socialcooling

Joined 9 years ago

Karma
119
Submissions
42
Member since
Jun 19
Privacy professional
View on Hacker News

Recent submissions

Comment · 2 years ago
Comment · 5 years ago
I've created an educational game in which you can try to fool a face recognition algorithm by making funny faces. It's 100% privacy friendly, running locally in your own browser.
Comment · 6 years ago
Technically. But as a European I'd already be in a lot of trouble for breaking the law, since the website promises to not collect personal data.

What you can do is check the javascript code for any 'http'. So open https://www.hownormalami.eu/main.js and then do CTRL-F and type in 'http'. See if any of the things you find seem to call home. Also scan over the code visually to check for any obfuscated code patterns which could hide additional instances of data transmission. In this case you won't find any shenanigans.

In general I applaud your critical stance.

Comment · 6 years ago
The main one is FaceApiJS.

One I trained myself (BMI), and the others I just scavenged from existing Github projects. So with most of these I don't know how they were trained, what photos they were fed, etc.

Comment · 6 years ago
That's difficult to say. You could start with getting to to know an issue and specialising in it. Reading books from academic writers (not the stuff sold at airport bookstores). Read Slashdot. Once you've got a grasp, think about how you could translate what you learnt so that your mom would understand it. Think about what your mom likes: quizes, human-interest stories. Lighthearted stuff. Then create your own translation.

It's probably easiest to connect with a local group of people who care about these issues.

Comment · 6 years ago
Ah, so that's it! Thanks!
Comment · 6 years ago
Comment · 6 years ago
Actually, the checkbox does nothing. It's just for show :-)
Comment · 6 years ago
Yes you are correct.
Comment · 6 years ago
You may enjoy this blogpost I made that goes into how the BMI prediction algorithm was developed. It's the only one in the project that I trained myself.

https://www.sherpapieces.eu/overview/predicting-your-bmi-fro...

Like you said, muscles matter quite a bit. It's attractive to use photos af athletes to train the algorithm, because there are a lot of websites that show their photos along with their weight and height (which you need to calculate a BMI). But since muscles weigh so much more than fat, athletes have high BMI with slim bodies, which warps the algorithm.

I didn't use athletes for this reason (it's trained on a diet of 50% Chinese celebrities and 50% American arrest records). But I found projects that do use these photos.

Comment · 6 years ago
Yikes. I'll look into it. I take it you tried the button in the top-left to switch?

Were you able to select the other camera if you reloaded the page?

Comment · 6 years ago
Creator here. It sounds like the main face detection algorithm (faceApiJS) wasn't able to load somehow. Then it will just try to keep you entertained by spouting funny loading sentences.

Try refreshing the page, or using a different browser.

Comment · 6 years ago
Creator here.

The project was created as part of my role as artist within the Sherpa consortium. It's a Horizon2020 research project whose goal is to figure out what Europeans believe are the biggest issues around AI that we'll experience in 2025. Most of the other partners are universities, and I'm the lucky one who gets to translate what we learn into art pieces.

Learn more about Sherpa here: https://www.project-sherpa.eu/

All the other works I've created / am creating for Sherpa can be found here: https://www.sherpapieces.eu/

Besides working as an artist I also work as a privacy designer, which means I specialise in creating things/products that protect privacy.

This website was a very cool challenge to build, a testament to the power of javascript.

Live from the official Hacker News API

A modern reader for Hacker News · SvelteKit + Skeleton