Why do systems not talk to each other?

In the world of networks you can barely move without being subjected to a computer network in one form or another. They're everywhere and affect all parts of our lives, so why do we trust them implicitly when they can't even talk to each other?

Traveling back to the Isle of Man today for an Easter break, I came across this problem of systems not communicating with each other. Because two systems weren't talking to one another it created confusion amongst passengers. The passenger information display system at Luton airport was showing my flight in a slot ten minutes earlier than the scheduled one and with a different airline code. After checking the screen and not seeing my flight I instantly checked my ticket to see if I'd arrived on the right day.

After a little checking I noticed the flight and headed to the check-in desk which had the same, different, details. They gave me a boarding card with the original details and pointed me to the boarding gate which showed the wrong ones. The tannoy announcements came out with the original details.

Which am I meant to believe? What if there had been a major difference? Why was there a difference in the first place?

Admittedly this seems to have been a bit of a rant about Luton airport but how often do you come across systems which could communicate more effectively with each other than they currently do? How much could those improvements improve customer satisfaction? The simple answer is 'often', and 'very much' respectively.

Transport for London recently improved their experience for travellers in London by capping the amount one person spends daily on their public transport usage as long as they use their Oyster card. By collating all of the data collected across their vast network of entry gates and other card readers they work out how much each user has spent. This stage they have to do anyway to enable a centralised charging system, but they have gone a step further and introduced the cost-savings benefit to the system as well.

I suppose at any one time the ticket machine you are using may not know what has happened earlier in the day - especially in the case of busses, where there is no physical link to the rest of the network - but by the end of the day the central system will know what transactions have been made on each card and apply the benefits after that. As long as money isn't debited from the card instantly (or there is flexibility in the process) the user knows that they can travel as frequently as they like across London's transport network within one day whilst holding just enough credit for a day travelcard.

Mental note to self: post things earlier in the day so my brain isn't only half awake and not fully able to explain things!