It's time to talk about consent at #PEPR21 starting with "Designing Meaningful Privacy Choice Experiences for Users" by Yuanyuan Feng, Carnegie Mellon and Yaxing Yao, University of Maryland
Notice and choice is a legal framework. There are privacy notices which tell people about the practices. The controls let people have limited controls.
But in practice the controls are usually difficult to find, overly simplified, and sometimes manipulative using "dark patterns"
But in practice the controls are usually difficult to find, overly simplified, and sometimes manipulative using "dark patterns"
Dark patterns manipulate people into making choices they might not otherwise make. For example, the terms/policy are linked in tiny type and there's only one button: sign up. Any choices are hidden behind this, which is suboptimal.
Or the pre-selected options may not be good.
Or the pre-selected options may not be good.
There's a lack of guidance for practitioners, even though there are requirements in GDPR or CCPA.
This work provides more design guidance.
This work provides more design guidance.
Introduce the concept of "meaningful privacy choices" to make privacy choice experiences better.
* effectiveness -- ability of users to make choices that align with their preferences
* efficiency -- ability to choose with minimal time and effort
* user awareness
* effectiveness -- ability of users to make choices that align with their preferences
* efficiency -- ability to choose with minimal time and effort
* user awareness
* comprehensiveness -- range of choices are wide enough
* neutrality -- choices should be presented in a neutral manner, free of dark patterns
* neutrality -- choices should be presented in a neutral manner, free of dark patterns
[ note that "neutral" and dark patterns are not opposites. The Chrome malware warnings are a really good example of where respectful design requires some extremely non-neutral design ]
privacy choice:
* type (e.g. binary, multiple, contextualized)
* functionality (e.g. presentation, enforcement, feedback)
* timing (e.g. at setup vs just-in-time)
* channel (e.g. primary, secondary)
* modality (e.g. visual, auditory, machine-readable)
* type (e.g. binary, multiple, contextualized)
* functionality (e.g. presentation, enforcement, feedback)
* timing (e.g. at setup vs just-in-time)
* channel (e.g. primary, secondary)
* modality (e.g. visual, auditory, machine-readable)
They provide a list of design options for each of these. They're not exhaustive and can be expanded
Use case: designing a privacy choice platform for Carnegie Mellon IoT assistant app
[ plays video about what this app does ]
[ plays video about what this app does ]
[ There's a BT-based user location tracking service. The video goes through the notice and choices. Apologies, this goes too fast to screenshot. Also apologies that I can't caption pictures fast enough -- please feel free to help and I try to give as much as I can in thread]
[ One thing I notice is that you need to manually refresh -- keeping settings in sync across multiple surfaces and applications is *critical* and people don't think about this enough ]
Q: Where can you find more info about this IoT work being presented?
A: Here's the paper: dl.acm.org/doi/fullHtml/1…
A: Here's the paper: dl.acm.org/doi/fullHtml/1…
[ Another great talk coming up, but I'm going to have to take a hand break again. Catch you again after the #PEPR21 ethics panel! Hopefully someone else can livetweet that one. ]
• • •
Missing some Tweet in this thread? You can try to
force a refresh
