Textusa – irrefutable proof

I enjoy Textusa’s blog most Fridays. There is usually a detailed trawl through the PJ files followed by some thought provoking revelations. Today’s “Irrefutable Proof” was excellent fare.

However, don’t read “Irrefutable Proof”. Textusa admits up front that it is a long post. And it happens to be mainly wrong, so do yourself a favour and read the short, accurate version here instead.

There are documents in the PJ files labelled as the Ocean Club check-in lists, and check-in lists used by the Tapas Restaurant. The latter are pages 615 to 638 of volume IIIa. http://www.mccannpjfiles.co.uk/PJ/TAPAS_BOOKING.htm gives you the lot, but you only need this post to understand what is going on.

Essentially, Textusa proposes that the Ocean Club records were extensively altered before the PJ got their hands on them, with the aim of hiding important names.

The issue starts with the PJ files. The link above describes the documents as check-in lists used by the Tapas restaurant. This is wrong. The prints are running tabs of what customers owe to the Tapas area, presumably so this can be paid at check-out.

Textusa points out that at least two days of records are missing, in a very odd order. Here I would agree, though I can think of one reason for that. If the ‘missing’ records contain information on T9 spending, they are relevant. If they don’t, they aren’t. I can’t tell from the records that are in the file whether that was significant or not.

Textusa thinks that as the information was produced at variable hours in the very early hours of the morning something strange is going on. I don’t. I think the Tapas workers keyed this information into the system up until when they clocked off around midnight, then the Ocean Club 24 hour reception pulled it from the system overnight, printed it and filed it.

Textusa thinks that the times recorded on the sheets, just one or two seconds per day, is much faster than any printer today, let alone in 2007, thus impossible. I think the time is the time taken for the database software to produce the print file, not the time to print it. One or two seconds is a reasonable production time. Plus the database software did not know how long the print file sat in the print queue, or how long it took to print it.

Textusa notes that various fields do not make sense. They change from a given value one day, to a second the next, to the first value on the third print. This flopping about, combined with values that make no sense, lead Textusa to conclude that the source of the sheets was not the database software, but something like Excel. Headings flop about, names flop about, locations flop about, column widths flop about. There are some really weird names on the lists, as in, they do not look like they are real.

From this, Textusa deduces that the lists have undergone major manipulation, the aim thereof to keep sensitive names away from the light of day. If that really was the aim, the oddities in the list draw the eye to the very distinct possibility of manipulation, so such an effort was futile.

Textusa started off by querying why such printed lists would exist in the first place, given that reception staff could look up the database at any time. That question basically answers itself – the lists were not for people who had access to the database system. They were for people who did not have access to the system. That would be people like the accountant (checking on the bookkeeper) and the owners (checking on everyone). The prints keep people honest. There’s no getting a bill paid in cash at the end and putting half of it through the till and keeping half for oneself.

As it happens, the daily reports show other important management data, notably occupancy rates.

By the way, check out those of the T9 who appear on this graphic and note something odd, but interesting. Most Mark Warner clients at the OC were booked in as adults only. Basically, OC staff didn’t get or didn’t record that children were in the party. That was because it was up to Mark Warner staff to provide cots for Mark Warner clients. The OC did it for other booking organisations, but not for Mark Warner clients.

So, what do you get when you look at the day I have selected?


Firstly, the records have punch holes. So the information was definitely printed to paper, then filed.

Second, look at the line of = signs near the top that goes bananas. Sometimes it has one _, sometimes it has two =, and sometimes it has 3. That is not Excel. That is the OCR software in the photocopier corpsing when the PJ took a copy. OCR mistakes explain all the flips and flops, and why there are names that any human being would think is not possible. And the variable column widths and the variable column headers.

What do I have now that I didn’t have before? Quite a lot, actually.

First, I have a map of who was located where in the OC. Which places were occupied, and how many places were vacant.

Second, I have a good idea of who arrived on the same flight as the McCanns. This was one of the earliest points at which people outside the T9 could be aware of Madeleine. It is hardly likely that someone flew out to Portugal with their family in order to snatch a child they were not aware of, but now I can identify these unlikely suspects.

Third, the OC records seldom recorded when children were present. The record for the McCanns shows a party of 5, but has them all as adults. So I know know that most of the OC staff were ignorant of the fact that 3 young children were going into 5A. This refines the possibility of which OC staff might be involved.

All in all, Textusa has come up with a complex, incorrect solution for a simple problem. However, as often happens, this has dredged up some good stuff, so I’m looking forward to next Friday.


2 thoughts on “Textusa – irrefutable proof

  1. Interesting read but you yourself are completely wrong about OCR. OCR software is used to convert printed characters into digital text. You simply do not use OCR when photocopying documents and the ones above are without doubt photocopied.

    • They are photocopied or scanned. Or quite possibly both.

      Photocopying makes none of the errors in the files, since it is a ‘dumb’ process. Scanning is essentially photocopying with ‘intelligence’ built in. My scanner puts documents through OCR as the default.

      I would say this information was first photocopied (giving a correct image), then the pages had notes added, notably the page number in the JP file. It could not possibly have had this page number at point of origin in the OC, so it was added later. Then the photocopied pages were filed in a manual file in the JP files. Then as the case was being shelved, the PJ produced a DVD by scanning all the relevant documents.

      There are scanning errors elsewhere in the files. The arrivals list happens to be particularly ‘buggy’ because software cannot do a grammar check on the content, as there is no grammar to check.

      Welcome aboard.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s