Talk:Record locator


 * this article contains simple inaccuracies. however, i was not able to edit it. a simple fact is a record locator can only contain 6 characters. Period. no 5-7 range. But, I am unable to edit the page******  —Preceding unsigned comment added by Envie (talk • contribs) 11:35, 11 February 2009 (UTC)

No, you are wrong. Locators as used by airlines were typically shorter than those used by CRS's, presumably because they only needed a smaller number of unique permutations than the CRS. Not sure if this is still the case.

As far as I know, most of the GDS systems use 6, but not sure if this is a mandatory practice now.

The whole article needs improving, especially as it does not make enough distinction between airline's internal systems, and the multi-access system that makes them accessible to the rest of the industry.--Dmol (talk) 11:46, 11 February 2009 (UTC)

seriously -- a record locator can only contain six characters. it is, in fact an annoying system limitation that means record locators have to get recycled every months while a ticket record remains active for 365 days, anywho, CRSs, GDSs, these are the same now. The article indeed needs much help. All the ones on these topics do. Envie (talk) 12:03, 11 February 2009 (UTC)

Number of locator permutations
I have a question: do the different systems take steps to avoid assignment of the same record locator for different PNRs between them? Is there any more information about how the record locators are managed to allow resuse at a later time? — Preceding unsigned comment added by 74.94.170.177 (talk) 02:38, 10 February 2012 (UTC)

I came here to answer the above question, but notice some grossly inaccurate info that has not been challenged. --Dmol (talk) 09:55, 12 February 2012 (UTC)
 * I use the Galileo system at work. Its locators are 6 place alpha-numerics.  That gives over 2 BILLION variations, so the claim about the numbers being recycled every few months is ludicrous.
 * Galileo is being replaced by TUD, a point-and-click version, which uses 6 place alpha-numeric. TUD also issues a Galileo locator for the booking, but can combine 2 or more Galileo bookings in 1 TUD locator.
 * The locator used by the airline, known as the Vendor Locator, might be smaller, and might be only letters, but that still gives an amazing number of possibilities. Even systems with five alpha-numerics gives 60,000,000 different permutations.  Five letters still gives 10,000,000, but I've not seen letters only for a while.  (These systems are not case-sensitive. They only show upper case.)
 * Many airlines, even large ones, use a GDS as their own system, so again we are back to the 2 billion possibilities.
 * Ticket records don't last in a GDS for 365 days, although potentially they could. The records are not normally accessible a few days after the last flight, and a month after that they must be specifically requested from the GDS company. They do remain on file with the companies, for what I presume is some years depending on the legal requirements for record keeping.
 * I would assume that there is some system to prevent two GDS's using the same block of numbers, but as we've seen there are billions, so it's not a real problem.
 * Just another thought, and I have no proof of this, perhaps the addition of the prefix 1A for Amadeus, or 1G for Galileo, further helps to differentiate the PNRs.


 * In case anyone comes here and sees the above claim that reuse is not a problem because 2 billion is such a large number, consider that there are tens of millions of flights every year, with dozens of passengers on each. If there were only 20 million flights in a year, 2 billion record locators gives just 100 locators per flight. If the major GDS systems issue duplicate locators, they might go a couple of years before running out, but it's certainly not "ludicrous" to suggest they might need to cycle them every few months due to additional limitations. - IMSoP (talk) 22:34, 13 November 2017 (UTC)