Objects Flashcards
What questions does object information answer?
Object information answers questions about PROPERTIES and RELATIONS of objects in the world, such as ‘where is this object?’ ‘How big is it?’
What are objects?
Objects are individual things located in space and existing in time. They can be uniquely identified and they can be described by their properties and relations.
What are some examples of objects?
Three geographic examples are:
- buildings: identifies through their addresses or insurance numbers, they have footprints construction dates, heights and owners and they are spatially related to other buildings as well as to the land parcel they occupy
- lakes: identified through their name, have sizes and depths, exist during a period of time, are spatially related to other water bodies (rivers flowing in and out of them)
- vehicles: identified by there VIN and license plate, they have an owner, a position and speed at any instant, and stand or move in public/ private traffic area at any time.
What is the duality between fields and objects?
Many phenomena can be considered either, for example, can identify soil type at any position on the ground using field view, or identify regions of a given solid type in an object view.
What is the single defining characteristic of an object?
All objects have an identity, which allows for tracking changes on their properties and relations.
All objects are bounded, but their boundaries and not always known or even knowable. The locations and boundaries of such objects are characterises by a central position (such as a mountain peak) and an approximate region ( a forest polygon). Can be delimited by a transition zone
What types of properties and relations are represented in objects?
All properties and relations of objects are either spatial, temporal or thematic.
Examples of a buildings properties are a buildings spatial footprint ( spatial) it’s year of construction (temporal) or its tax value (thematic)
Can spatial and thematic properties of objects depend on time?
Yes, the location of a moving objects the size of a greeting or shrinking object (think of glaciers) and many of the thematic properties and relations are a function of time. At any given time, these objects properties and the relations the object participates in are said to define TOGETHER the state of an object.
Can objects be disaggregated?
Parts of objects can be treated as objects, if they have their own identity. Conversely, complex objects can get aggregated from simpler ones as parts. Part-whole relations between objects and their parts are common, capturing spatial, temporal and thematic aspects.
E.g. a condominium is located in a building, which in turn is located in a building block.
How do we define object classes?
Object classes are defined in their shared properties and relations and form object class hierarchies
How are objects represented?
To deliver spatial information, object representations need to capture at least location and identity information. The smallest object models need to capture a position and record an identifier.
Is enriching object models always beneficial?
Not necessarily, object models that include additional spatial, temporal or thematic data can make them arbitrarily complex. E.g spatial enrichments of object relocations as positions, regions of shapes and sizes, is arbitrary.
However, enriching temporal and thematic data allows for representing objects with multiple complex histories and attributes e.g. turning a public into a pond in the wet season, freezing in the winter, and drying entirely at other times of the year
Why is modelling and tracking object identity theoretically and practically hard?
It is impossible to make sure a computer representation that has been assigned a unique and stable identifier remains pointed to the same thing in the world. E.g. natural objects are harder to separate from their environments and to uniquely identify over time.
What is the difference between object views of spatial information and object oriented representation?
Object view of spatial information of the world (in terms of identity and location) where’s the latter is a software design technique that groups data and the operations applicable to them.
Object-oriented techniques are ubiquitous and can be used to represent fields and other types of spatial information.
These two object notions share an emphasis on object identity properties and relations and their preservation in software models.
How are object data used?
Exploits location data to compute sizes, shapes, neighbourhoods and spatial relations for individual objects as well as distributions and intersections for multiples
How do we construct objects?
- from fields
- from locations and thematic attributes
