Product Designer

Disintegration Anxiety

Design Journal | code | thought process | Class Homework | Project Documentations & more. WIP workspace. Current Portfolio at: atharvapatil.github.io

Population infinite: Obfuscation

With me spending most of my time on the browser looking up stuff this is the apparatus which collects the most amount of information about me. For this weeks assignment I decided to create a fake search history in order to confuse the machine that analyses. For this I chose to use the puppeteer package we saw in class.

There was one problem if I chose which words I want the system to search I will be biased against the words I use. There is still a chance of my personality leaking into the search terms. So I used another package called, ‘random-words’ to get the words for me and search for them.

This is how far I got. But I know there are 2 problems with the obfuscation approach I am taking. I wouldn’t search for completely random words and I won’t do search for them within a 2-3 minute window.

WhatI would need is something which generates a larger amount of text which is deviant enough but not too deviant(I guess mathematically it would something similar to Perlin noise). My searches begin to deviate but not drastically. They keep deviating as though I am researching something.

The second will be I won’t search for hundreds of things at the same time I would need some delay which would make me distinguishable from a machine.

To improve what I have made currently I would like to generate better fake text and add a human like quality along with it. For the current assignment the code is here: