train scenarios

Consider a train that has various forms of location sensing within it. This will clearly make a 'space' of locations within the train, but this space itself both moves and bends within the stationary space of track, buildings etc. (fig. 1).

figure 1. train spaces

spaces:

Outside the train, the track etc. - a stationary space mapped to OS or perhaps GPS
The whole train as a largely linear space. Regarded as a straight line even though it clearly bends a bit. However, required granularity is low - "I want to get from seat 3F in carriage G to the buffet car"
Carriage space - coordinate space that does not bend!
Seat spaces requiring fine granularity. For example, to place 'reserved' tickets in seat back requires millimetre accuracy.

scenario - brief encounter

Alison is on the train and Brian is on the platform. As the train starts off Brian keeps pace and starts to say something to Alison through the window. Alison cannot hear through window and the noise of the engine, but can see his lips moving. He runs along beside the train until it gets too fast and he is left behind. It is one of those old fashioned trains with open steps and as the open doorway passes he swings himself up into the train.

points to note:

Alison's location is measured relative to the carriage and seat spaces and she is not 'moving' relative to them
Brian's location is measured with GPs and is moving
The boundary between them (train window) is permeable to vision, but not sound or movement.
This boundary is also moving relative to the stationary space, but still relative to the carriage space
The train doorway is also moving, bit is permeable to movement
Brian moves from being in a state of movement in the stationary space to being stationary in the moving train space.

scenario - oops

Brian is walking quickly down the train to find Alison. Someone's feet are sticking slightly into the gangway and he half trips over them.

"oops", he says and he turns
"sorry", says the person sitting at the seat

Then he recognises it is Clarise another work colleague.

points to note:

Brian is mentally moving primarily in the train and carriage space.
When he trips he shifts attention to the seat space.
This is an unplanned change - it is unlikely that we can have a simple way of telling when Brian's change of focus occurs
so what does this mean computationally?
one solution (for some applications/domains) is to use time of residence within a zone as a measure of focus

scenario - I've heard that one before

After a polite chat with Clarise, Brian continues down the train. As he passes the seat of Dave the dustman from Dunstable, he meets the ticket collector. Not only does Brian have no ticket, but he left his wallet in the office. For some time they stand by Dave's seat discussing it.

Happily Brian manages to find some cash in his back trouser pocket, but while the ticket collector is issuing the ticket, the train stops at a station and then starts again. As they finish, Brian looks out through the train only to see Alison slowly moving past the window. She is standing on the platform looking at the information board while the train moves off to the next station.

Brian's words are unrecorded.

Points to note:

Brian spent some time next to Dave's seat. If time in proximity were used as a heuristic for entering the seat space, Brain would have been regarded computationally as in Dave's seat space.
At the end Alison was stationary relative to the station space, but moving relative to the train space.
Brian was looking at Alison, but she wasn't looking at him, although the boundary is permeable to vision you have to be looking through it. (focus)