Studio Session: Planning your story aim, data, and text components
Having critiqued a webpage last week, and having generated an interactive info vis for the web in the tutorial sessions, we have a better idea of the main components that will make up our story for Assessments 2 and 3. Let's outline these for a ready reference:
- A title
- Sections of the story: sub-headings
- Text for the story
- A main interactive info vis
- Static info vis to highight main points for the sections
For this week, let's start to develop ideas for items 1, 2, and 3 on that list:
- Before we do anything else, here is some food for thought: the process of design is often laid out in a linear manner. For example, here I am asking you to focus on the title, section structure and text of your story. However, there needs to be, inside your head, a parallel thought structure evolving, with tight links forming between your story idea, your info vis, and the data that is going to drive the story and the info vis. Consider these points:
- Breadth versus depth, complexity versus simplicity, spread versus precision: For example, while each of us really does want world peace, it is no good having a story idea that says "The aim of my story is to present data that convinces everyone in the world to become peaceful". While a laudable aim, it is at once too broad, too general, and too spread out - and when it is all those things, it may fail to have an impact. On the other hand, you may land a fantastic dataset on all the wars the world has seen, and on the resulting social, economic, or cultural impacts. So, you may then consider a story titled "The price of war: is it too high?" (or "The aim of my story is to document the social and economic impacts of war in world history") which conveys more or less the same emotion as before, but is much more tied into to your available data. So, there is always a trade-off between the factors listed above.
- Never lie with data: Two, we should never massage our data, our text, or any info vis we create, to claim something which is not explicitly present in the data. For an extreme example, we cannot have arguments of this form: "This data set shows me that current science cannot explain how our brains think. Therefore, the only other possibility is that an extra-terrestial green coloured, frothy, higher level intelligence must have planted intelligence in us." No, really, the data only shows you the current limitations of our knowledge: we don't know "to date". We may at a future date. Period. This type of reasoning may seem far fetched and crazy, and it may make some of you laugh, but believe it or not, while researching for writing this tutorial I found that this is exactly the kind of argument that well-endowed, extremely rich but crazy institutions like the Discovery Institute use: they "appear" to use scientific information, but in essence they are using information to intentionally deceive. The Info Vis book, The Truthful Art has a whole section discussing this type of deception with data, including the above example, and many other interesting examples from climate change and other world-relevant issues on the truthful presentation of data.
- Evidence, not belief: Three, if you start with a belief, and then go out looking for data to support it, you will always find some data to support your beliefs. As the old saying goes, "if all I had was a hammer, then the world seems full of nails". Or, if all I had is a square frame, the whole world seems full of squares, because that is all I have to look through. If you search on the Internet, an astonishingly high number of what passes for infographics and info vis is of this category. And yet, in essence, this is scientific data and info vis crime. Instead, the scientific and honest way would be the following: You state your belief as a hypothesis. Then, to the best of your ability, you try and find data to support it, as well as to negate it. That is, as Richard Feynman once said, you have the integrity to try and bend over backwards to find proof that negates your own hypothesis. And then, you put together ALL of the evidence to honestly bring out the "truth", and acknowledging that there is always going to be some data that you have not exhaustively captured, which may change your findings.
Ok, we veered into philosophical territory there - but its important to have the ethical and philosophical bases of our work all sorted out, because it deeply affects what we produce. Now we will follow some simple design and ideation steps, modified slightly for producing the stucture of a "story":
Brainstorming Aims (20 minutes): First, hoping that you would have already spent some time thinking about data sets and stories, write down about 5 or 10 one-sentence descriptions of the following form:
The aim of my story is ...
At this stage, its ok to write multiple such aims. That is, even with a single data set, you could develop several aims. Or you could have aims associated with different data sets. Don't worry too much about finalising something - just treat this as a brain storming session, and write down your aims (you might want to write down the main data set(s) along with each aim). The only restriction is that it should be one-sentence aims. These one sentence aims will help you think about titles.
Brainstorming Titles (15 minutes): Second, associated with each aim, think up some catchy titles.
Develop User Personas/Target Audience (30 minutes): Third, write down user persona features for your target audience. Its ok to have a niche story, for a particular kind of audience. For example, music, or an art form, or a hobby, a particular time in history, a social issue, etc. But just identify who would be affected by and interested in the story. Who are you writing for? Who do you want to reach? This will help set the tone of your text.
Develop Story Structure (45 minutes): Fourth, think of at least three (or more) sections with the broad unbrella outline of Introduction, Body, and Conclusion. The introduction is the place where you explain the main aim, what your story is about - the "what" part. It is also the part, where, with a bang, you introduce your most important, most relevant, most explosive, finding or message. The surprise element, the importance element. The Body is where you develop facts, arguments, explorations, and explanations. The Conclusion sums up, and reiterates some of the main findings, and points to the future. Write down a bullet points structure, with just a few words. And then, gradually, begin to develop the text part under each of the sections.
As you develop your main interactive infovis on the whole story, and static info vis for the Body, this section structure and the developing text content will help you to think about the planning of the interactive and static info vis pieces. Also, at this stage, start thinking about the visual layout of the pieces: do some quick wireframing.
Lastly, as you develop the section structure, through your research and reading, start to write down all the references (academic and news sources), your data sources, and your code sources.
In summary, what you produce through each of these steps will all go into developing parts of your Assessment A2 report. Its ok to not completely finish all of these steps today, because a lot of it will iteratively develop and change as you begin to play with your data and develop your info vis pieces - just use the studio structure to start developing these. Also remember, you are developing each of these in tight coupling with your data, so feel free to use Excel to experiment with quick analytics on the data. For example, when you generate an aim or a hypothesis, it might be handy to check quickly whether it is possible to achieve what you want to given the data set you have. The teaching team will come around for design crits on each of these aspects.
Sarkar, S. and Hussein, D.A., 2017, D3 Tutorials for Information Visualisation Design Studio, University of Sydney.
Email: email@example.com, firstname.lastname@example.org