I'm following Chandrababu Naidu's interest in IT-enabled movie distribution in India. This early 2000s initiative explores digitizing cinema delivery to replace traditional physical film prints, marking a significant shift in the media distribution landscape.
I found that force-fitting a normal distribution to banking and bond data drastically understates tail risk. By ignoring power-law distributions, I underestimated worst-case scenarios that actually occur much more frequently than the bell curve predicts.
I used to normalize everything from work performance to movie ratings, but I’ve realized that blindly assuming a normal distribution is often incorrect. Many real-life variables, like stock prices, violate the Central Limit Theorem’s requirements.
I explore how to use Bayes' Theorem to update a probability distribution iteratively. Starting with a flat prior, I show how successive coin toss results refine the Beta distribution to better estimate unknown likelihoods.
I explore the MPAA's decision to use digital fingerprinting against P2P movie sharing. This shift indicates the industry is finally moving from legal regulation toward technological solutions to manage content distribution on file-sharing platforms.
I built an interactive jigsaw puzzle using jumbled movie stills. You can drag and move the blocks to reconstruct the scene and identify the film. It’s a simple web-based quiz format for movie buffs to test their recognition.