I’m doing some form of data visualization learning for 180 days straight because I need to #JFDI. See post explaining how and why I’m doing this.
Definitely got into a good groove in starting earlier today and managing my energy right. Getting a good pace in the first week of starting any new habit or process can be difficult, so I’m glad I’ve gotten there already.
Visualization Worked On or Created:
–USWNT Visualizations done yesterday. Added a bit more narrative and styling to this.
-I started exploring pieces of an old project at I did for a Techniques of Data Visualization course I took in grad school. This course was much less about code and more about critical visual theory, a healthy skepticism about data, and storytelling. I put together a project on looking at how US demographics have changed through time. I’m planning to beef this up into possible a portfolio type piece using visuals generated in Tableau and D3.js eventually.
Decomposition of a Visualization:
-Break – > code learning catch-up day
Udacity Problem Set One Finished – On section 2a Design Principles
- Visual Encodings + Data Types + Relationship = Chart Types!
eg. x coordinate and y coordinate with a shape of circle + two continuous variables + y value is dependent on x value = scatter plot
- Position along an aligned scale is most accurate for human visual perception while color hue and color saturation are least accurate for quantified human visual perception.
- Data visualization takes advantage of human perception because
- Humans can process visual information in parallel (all at once) as opposed to taking in information serially, like when reading text word by word.
- The human eye is excellent at identifying differences in placement and color.
Treehouse D3.js – Styling and Scaling with Data – Using Linear Scales
- In D3, scales are used to map a domain in your data to a visual range. It can be a pixel position, size, or color. Think of scales that have different methods as well as attributes.
- return parseInt(element.TMAX) // returns a string, using d3 will check the position they appear in the array arther than value, parseInt checks for maximum numeric value. Using a + sign will also coerce into a string, but careful doesn’t work with strings
- Scales can also take an inverted value takes in number in domain and spits back corresponding range value.Reading and Learning Data Visualization Theoretically/Critically:
- Thinking about how to make tables of relations, tuples, etc, dealing with functional dependencies, etc. and challenges how to encode and express those relationships in psuedo-wireframes.
- How to formalize conventions by taking view that graphical presentations are actually sentences of graphical languages that have precise syntactic and semantic definitions.
- Precise definitions of graphical conventions used to design and interpret information presentations make theorems possible for an engine.