It is common to encounter the error of supposing that we can simply begin with the raw bearing we have applied to the map, couple this with the stated scale of 6.4 miles to the inch and then just run the math. It is, however, not as simple as this.

The Phillips 66 Map used by The Zodiac is, as we shall find, actually not scaled at anything close to 6.4 miles to the inch, nor is the straight-up direction (Map North) properly aligned to True North. It is for these reasons that, through ignoring these combining deviations, we find ourselves regularly presented with proposed bomb locations that do not match the location that would have been obtained by simply following the instructions and laying off the coordinates on the map, as given.

The starting point, then, must be a detailed analysis aimed at identifying the precise degree of each of these primary deviations. And, to do that, we must start with the map itself.

Topographically we have the sloping hillside with adjacent road, surely frequented by school buses for both day outings as well as regular school runs, all combining to meet the logistics of the detailed mechanism we had drawn for us. Yet, more than just this, there is something oddly compelling about the existence of precisely these features at such an elevated and visible landmark—there for all the inhabitants of San Francisco to see and notice daily.

Were there to have actually been an explosion here, people would have been able to stand and point towards it from wherever they found themselves, with only the minimal of effort. With a telescope or an excellent pair of binoculars, it could even be scoped from the very peak of Mount Diablo itself. Even a search here for an ultimately absent device would have been enough to draw the daily attention of tens of thousands at a time.

The North-East face of the twin-peaked hill—precisely that face where the bomb would have to be placed to catch the early morning Summer sun—is just the very face which looms in the distance from many a vantage point along the majority of Market Street, heading west.

And maybe—just maybe—it is a schematic for these very Peaks that we have encoded in the otherwise mysterious symbol added to the Halloween card.

In any event, if there is any new potential solution I see offered to the Z32 cipher, my first thought is always to check whether it will reach from Mount Diablo along the "eighth radian" to somewhere around the middle of the San Francisco Peninsula, and I am all the less excited whenever it doesn't.

1. the first and the twenty-sixth letter must be the same;
2. as, also, the second and the final (thirty-second) letter;
3. and, finally, the same for the sixth and the fourteenth letter.

This is all that we have to work with.

Yet, it is perhaps worthwhile considering the question as to why it was these specific three constraints that the author chose to give to us at the outset, and not any of the alternatives that must have been equally possible.

Why, in particular, does he seem especially keen to direct us to the relationship between the very beginning and end of the cipher?

Here I'm going to be using only an old version of Microsoft Word and a copy of the Zodiac Phillips Map. The example I'm going to be working with is one chosen completely at random, but which nonetheless has a nice, natural 'read' to it.

We use, then:

AT FOUR AND A THIRD INCHES RADIAN EIGHT

from:

'RADIAN EIGHT', here, assumes the given division of the superimposed compass rose over Mount Diablo, which The Zodiac has suggested we divide into twelve equal sectors of 30° each. Radian eight, therefore, it that "radial line" extending from Mount Diablo at a bearing of (8 × 30°) + 17° = 257° from map north, the last term added to make the adjustment to Magnetic North, as indicated.

Let's get into it, then.

"If Baber’s interpretation is correct, the Zodiac killer’s taunting letter and Z32 cipher pointed to his first potential victim - Short, Margolis’s former lover who was murdered two decades before the serial killing spree in the San Francisco Bay Area began.

However, cryptography experts who previously verified Baber’s solution to Z13 - and confirmed to the Daily Mail that they believe those findings to be correct - have been unable to confirm his solution to Z32.

Ed Giorgio, the former Chief US Codemaker and Chief US Codebreaker at the National Security Agency (NSA), described the finding as ‘plausible’ and ‘consistent with other people's theories’ about the map and coordinates, but said it lacked cryptographic evidence.

‘There’re two parts to it: the projection of the coordinates and the map. So all that is very intriguing,’ Giorgio told the Daily Mail. ‘But the issue for us is how do you decrypt those 32 characters?’

He has no code, no whitepaper.  The way to decrypt those 32 characters is to test all plausible combinations of radians/inches, then rank the crytographic survivors by proximity to known Zodiac activity.  That will yield IN THREE AND THREE EIGHTHS RADIANS TEN as the top ranked candidate solution, due to it's proxiimity to the Blue Rock Springs attack site (1.15 miles).  When you check that location on the map you will see a big triangle landmark on the ground (see my profile pic).

Ultimately it seems like this solution is designed to create a connection between the Zodiac and Dahlia case.  There isn't much substance to it, although it is being reported globally.  The narrative does well to achieve clicks, in that regard it is a good attempt.

We don't have even this much currently for the Z32. That said, however, it is still possible to propose the 'form' of the original artifact in this case also. Essentially, but by way of example only, uncomplicated logic must have us believe it would have been nothing any-the-less concrete and tangible than this:

But here's the thing - and this is where most attempts go astray right from the start. We have to first ask where it is that we are searching for our bomb location using what we have been given, and the ubiquitous answer to this appears to be "on the ground." Of course this is, however, not the case at all.

Surely, were we to one day reach the actual solution, we will have a point identified on the solid earth. That said, however, our process of searching involves only our proposed cipher solution and the Phillips 66 map.

In and of itself, the map we are given is very poorly detailed and sloppily scaled. Despite the almost universal assumption that a translation of any proposed solution into lat/long or GPS or any other coordinates on the ground will match the equivalent point as marked on the map with the same bearing and distance from Mount Diablo, it won't!

Even a cursory attempt to create a semi-transparent overlay of the Phillips map on top of the equivalent terrain on, say, Google Maps, will show just how badly the attempt fails. And, in 1970, we didn't have Google Maps either. Basically, to be authentic in our search we just need to step back from the modern day, close down our Maps application, and stick with the Phillips map that we were actually given.

There are, essentially, two possibilities for what we are looking for. Firstly, it may be the case that all we have been given in the cipher is the relevant coordinate pair for the feature we are guided to identify or, secondly, we are directed to a much more general area and given an additional clue as to the specific ground feature sought.

In the first instance, the map must give us everything. The feature looked for must be actually present on the map. Of course, there are very few features actually shown - the bomb site must be somewhere along a pretty major highway or main road - and thus this realisation alone must cause us to give especial consideration to the second possibility. In the latter instance, then, we require something extra, and very likely the last characters of the cypher would give us this further clue.

For example, suppose the intended location was back at Blue Rock Springs. With very little precision - since the map doesn't permit precision of any kind - we might find that the decoded coordinates point from Mount Diablo over Benicia toward eastern Vallejo in general. That, in itself, wouldn't give us much at all. But suppose something like 'LAZULUM' were added at the end, then now - and only now - we can pin down something not immediately present on the map.

The point is, however we are to proceed, we simply have to get away from thinking about coordinates, bearings, distances on the ground. If that had been the intention, we would have either been given a better map or, more likely, none at all. Essentially, in this one instance, if not in others, Google is actually your Nemesis, not your friend.

I wanted to share some of my findings regarding my work on Z13.

To begin, I must address the unicity distance problem, and the fact that multiple possible solutions exist for this cipher.

To account for this, I decided that I needed a list of all adult male names that lived in Vallejo between the years 1968-1969.  The list of possible names must be constrained to people that we know lived in the relevant area, during the relevant time period.  

I was able to obtain a scanned copy of the Vallejo 1968-1969 city directory, and OCR'd all of the names, and addresses.

I then augmented this list with common nicknames.  For example a name list as a "Robert Smith" in the city directory may also lead to entries for Bob Smith, Rob Smith, ect.  

I then took this list of names, and created another list that formats them differently.  For example, a name such as Robert A. Smith, might be tested in the forms Robert A Smith, Rob A Smith, Rob Smith, R Smith, ect.  We are testing all plausible ways of writing the name.

With this final list, names that do not meet the symbol repetition patterns of the cipher are structurally rejected.  This greatly narrows the list.  Then come the positional constraints of the cipher text symbol.  These are incredibly restrictive, rejecting every name in the city directory.

And so, I tested all of the names that meet the symbol frequency constraints using as many basic transposition steps as possible.  Now I am starting to get very close matches on a small subset of names.

However, I did not find a perfect match.

There was a single name in the city directory, whose name was a nearly perfect match on Z13, using both his nickname, and his given name.  I did research this person, and they worked for the city of Vallejo, on public infrastructure.  They played JV football for Vallejo High School.  Their obituary did not mention family members, and they passed away in Vallejo a few years ago.

I will not name them (unethical), and I do not consider them to be the Zodiac.  Everyone is innocent until proven guilty, and his name was not a perfect match on Z13, just one character off after a basic transposition. 

The only thing that would prove a Z13 solution is a DNA and/or handwriting match that is verified by an unbiased, qualified expert.  They will also need to have lived in the area during the right time (bonus points if they left around 1974).  They would need to resemble the sketch, and be within the age range of witness descriptions.

I wanted to share my work on the cipher, even if I am not presenting a solution.  I do have a copy of the city directory for those that want to do their own research.  You can send me a private message if you would like it.