Introducing Visual Deconstructions
Taking visualisations apart to understand how they were made
Have you ever looked at a map (or any data visualisation for that matter) and thought, I wonder how that was made? If so, then a new concept that the GeoDataViz team at Ordnance Survey are calling Visual Deconstructions, could help.
What is a Visual Deconstruction?
A visual deconstruction is a concept that allows them to record the styling rules for a given data visualisation. It is made up of a title, a description, a url where relevant, keyword tags, an image, plus the draw order and styling information for each layer of data from which it is compiled.
It is a form of documentation that allows you to quickly reference and recreate styling rules, as well as being able to share it clearly with others. It is also a great way to learn how something is made and therefore is a useful tool for someone designing their own visualisation.
For a better idea, here is a minified version of what a Visual Deconstruction looks like:
Useful documentation
Our GeoDataViz team have been using Visual Deconstructions for a while now and find them really useful for documenting work in a visual manner. They allow for quick future reference and help ensure the consistent application of styles that can be shared with other teams too.
The title and description give the all important context and purpose. When trying to understand how a map or chart was made, it’s also important to know why, and who for (the audience). The list of layers then breaks the visualisation down to reveal each slice of data that makes it up – in the order in which they’re drawn. This mimics the layers in a GIS or graphic design software.
The styling rules for each layer of data are then laid out clearly. Data types include Points, Lines, Polygons, Text and Image (often used as a baselayer). Here is an example for each of the data types:
Points
Lines
This example shows how we handle stacked or composite styles
Polygons
Text
Image
You may notice that we don’t record the sizes of features such as line widths or font sizes. This is because they vary based on the software units, dimensions of the visualisation and/or map scales etc. The image itself should help you gauge the relative sizes of features if you are copying the styles.
Now that we have explained what a Visual Deconstruction is, you can see a couple of our recent examples here and here.
In the future we will be hosting Visual Deconstructions on our website and hopefully they will prove to be useful resources. Once we have enough they will also serve as a gallery of our work.
We would love your feedback so please get in touch if you’re interested in this concept and let us know if you are already doing something similar.