Tag Archives: Information is Beautiful

27 Feb

Actors Working in Teams [infographic]

Sometimes I’ll start a data project, and duing the process one of several things may happen:

  • I lose interest
  • something more important comes along
  • I realized it’s just not “working”

The latter happened as I was working on a piece for the latest Information is Beautiful challenge – which involved chomping through a large and very interesting data set regarding Hollywood movies.

These challenges encourage you to use a data set provided by the website to create either a visualization, napkin drawing ( sketch) or an interactive piece.

After several hours of looking through the data, looking for interesting angles, and hunting for more data I could add to the set (via Google Refine) I settled on looking at the connections between the actors involved in the top films.

Ask anyone who’s watched a film with me and they’ll tell you that I have a VERY annoying habit of opening Wikipedia to find out where I’ve seen a particular actor before. It’s annoying in everyday life, but for this – it was a dream.

I was also intrigued by the success of the films starring Seth Rogen / Jonah Hill et al, and the idea of teams working together – and how prevalent this was across other areas of Hollywood.

(see below for more details and why I eventually shelved it)

 

 

 

What it Means

Yes, it’s a bit of a headache isn’t it? The original dataset with featured the films coming out of the major studios, over a certain time period.

I added actor information to the cataset from Freebase (within Google Refine) and worked out which ones had appeared in the most films over the time period. I then cross referenced the films and created the above chart.

I had originally intended to give each film a different colour but this became unworkable – so I limited the colours to the films featuring 3 or more actors on the chart. The rest I coloured in grey.

Why It Didnt Work

For one, it was too damn complicated – no chart should take 3 paragraphs to explain. Kinda defeats the point, right?

Secondly, the choice of resulting entries was nonsense.

  • Original Data (limited to major studios)
  • Actors added to each film (according to Wikipedia, via Freebase)
  • Top 26 hardest working actors selected (based on original list, so ignoring independent or smaller budget films)
And thirdly – it didn’t really say anything. I always put the success of my last Information is Beautiful entry down to the quirky subject matter (comparing the lifespan of important earth resources to celebrities). This didn’t have that element. Yes, it was interesting to see which actors work together, but the dataset was too limited to show any major patterns. 
I eventually decided to stop working on the design (hence a few wonky areas) and shelve it.

So why am I publishing it here?

Because I spent all day on it, I like the IDEA and design and I wanted to share my experiences of when to walk away.

I’d love to hear your experiences of when you’ve had to walk away – and why.

26 Feb

Who is my data idol?

There must be something in the water …. twice this week I have received emails from online journalism students asking for my thoughts on data journalism – and more specifically, my data idols.

The first was an email from Germany – or student Katarina Bons to be precise – asking for information about any studies into data journalism in the UK and who the key players were.

During the completion of my final MA Online Journalism project, during a desperate need to relieve tension, I pasted the text of my MA final dissertation into Wordle. Whilst I regrettably did not save the final image there were some surprising, and not so surprising results. Data, was of course – one of the most commonly used words (so much in fact that the A key on my laptop broke off and flew across the room).

Another word that cropped up, worryingly often, was McCandless.

Poster boy for the data generation, but not without his critics, David McCandless is definitely my data-crush – his simplistic yet visually stimulating work is a definite inspiration to me and – in times of creative drought – I have asked – “what would McCandless do?”. (to make the shortlist of an Information is Beautiful challenge was one of my personal highlights)

Then I received an email, a few days later, from BCU MA Online Journalism student Duarte Romero Varela asking for a recorded interview about data journalism.

Being an alumni of that particular course, and a self-confessed data geek, I was more than happy to hold forth.

Meeting in Cafe Blend in Birmingham, Duarte interviewed me for a podcast (listen here) covering a range of issues including:

  • Q: who is my data viz hero? (A: see above)
  • Q: what tools do I use? (A: Excel > Tableau > Illustrator)
  • Q: what is more important, how clear a viz is, or how it looks? (A: both – a clear ugly chart is like a badly written article – who’d want to publish it?)

This question was of particular interest to me – the tide does appear to be turning against data viz / infographics at the moment, thanks to the tsunami of terrible examples finding their way into web content, newspapers and onto billboards, bus stops and the side of coffee cups.  There are cases, and I am definitely guilty of this at times, of being seduced along a path of beauty, and forgetting about the practicalities and the journalism.

Here’s how I see it …

Journalism: what are you trying to convey? What is the story?

Clarity: it has to tell that story

Design: it has to look attractive for people to want to interact with it

 

It’s a shame that there are so bad examples out there, outweighing the good and giving the whole area a bad name.

There is a real need in some cases for highly complex information to be reworked into a visible format and it would be a shame if we threw the useful bar chart out with the overdesigned viz

All content (c) Caroline Beavon 2020