Including time dimension

 

[1]Seems to be a nice demo of how the linked geo data has been used,
demonstrating the first 2 use cases of [2].

 

So continuing on Karl's thought on POI, if we consider time dimension, I
see that

 

A place is a location/anchor

A POI is any 'linked data'  which has place and time attached to it,
both of which could vary.

 

The variations for place are 'fixed' and 'varying'.

The variations for time are 

't1-indefinite'( things which are existing now and we don't know how
long they exist)  ,

't1-t2'( things of past)  both of which can be termed as 'fixed' and 

't1-t2'repetitive (things which become active periodically) which can be
viewed as 'varying'.

 

The various categories to which a POI can belong to are

 

A river/house/hotel/virtual tour  POIs  have 'fixed' place and 'fixed'
time(either 't1-indefinite' or 't1-t2').

A car/person POIs have 'varying' place and 'fixed' time(either
't1-indefinite' or 't1-t2').

A taxi POI  has varying place and  'varying' time (as the taxi driver
runs it every day from 8 am till 4 pm).

A movie/local ad/dinner at a hotel  POIs have fixed place and are
'periodic'(for similar reasons...).

 

 

Periodic time frame  can even have several t1-t2 pairs.

 

A POI may change from one of category to another based on requirements.
For example, 

A taxi can give service 24 hours and become ' varying' place and fixed'
time category POI. 

A hotel destroyed can become a 'fixed' place and 'fixed' time (with
transition from 't1-indefinite' to 't1-t2' )category POI. 

 

[1]
http://www.uni-koblenz-landau.de/koblenz/fb4/AGStaab/Research/systeme/cs
xPOI

[2] http://www.perey.com/ARStandards/Three_Use_Cases.pdf  

 

 

Just a line of thought on including the 'time' factor.  

 

---------

 

Vinod