Page 1 of 1

Thoughts on the Zodiac 340 in light of typography

PostPosted: Tue Dec 06, 2016 6:22 pm
by Soze
Recently I have been reading up on the subject of typography and, typesetting, as it relates to typography. Through the course of this reading I came across a mention of "minding your P's and Q's".

In the early days of composition for printing, individual letters or symbols, called sorts, were cast from lead and kept in a box according to their font style, size, weight and whether they were upper or lower case.

It is particularly important that, when finished with composition, each sort is returned to the appropriate slot for storing. If they are not stored in the appropriate location, when starting a new composition, you may find that you have pulled a "b" when you needed a "p" or a "q" when you needed a "d"; they all have the same shape and only take form given its position. Letters that carry the same shape are: b/d/p/q, n/u and m/w.

When I think of the above I am reminded of two things:

1. 11/9/1969 letter, page 3, regarding the Zodiac's confusion over the words "barking" and "parking".
2. The backward capital letters of the Zodiac 340 cipher.

Point 1 above, I feel, needs no further explanation as it's point is raised in the paragraph above it. With regards to, point 2, I offer the following:

The alphabetic symbols of the 340 are in capital letters. If you reduce the symbols to the lowercase form, the vast majority of the symbols remain unchanged except for, the backwards letters. When changed to lowercase form, a backwards "B" becomes a lowercase "d", a backwards "Q" becomes a lowercase "p" and a backwards "D" becomes a lowercase "b". The importance of minding your P's and Q's.

This changing of letter form from one to another affects the frequency count for each symbol. A "B" symbol for example, which has a count of 12, would now have a count of 17 by adding in the backwards D (b).

Prior to this writing I had merged symbols according to what I have been describing above and reduced them to two columns of 24 symbols; letters and symbols. My initial thought at the time was that maybe this cipher was a bilateral Baconian cipher given the two columns, the count of 24 and it's use within steganography. As I read more into typography I moved away from this thought.

Before I go on I would like to add that there are two types of frequency tables; one that we read about all the time and another dealing with the printing industry. I have listed both below with differences between the two marked:

Letter frequency table:

ETAOINS(RH)DLUCMF(YWGPBVKXQJZ)

Printing industry frequency:

ETAOINS(HR)DLUCMF(WYPVBGKQJXZ)

The list above will fall into play later.

Typography began as a manual process for printing the written word in newspapers but also for books. Casts were made of individual letters in lead, placed in molds to form words and then, sent off to be printed. How these casts were placed in the mold depended upon which field you were working in; newspapers or books. Newspapers I believe played their type from right to left and top to bottom. Books were upside down, left to right and bottom to top.

From there things went to the mechanical: Linotype and Monotype machines. Linotype, which was the least expensive of the two, was predominantly used by newspapers and Monotype was predominantly used by book makers.

Both the Linotype and Monotype used a keyboard for typing text and both made casts of the letters typed for printing. Where they differed was in the number of alphabets used and the casting process.

The Linotype had 1 keyboard with 3 columns: lowercase, symbols, uppercase; 2 full alphabets. The Monotype had 2 keyboards and began with 5 alphabets but ended with 7. Jyst as the Linotype, the Monotype, had lowercase, symbols and uppercase. One of the alphabets, in both cases of the monotype alphabet, were for justification of text.

The Linotype had 91 columns that held both alphabets and symbols. Two of the 91 were for the letter E given its high frequency. As the operator types, each letter falls from its slot to a carriage. When all letters of a line are typed the whole line is positioned in a chamber that fills with lead to produce a cast of the whole line typed, hence, Linotype.

The Monotype is a bit different in the casting process. I can't fully describe it as I'm still reading on the subject but can give you the gist. Instead of having slots for the alphabet, the monotype, has a casting plate. This casting plate, empty, looks and functions, similar to a chess board. You have all these connected sections that run off of columns and rows. Monotypes began with a 15 x 15 matrix that would hold 225 characters, moved later to a 16 x 17 that would hold 272 characters and then later (1967), a 17 x 20 matrix that would hold 340 characters.

These casting plates had both alphabets, symbols for printed text and symbols for justification (which was only read by the machine). There were standard plates but an operator could make their own casts or order them through the company. When filled, caster plates look like what the 340 looks like. All these symbols will fill, in various order, every block within those columns and rows.

When the operator types the text to be printed, instead of casts of letters falling from a slot like the Linotype, a roll of paper is punched. This punched paper looks like the roll of paper in pianos that play on their own. You would think that the punches in the paper indicate characters but they don't. Like a chessboard they indicate positions, column to row, on the casting plate. Unless I'm confusing one cipher for another, this reading of columns and rows, is like a playfair or a vigenere.

After the operator has typed the desired text, the paper with punches is removed and, fed to the casting machine via the last item punched. The reason for this is because, at the end of every line typed, there is justification code that tells the caster how to space each word and how to end a sentence. Putting the punches in backwards basically sets the machine for casting.

I'm still digging into the Monotype so I can't give a full list of the codes used and I'm not entirely comfortable in some of my future mentions of code. However I do hope that, even if I were slightly wrong in code description, you will still see the point I will try to make with regards to how I'm thinking the Zodiac 340 cipher was set up.

I'm not really sure if the keyboard operator types this or if the casting operator does this just prior to inserting the punched paper but an H is typed. This H tests the machine for type. Then the paper is run beginning with the last item punched.

So in looking at the 340, as is, the H in the first column from top is that testing of the machine for type. The reading of the cipher is then read from the bottom up and read backwards. I say this because when the punched paper is inserted it is inserted by the last item typed first.

There are codes that the machine reads that pertain to justification. A backwards k is used at an end of a line. From what I'm understanding it could mean an end of a sentence or sending a signal to the machine that there is still unused space on this line; make the justification. An empty square is in similar fashion. There is a code for a vertical bar similar to what looks like a lowercase L in the cipher. Double letters or symbols act in a fashion similar to an em.

I'm stopping here because I don't have a full list of the codes and how they operate. I suppose that the point I'm trying to make here is that the justification codes in monotype (like what I'm describing above), should be looked upon in the 340 as nulls and that the information between them, should be read from the bottom up after frequency and substitution.

Soze

Re: Thoughts on the Zodiac 340 in light of typography

PostPosted: Tue Dec 06, 2016 6:44 pm
by Soze
To add....

The 17 x 20 matrix created in 1967 containing 340 characters was monophoto filmsetter mark 4. This was the start of digital printing and a goodbye to the mechanical for the most part. The New York times held out until 1978.

Soze

Re: Thoughts on the Zodiac 340 in light of typography

PostPosted: Tue Dec 06, 2016 8:05 pm
by traveller1st
Interesting indeed. Good stuff Soze.

Re: Thoughts on the Zodiac 340 in light of typography

PostPosted: Wed Dec 07, 2016 4:58 am
by Jarlve
Hey Soze,

I'm not sure if I understand correctly and I have a bunch of question.

Do you think it's possible that this 17 by 20 matrix was used in creation of the 340? This 17 by 20 matrix has a fixed size? Can we correlate its size to the 340?

Do you think there could be a link between Zodiac and typography, it seems very plausible. As traveller1st pointed out somewhere else, Zodiac used different "fonts" throughout his letters and they display almost uncanny spacing habits. See: viewtopic.php?f=48&t=2586 I think allot of people have pondered Zodiac having an artistic or graphic background of some kind. In the HQ 340 it can be observed that symbols were drawn in the same way every time, a display of typographical consistency.

If you are suggesting merging of symbols, could you summarize a list or is it only the "B" symbol as you suggested?

You are suggesting a different direction, do you mean from right-to-left, bottom-to-top? You also mention column.

Soze wrote:When the operator types the text to be printed, instead of casts of letters falling from a slot like the Linotype, a roll of paper is punched. This punched paper looks like the roll of paper in pianos that play on their own. You would think that the punches in the paper indicate characters but they don't. Like a chessboard they indicate positions, column to row, on the casting plate. Unless I'm confusing one cipher for another, this reading of columns and rows, is like a playfair or a vigenere.

This sounds interesting but I don't understand what you mean, could you give a small practical example please of how this could encode a plaintext? I think you may be referring to playfair.

Re: Thoughts on the Zodiac 340 in light of typography

PostPosted: Wed Dec 07, 2016 3:19 pm
by Soze
Letters merged:

1. Backwards L, K, F, Y, C and J merge to their respective capital letter symbols as, even in lower case, they are the same.
2. Backwards B (d) merges with capital symbol D.
3. Backwards Q (p) merges with capital symbol P.
4. Backwards D (b) merges with capital symbol B.
5. Filled P (p) merges with capital symbol P. *An error in casting is when the bubbles of letters are filled with lead. Unfilled, this letter, would still be a P.
6. Backwards P (q) is made to a q. There is no lowercase Q so this only changes the appearance of the symbol.

Note: I don't include the upside down T or V. Most of the mirror symbols flip on the horizontal and these flip on the vertical. Because of this I chose to continue with the horizontal and treat the upside down T and V as unique symbols. You may think differently and may want to merge them with their respective upright uppercase letters.

Symbols merged:

1. > and <
2. Left empty and right empty square
3. Left empty and right empty circle.
4. Left and right I dot.

Note: I treated the top empty and bottom empty circle in the same fashion as the upside down T and V.

My mentioning of the merged symbols is because, if the Zodiac was someone very familiar with typography (which i think he was), then this would be just the sort of thing he would do to mess up the process of cracking the 340. My thoughts are, when you get to the point of substitution, you make the corrections. Without knowing anything about cryptography you may wish to do it before.

My firm position with regards to reading the cipher is from right to left and bottom to top. Everything in monotype casting is in reverse. So the start of his actual message will begin bottom right.

In the monotype system there is a chart for manually reading a punched hole. This chart is 6 or 7 inches in length and approximately 1 inch high. I've only seen this in video and nowhere else. I don't know what it's called and it's position in the video didn't give clues to how it looked. All I know is the short and skinny on how it was used.

Placed against the punched paper, it tells via column and row, how to match that hole to a symbol or letter on the casting plate. Originally my thought was that this chart would correlate to the 340 and tell us what that symbol meant. Originally I thought that it would essentially work like a playfair or vigenere. That was what I gathered based on how everything looked and what little I knew of ciphers. However, none of that made sense as, all we have, is what "looks like" a casting plate, I.e. the 340. So the chart is useless and so was my original thought.

The Zodiac, he knows everything there is to know about typography from the casting of individual sorts to the mechanical and then ultimately the digital by way of his 17 x 20 matrix of 340 characters. He knows all about justification and page layout. I can prove this later.

In thinking about the above and, trying to figure out how he would encode a message, I came up with the idea of codes mentioned in the latter part of first post. I apparently felt that I didn't do a good job of explaining it as I dreamt about it all night. The practical example I came up with in dream: HTML.

In HTML coding you have code, then you have text, then more code and so on until the end. Some codes pertain to the overall structure of the page and some code pertain to a particular sentence or bulleted text. Occasionally you will find some code in the middle of sentence, at the start and end of words, indicating bold or underlined words. The end result is a completely formatted page to ones taste but without the code. The code in HTML are nulls and never to be seen unless you make a mistake in the coding process. HTML code is a technical writing process only.

My thoughts regarding the 340 is that the 17 x 20 340 matrix should be left as is. No swapping columns. No putting rows on top of another. Do nothing. The 17 x 20 matrix is just a page layout from the era of digital. Its what he knows. It's what he has on hand. A mold so to speak inside of which he wrote his message. His message, I think he did just like, HTML. He wrote it in technical form complete with code but in justification code. You will have code at the start of sentences and the end of sentences. You will have code in the middle of sentences because, if typed on a Monotype system, there are only going to be a certain number of words you can fit per line. This code in the middle will be justification to the machine to make this line square. So there are going to be codes throughout and perhaps a great number of them. I wouldn't be surprised if his message was no longer than 2 or 3 sentences.

While I don't have a list of all possible codes used as of yet, and I am hoping Trav can help, the backwards k is a very good start. This symbol ends a line which means it ends a sentence or breaks one up for justification.

Anyway I have done the best I could to answer all of your questions except one. If you have additional please ask or give input as you see fit.

With regards to the unanswered question of " Do I think there could be a link between the Zodiac and typography" my answer to you is a flat, hell yeah, yes but that I need to write in another post to explain and will work on that next.

Soze

Re: Thoughts on the Zodiac 340 in light of typography

PostPosted: Wed Dec 07, 2016 4:14 pm
by Jarlve
Thanks for all the information Soze.

The link between Zodiac and typography is very interesting.

I don't fully understand your symbol merging instructions, do you have a version of the 340 were you did this? What I could do for you is check if some properties of the cipher improve and run it through my solver. The thing is that merging of symbols would likely not do much good towards a solve since in the best case that essentialy only reduces the number of symbols the solver need to process.

Code: Select all
HER>pl^VPk|1LTG2d
Np+B(#O%DWY.<*Kf)
By:cM+UZGW()L#zHJ
Spp7^l8*V3pO++RK2
_9M+ztjd|5FP+&4k/
p8R^FlO-*dCkF>2D(
#5+Kq%;2UcXGV.zL|
(G2Jfj#O+_NYz+@L9
d<M+b+ZR2FBcyA64K
-zlUV+^J+Op7<FBy-
U+R/5tE|DYBpbTMKO
2<clRJ|*5T4M.+&BF
z69Sy#+N|5FBc(;8R
lGFN^f524b.cV4t++
yBX1*:49CE>VUZ5-+
|c.3zBK(Op^.fMqG2
RcT+L16C<+FlWB|)L
++)WCzWcPOSHT/()p
|FkdW<7tB_YOB*-Cc
>MDHNpkSzZO8A|K;+

Re: Thoughts on the Zodiac 340 in light of typography

PostPosted: Wed Dec 07, 2016 5:40 pm
by Soze
No. Don't worry about it. My interest, despite its location, was not about solving the cipher. I could care less about that as, I don't think the message, will reveal anything useful in terms of catching him. In my opinion the 340 was a diversion.

My post was nothing more than trying to add the 340 to what I already know of the Zodiac, basically, here is some things I know, here is some things I have thought about in relation to what I know, could this be plausible in terms of the 340. If it worked out to be useful (which it doesn't sound like it is or will be) great but, if not, just the same for me. Outside of that I have no interest. I thiink you will see what I mean when I go to talk about the Zodiac again in relation to typography.

I do thank you for your offer though.

Soze