Ivy Turk
Academic Site for Ivy Turk
Project maintained by ivyturk
Hosted on GitHub Pages — Theme by mattgraham
Home // Publications // Academic and Employment History // Teaching Experience // Project Ideas
Notice: I’ve had multiple affiliations with different project lengths, but most ideas here can be adjusted to fit these varying timescales. The suggested timescale for some projects is indicated with [N months], for example [3] indicates a three month project.
While I have previously supervised more technical projects, this list has recently been shortened to better fit my present areas of expertise.
- [3-6] Analyse the effectiveness of different hiding mechanisms for applications. Examples include BrightSky (weather app), Aspire (news app), Krila (period tracker) - all abuse support apps. Examples from other contexts include Grindr (changes icon to hide app) and vault apps (file storage, sometimes provide lock screen with secondary PIN to hide main profile). These hiding mechanisms are praised for allowing an app to be covert when needed, but how effective are they at hiding things? This could be evaluated (for example) with a deceptive user study that has users interact with either a disguised app or the various apps on a phone and see if they identify the hidden use, or a non-deceptive variant asking directly for opinions on the covert mechanisms in place. An extended version of this that systematically analyses many apps could be adapted into a full paper for submission.
- Designing and implementing novel safety tools to aid those experiencing various interpersonal harms
- [3-6] Evidence gathering tools. Goal is an application which provides secure storage of information and media even if the device itself is compromised (e.g. domestic abuse scenarios where the partner has access to the phone). Our prior work on vault apps is here as possible inspiration. Possible extensions could look into ensuring digital evidence is admissible in court, or at least providing integrity and authenticity.
- Covert information sharing tools. Simple implementations [3] could set up a private channel for communications with trusted third parties, or extended versions [6] could implement either existing crypto protocols for secure (+anonymous?) comms or have a go at designing your own and attempting to break it. Extensions could include authenticating endpoints are not compromised (e.g. shared phones) or designing a system for police or NGO contact.
- Any prototype implementation with evaluation for novel privacy and safety tools. Some suggestions might be
- Location privacy and spoofing, building off of novel suggestions from our unwanted trackers research or other contexts.
- Shared access home systems (such as IoT networks) and compartmentalisation, designing for multiple superuser environments and/or allowing for delegation of access controls while considering adversarial scenarios
- Abuse escape planning tools.
System Evaluation
- [3] Evaluating any existing tools (ideas above) for safety, security, usability.