Design & Deception: A Crash Course in Data Distortion

[mozfest-bot]

[ UUID ] 4d1ab6ea-3351-412a-b1bf-b5862e84de53

[ Session Name ] Design & Deception: A Crash Course in Data Distortion [ Primary Space ] Decentralization [ Secondary Space ] Open Innovation

[ Submitter's Name ] Jess Sauer [ Submitter's Affiliated Organisation ] InfluenceTX [ Submitter's Github ] jessvsmachine

[ Additional facilitators ] Gilbert Slade,Robert Friedman

What will happen in your session?

We’ll begin with a group brainstorm of “best practices” for data distortion, covering design and data science methods for transforming raw data into propaganda and highlighting the overlap between unintentional misrepresentation and ill-intentioned manipulation.

Participant groups will each receive a data set and a description of the group they represent and its political aims. Each group will apply the methods they've brainstormed to manipulate their data set and meet their political goals.

At the end, each group or participant will present their work to the rest of the workshop, who will guess at what methods they employed. Finally, each team will reveal their data set and how they’ve distorted it.

What is the goal or outcome of your session?

We aim to illustrate that those who own the data own the message while giving participants a vocabulary they can use to question the methods used by propagandists to achieve political aims. By putting themselves in the shoes of someone intent on misusing data, participants will learn about ethical information design and the importance of open data.

If your session requires additional materials or electronic equipment, please outline your needs.

We’ll need a projector and, if possible, four laptops so that any groups without one can access the data set and visualization tool.

Time needed

90 mins

[omnignorant]

@dvigneshwer @cubicgarden flagging this one for y'all to keep an eye on. @jessvsmachine is a real gem with invaluable experience in UX & big data, big business.

[cubicgarden]

Feels better in the privacy & security space?

[jessvsmachine]

Hi @cubicgarden—our thinking in choosing this space was our primary concern that “black box” datasets and algorithms ensure that data owners have complete power to shape the message through data and design decisions. The more limited the access to that data, the more consolidated the political power. This was our thinking in having each group “own” a unique dataset and manipulate it with a certain end in mind, then revealing the dataset to expose the distortions to analysis and critique. We’re totally open to another space, though, if you think it would be a better fit. What's your thinking on privacy and security?

[cubicgarden]

Thanks... for the explainer @jessvsmachine

[dvigneshwer]

@jessvsmachine Thanks for the proposal. I definitely feel that decision science is a next gen skillset which people should focus on for building their muscles to become more informed.

I was curious about learning more on the following questions:

• How do you plan to help the audience generate insights? I am assuming you would share some simple scientific notebooks with basic data science techniques
• What kind of dataset do you plan to distribute?
• Would you be interested in selecting some open data set published by government organizations to evaluate policies and discuss the power aspect of the space narrative? I am happy to collaboratively work with you on this idea.

[jessvsmachine]

Hi @vigneshwerd! Yes, for insight generation, we planned to use pre-written Python notebooks as transparent as we can make them. As to the datasets we distribute, your suggestion is very much in line with what we were thinking. I've been browsing data.world's international datasets for ideas but haven't narrowed it down yet—for example, there is a wealth of UN Development Programme aid data that could be interesting to combine with different population health metrics over time. We're not tied to any one idea and would love to collaborate with you! Thanks for the offer. Are there any issues you think might be particularly enlightening to show through a prism of different agendas?

[dvigneshwer]

Thanks for the explainer @jessvsmachine

[maboa]

@jessvsmachine - we love your proposal - thanks so much for submitting and following up in detail. We have a limited number of travel and accomodation stipends to go around and these are now pretty much all assigned. Do you think you be able to come and run this session without the aid of a stipend?

[jessvsmachine]

@maboa Thanks for the kind words! Unfortunately, I would not be able to attend without a stipend.

[cubicgarden]

Thanks for getting back to us @jessvsmachine

[amirad]

@cubicgarden @maboa @vigneshwerd - are stipends still being allocated for this space? Want to do my best to get Jess to Mozfest and trying to understand the options here.

[dvigneshwer]

We will get back to on this @amirad

[amirad]

Thanks @vigneshwerd - appreciate it!