iiinova — where your digital self keeps evolving

LK Seiling

@lkseiling.eurosky.social

219 followers 306 following 68 posts

DE/EN. 📍 Potsdam / Berlin coordination of @dsa40collaboratory.bsky.social, various research at @weizenbauminstitut.bsky.social among other things: zusammenfuergleichstellung.de

Posts Replies Media

LK Seiling @lkseiling.eurosky.social · 07/05/2026

Just migrated my data to @eurosky.social on servers in Germany - and can keep using bluesky like nothing happened. Next: re-starting to blog using @leaflet.pub. Genuinely excited to see where this goes. In case you are also considering to move 👉 move.eurosky.tech

a screenshot of my Eurosky dashboard, detailing the number of posts, followers, following and other applications also using ATproto

2528

LK Seiling @lkseiling.eurosky.social · 26/03/2024

To lay the groundwork for such communication, we suggest basic information categories to inform users about to enable proper risk perception and better informed decision making.

A table with to columns: "Super Category" and "Proposed Information Category".
The super category "Latent Consequences" includes: Informational Power, Decontextuaizaton, Emergence of Information, Discrimination, Manipulation Identity Theft, Defectivenss of Information, and Permanence of Information.
The super category "Tangible Consequences" includes:
Financial Losses, Stigmatization, Physical Harm, Publicity Damages, and Psychological Harm.
The super category "Risky Data-Processing Operations" includes: Profiling, Automated Decision Making, Scoring, Behavioral Monitoring, Systematic Monitoring of Data Subjects on a Large Scale, Data Collection from Third Parties/Data Combination, Data Disclosure, Large Scale Processing of Personal Data, Storage period, and Off-Device Storage.
The super category "Risky Data Types" includes: Sensitive Data” Genetic Data, Health Data, Biometric Data, Location Data, Unique Identifiers, and Financial Information.
please see the paper for the full table

100

LK Seiling @lkseiling.eurosky.social · 26/03/2024

As our research into privacy risk shows, the processing of personal data can lead to two kinds of consequences: latent consequences and tangible consequences.

At the top of the picture are circles with data processing operations in them (like collection, storage, inference, transmission). They are linked - by downward pointing arrows - to rectangles with dotted outlines (latent consequences, such as Defectiveness of Information, Re-Identification, Emergence of Information, Informational Power, Identitiy Theft, etc.) which are again linked to rectangles with normal outlines (tangible consequences). Some of these rectangles have a bright colouring (immaterial consequences, like psychological harm, manipulation, loss of privacy or publicity damages), some have a darker colouring (material consequences like stigmatisation, financial losses and physical harm). The arrows connecting processing operations to latent consequences and latent consequences to tangible consequences also have red and green text on them, which indicates with aspects could increase or lessen the likelihood of these consequences occurring (for details please see appendix)

100

LK Seiling @lkseiling.eurosky.social · 26/03/2024

To answer this question, we first needed a framework for privacy risk. We merged the Model of Perceived Risk with the concept of Contextual Integrity (by Helen Nissenbaum) into the Contextual Model of Perceived Privacy Risk.

On the left side of the image is a figure which has three columns. The first includes sources of risk (shown as trapezoids with the letter S in them) which are linked to various events (circles with the letter E in them) which again are linked to consequences (squares with the letter C in them). The heading below says "Perceived Risk Model" with the description "a consumer perceives risk as some source exposing them to different types of harm through an event". Below this description is a plus sign that connects the Perceived Risk Model to another heading: "Contextual Integrity" with the following description: "contexts are defined by their ends and values, the entities involved in information exchange, the type of information exchanged, and the privacy norms within this context". In the middle of the image is an arrow to the right, pointing to the combination of both approaches: the Contextual Model for Perceived Privacy Risk.

100

LK Seiling @lkseiling.eurosky.social · 26/03/2024

10 years ago, researchers still included evaluative information in their proposals (some icons even made it into a GDPR draft). Today, such evaluations are gone. What we see instead are topic icons, acting like a headline (helping to find relevant info) but doing little more.

The figure has a heading with says "Privacy Visualisations" and is split down the middle into "then" and "now".
On the "then"-side are image of the Privacy Label (a big table that highlights fields representing potentially risky usage with a white exclamation mark on red background), PrivacyBird (a bird of either green, yellow, red or grey colour depending on the agreement of the data processing with the user's personal preferences), and icons proposed in a GDPR draft (circles with a red outline that use red and black icons to indicate the danger of collection, storing, selling, sharing and other processing operations). The "now"-side shows Apple's and Google's current icon implementations which are bland and either grey, black or blue - containing no more evaluative information.

100