Hi everyone,

This being my last college year(ofc if I don’t fail :P) , I’m looking forward to take part into GSoC ‘21, and I’m interested to contribute to the Tor Foundations, specifically I’m interested in their project: https://community.torproject.org/gsoc/alexa-captcha-monitoring/

What’s the Problem:

We are an international team who believes everyone should have private access to the uncensored web.

Tor fights for privacy and annonymous viewing or accessing of the uncensored web, but a large number of people have reported facing trouble accessing popular websites like: Reddit, Instagram, Google Search as they keep getting a lot of captchas, showing that these websites don’t trsut annonymous viewing. This website : https://antcpt.com/eng/information/demo-form/recaptcha-3-test-score.html, scores browsers according to the reCAPTCHA v3(Google).

All possible values are from 0.9 to 0.1. Where 0.9 — probably a human, 0.1 — bot. Values 0.3 and below are considered as very suspicious activity, these users are interpreted as bots.

It generally produces 0.1 for Tor based browsers. Check out for yourselves, what scores you get :)

The Proposal:

Tracking how frequently The Alexa Top 500 websites return Captchas to Tor clients. Check for differences in non-Tor exits and different Tor exits, and track metrics like:

  • “If certain exit relays are getting hit with Captchas more frequently than others?”
  • “Does location have any specific impact with Captchas?”

Comments in ticket 40013 and the original project ticket could help with more ideas.

Mentors:

Resources: