thanks.

Since I have not registered, I do not have a name/id on Wikipedia, so, I sign this as "Lurker". 18:55, 03 February 2014 (UTC) — Preceding unsigned comment added by 49.15.194.170 (talk)

GA sweep: article listed at Good article reassessment
This article has been reviewed as part of WikiProject Good articles/Project quality task force. Whilst it is fairly well-referenced and meets many of the Good article criteria, the focus and coverage of the article is more problematic. A very large portion of the text (in fact the majority) is not about the article subject himself, and were this to be removed there would be scant content remaining. For this reason I have listed the article at Good article reassessment, where its status will be debated by GA reviewers and other interested editors. Please feel free to contribute to this process by commenting in the appropriate section on the GAR page. Regards, EyeSerene TALK 13:21, 19 October 2007 (UTC)


 * As a result of a lengthy discussion at good article reassessment, it was decided by consensus to delist this article. It should be noted that several suggestions for meeting the good article criteria, including some as simple as renaming the article, were discussed and suggested.  For a full archive of the dicussion in question, see Good article reassessment/Archive 33. --Jayron32| talk | contribs  07:17, 19 November 2007 (UTC)