Naadloos van schets naar model

Data visualisation or datafication of a visual ?

Humans are visual creatures and graphs are beautiful to behold, like fireworks displays or flocks of birds. Yet, one often fails to see the forest for the trees. Human domain experts tend to be bewildered and daunted when tasked to read information from graph diagrams.

Instead, we are used to working with diagrams that schematically represent information. Domain experts, architects, machine builders, managers, organisation experts, etc. are trained to read schematic drawings that follow well defined conventions.

This begs the question whether it's possible to bridge the gap between schematic diagrams and linked data.

This story shows how Visio, which in fact is powerful 2D CAD software, can indeed produce Linked Data from railway signalling diagrams.

Signalling schemes are diagrams that are subject to strict design rules. The shapes in such drawings are templated and carry properties well understood by signalling experts. Typically, designers avail of stencils with master shapes, e.g. representing signals, tracks, railway crossings, train detection kit, etc. The designer drops these master shapes on the drawing sheet, close to a linear shape representing a track. The track is sectioned by shapes representing rail joints. Level crossings are drawn where the tracks are crossed. Text annotations reveal to the domain expert which interlocking equipment controls which trackside kit.

The attentive reader will note that shapes on technical drawings are representations of linked data - for expert eyes only. Master shapes in a stencil are types, shapes in the drawing are instances thereof. Crudely speaking, the position of a signal with respect to the track is a link between objects of type signal and track. This is the kind of information that is visible in drawings and that can be expressed as Linked Data.

An uplifting story

In the beginning there was a railway. It grew up to become a network.

Then we got the data. But the data sat on paper. Then in computer files. Lots of files that describe one and the same railway network.

Yet, the data in the files lived their separate lives, grew older and then grew apart.

This is the beginning of a beautiful story in which the data become as linked as the railway network that they describe.

Location or position ?

This is a very short graphic novel that shows how we think about positioning and locating networked equipment.

We consider a network to be a topology that consists of connected linear elements. Equipment has a function, a position on the surface of the earth and a location in the network. The latter is distinct from "classic" GIS thinking where objects have a position on the surface of the earth expressed in coordinates.

Short, position and location both inform whereabouts:

• position where on the surface of the earth
• location where in the network

Equipment that operates in a network topology, be it railways, telecoms or utilities, is attached to linear elements, e.g. pipes, wires, or rails. More often than not the map position is less relevant than the location in the network. This said, depending on their roles, people may be interested in either map position or network location. A network designer may only need to know about routes whereas a maintenance engineer needs to know how to drive to a piece of kit by car.

Spotemkim

Location with respect to a curvilinear feature such as a railway track often is more interesting than a position in euclidean space or on a map.

This diagram shows that location is given with respect to a "curvi-linear element of the network" which typically is the centre-line of a track (or pipe, road, cable, etc.). The spot is then said to be located at an "intrinsic coordinate" which is the relative distance along the feature in the range [0,1]. This allows ordering kit on the linear feature even if the absolute distance along the feature is unknown. Note that the linear element can have a length that is expressed in metres or whichever unit is appropriate. This supports the idea that lengths, positions and locations can result from observations of the real world. These observations can have meta-data informing quality, source, date of observation and much more. To the casual user, the meta-data may not be of interest. The façade design pattern is the elegant solution because it states that a thing has a spot location that hides complexity.

The position on a map, sheet of paper or the surface of the earth is optionally informed.

Against this backdrop, the railway model thinking can be stated like so

Function:

• A signal is a lamp on a stick
• A signal has a set of aspects
• A signal aspect sends instructions to the train

Topology for network location

• A signal is located next to a linear element

• A linear element is a piece of track that runs from point to point (or buffer stop)

• A linear element is an element of the topology

• A linear element links to another element (along which a train can travel)

• A relation is positioned at the end of two elements Linear elements are "glued" together, head-to-head, tail-to-tail or head-to-tail.

• A Signal is located at a given Spot

• A spot location can be attached to a linear element

• A spot location is located at a given distance-along a track

Topography for map position

• A spot can have geographic coordinates and/or
• A spot can have a position on a display or sheet of paper