AF447

[stickyb]

ACARS

I have gone back again to look at the ACARS messages. I posted in #843 on 9th June that at least some of the messages had been truncated, but not many people seem to have picked up on that fact.

Now, looking at the format of the printout, what else can be discerned?

1. the listing seems to be a subset of messages extracted from the total acars info available for AF. The selection criteria appear to be a) by reg no. FGZCP and b) by time - 120509 thru 010609

So someone has run a programme that does "show me a list of all acars messages received from FGZCP between 150509 and 010609

There could well be other filtering applied, as I am led to understand that other things are reported on acars on the normal takeoff/landing event.

2. The list appears to be displayed in earliest flight first order, and latest message first within flight. This seems slightly confusing, as you have to then read down the printout from page 1 for a chronological sequence of flights, buit then having found the flight you are interested in go to the bottom of the list for that flight and read backwards to get a chronological sequence of messages.

3. The timestamp displayed as part of the messages is not the timestamp used for sorting the messages into the displayed order, as some messages "appear" to be out of sequence. Maybe there is a separate undisplayed timestamp such as arrival time, and maybe the messages are not even transmitted in chronological sequence - some messages could be deemed more important than others and therefore jump the transmission queue.

4. The appears to be a "trailer" message at the end of a flight. This is the message with just / from what i assume is the previous flight, along with the PFR message from 2013. Maybe someone can decode that?

Therefore my assumption is that the first 27 pages contain messages from previous flights of that airframe. What puzzles me is that the list seems to go on for 256 pages, implying that there are many more messages following on from the ones we have seen. It is certianly an odd coincidence that the list would end exactly on a page boundary. I would also add that 256 is one of those "magic" numbers in computers, and could just indicate that a lazy programmer couldn't get the code right to count the number of pages that were going to be printed.

All this simply leads me to caution on attributing too much authenticity to these messages