How can a deeper understanding of images and data lead to artistic interventions in the context of climate change?

/ Five-week Live* Online class begins 29. April ends 27. May

/ Every Thursday, 7pm-9pm, CET

/ Small class of participants

Course Description

The act of producing digital images is now embedded in our everyday life as we share our experiences with internet-connected digital eyes. In 2017, 1.2 trillion digital photos were taken just with our smartphones. Such digital images become networked when we share, like and comment on them online.

Digital images rarely stand alone. When online, they are augmented with a number of metadata (such as hashtags, comments, geolocation, timestamps or likes) and also formatted, filtered and recommended in different ways by the platforms in which they are circulating.

On the one hand, this hyper-connected digital visual repertoire is an opportunity to find new ways of studying online images and visual culture. From a design perspective, there is also the need for new tools and strategies to enable the observation of large collections of digital images. How can we make a folder of images ready for analysis?

Furthermore, considerations can be made on the role of digital images in the processes of a collective environmental memory: not only humans, but also machines (such as satellites) are constantly recording pictures of our planet, generating a tremendous resource that could contribute to the preservation of shared cultural heritage about the changing state of our climate.

In this course we will navigate the stream of such human-snapped and machine-created images and take a stance on this highly interconnected visual context by designing our own image datasets.

“Wherever large amounts of data are collected, there are often empty spaces where no data live,” says visual artist and researcher, Mimi Ọnụọha. Who and what is visible, who and what is not? The making of a collection as such can also become a critical practice, or one of (an ethics of) care.

In this course, we will apply visual methodologies to the theme of climate futures using science fiction as a speculative storytelling genre, as a means to look forward and (re-)imagine the future in a way that is unbound by today’s limited possibilities.

Science fiction as a tool for future speculation is used by governments to imagine terrorist threats of the future, and by tech companies to try and develop the seemingly impossible technologies imagined in movies and literature.

The climate change debate similarly resides in a realm of forecasting and modeling of the future: where are we headed if we do not change our ways, what will disappear, which parts of the world are the most vulnerable for the effects of climate change, what is at stake for the near and distant future?

These questions are of course important and crucial to find solutions. But certain images from scientific literature (the scary graphs), journalism (the polar bears), or even the entertainment industry (the tornadoes), are presented to us over and over again until they stick. To overcome the so-called apocalypse fatigue, researchers, activists, and educational institutions are pointing out the need for finding new stories and images for people to engage with climate change.

In this course you will be introduced to:

· Digital methods and tools to collect images from different social media

· Data curation with spreadsheet softwares

· Automatically tagging your image dataset with computer vision algorithms (Clarifai API and Google Vision API)

· Tools for visual analysis of sets of images (Image Sorter)

· Network visualisation and analysis with Gephi

· Design of Composite Images with Vectr

· New and different images about future imaginaries

Outline

Week 1: Designing image datasets with digital methods

The first week will be dedicated to get to know one another and our work. You’ll be invited to share your expectations for the course, and together we’ll set some goals. We will introduce and discuss our personal research on climate futures, as well as visual Methods for digital images, with a showcase of projects that design and explore collections of images for social and cultural research.

The second part of the workshop will be dedicated to a web scraping tool medley, an overview on tools and techniques to scrape images and metadata from the web and social media, and how to design a dataset starting with a question driven approach.

Week 2: Categorising an image dataset

The second week will start with a round of feedback on your data collection. Each of the participants will present the query and the images collected during the previous week. We will then explore some simple techniques of data curation with Google spreadsheet. In this simple tutorial you will be introduced to the basics of spreadsheet magic. How to import data in Google spreadsheet, how to sort and filter data in different ways. No experience needed!

The second part of the workshop will be dedicated to the exploration of a collection of images through their visual attributes. We will explore our collections through Image Sorter, an image browsing application. You’ll be introduced to automatic content analysis through computer vision algorithms like Clarifai API or Google Vision API.

You’ll be able to automatically tag your images and visualise the most recurring tags in your dataset.

Week 3: Exploring an image dataset

In the third week we will discuss when and why it is interesting to visualise networks, and we will do so by exploring the basics of Gephi, an open-source network analysis and visualization software. We will format a dataset for Gephi, and use different visual features and layout algorithms to visualise the dataset as a network. We will also discuss principles on how to read a network visualisation and make sense out of it.

At the end of the afternoon, you’ll be able to explore and visualise your own image datasets.

Week 4: Visualising an image dataset

We’ll start this week with a round of presentations of your first image networks. Then, we’ll explore a set of recipes for composite images. These techniques are useful to compare differences in formats and styles across (small) collections of images. By overlaying images on top of each other, or by creating an image wall within a spreadsheet, one is able to produce a “visual summary” that can be used to compare multiple collections.

Week 5: Project Work

The last week will be dedicated completely to group work. We will spend time together and work one on one and in small groups with the techniques that you liked the most, or the ones that you’d like to explore further. At the end of the day, you will be invited to present your visual experiments to the class.