Earlier this year I fulfilled a life-long dream of mine… I entered a piece of my art into an actual exhibition!!! It wasn’t quite my ‘normal’ style (light/ dark or close up oil portraiture is my go-to normally) but I really loved the process of this one and am quite pleased with the final piece!
This summer will be the 10th year in a row that my school friend and I have been to the Summer Exhibition at the Royal Academy of Art so we thought that it was apt that we both submitted a piece to this year’s exhibition. If you didn’t know, the RA SE is the largest and oldest open submission of art and anyone can enter.
When thinking of ideas back in December for the February 2019 Royal Academy of Art Summer Exhibition deadline I was a bit stuck. Given that I was in my last couple of months of Data School training it’s no wonder that I turned to data for inspiration!
I have always loved the way that data can be presented beautifully and that sometimes you don’t even know you’re looking at data!
Another achievement of mine last year was running my first half marathon in October (see here for a post about my 2018 reflections on the DS blog). I had already downloaded my Strava data and had started exploring it in Alteryx and Tableau.
In Alteryx I isolated my 2018 runs and extracted three measures to investigate: speed, distance, time. I had to then scaffold my data onto a dataset that had every date 2018. This is to give me the full days within the charts rather than just showing the ones I ran on – i.e. adding onto my dataset the days that I didn’t run at all (more than I’d like!).
I then played around in Tableau to work out various styles and colour schemes. This is where the art process differed. ‘Usually’ an artist will sketch out different set ups and try various mock ups and colour studies in paint or pencil to see what worked best. For me, I got to use Tableau to do almost all of this which I found not only made the process a bit quicker/ smoother, but also meant I could play around with various possibilities more.
I always envisioned a white background with some clean primary colours to indicate the data without gridlines or axis giving the game away.
I tried bar charts and line graphs but for this particular piece I went with one of my favourite charts: a heat grid. This is where you have different temporal scales across either axis and the data fills in the table with different colours. I personally love it because it allows the viewer to see patterns in the data that you hadn’t realised were there. For example, in other datasets you can see commuter times if looking at time and days of train usage, or, apparently, I ran on the 24th of every month in the second half of 2018!
I then played around with the metrics, initially thinking of doing a triptic or diptic with one each (like in the image above). But there were only very subtle differences in the colours across the three measures so I decided to stick to one metric and looked into different ways of displaying the heatgrid.
I ended up with months on the coloumns and date of the month on the rows. I toyed with keeping the borders in and actually showing the grid, but given the differential in month lengths, decided that having the non-running days sort of blend together in white worked quite well.
I then used Tableau to try out different colours – blue, orange, purple and had to create a category for the distance. I found Tableau tricky as it doesn’t like setting zero as white within a custom colour palette within Tableau, so I created a few custom colour palettes (saved to My Preferences) to try out.
The final choice was a distance heat grid – see below for one with and without numbers – in a dark blue to light blue.
From screen to Canvas
My first attempt onto canvas didn’t turn out quite as nice as I’d planned. I drew the heatgrid onto the canvas with pencil then calculated which ones were to be filled in with what scale of colour. The pencil marks afterwards were hard to rub out, and the white paint that I used for the no running ‘days’ was hard to do.
My second attempt went much better and this was the one I photographed and submitted! I used a big a2 sheet of paper and drew the grid and numbers onto that. I then cut out the running days and laid the paper on top of the white painted, primed canvas to draw around the ‘gaps’. This worked out much better and I could paint more easily without too much ‘cleaning’ with white paint later on.
The final piece
….and the result was….
Sadly I didn’t get through to be hung as part of the exhibition, although I’m so proud of the fact that I got though to the final round and got to actually submit a piece for judging ‘in person’.
I’m really looking forward to attending the exhibition this summer for my 10th year in a row and seeing how other artists answered the Contemporary Life brief.
What have I learnt from digitising part of the art process?
I think that it’s been a handy experience to play around digitally with a composition before putting paint down. It’s something I don’t think I’ve done before and Tableau was great to work with, allowing for little changes to be easily made – and easily undone!
I think I’m glad to have had a go on the first canvas (and now I have a duplicate for my home!) as that allowed me to get the actual technique down versus the styling/ design from the data/ Tableau.
But you never know, there may be others in this series in the future… either with different data, chart types or even some abstract take on the Alteryx workflow!
What a way to kick off my first official blog on here – hope you enjoyed the longer than anticipated read! Catch up on my previous Tableau and Alteryx blogs on the Data School blog here, but I might be re-posting a few of them on this site too as well as new topics.