Earlier last month, a security researcher called Victor Gevers sent out a tweet stating that he had stumbled upon a number of open MongoDB databases hosting what was apparantly logged chats from a number of Chinese applications such as Weixin (WeChat) and QQ:
Around 364 million online profiles and their …
About two weeks ago, Hacker News was discussing a new R package called brickr, which relies on another relatively new and excellently crafted R package called rayshader to enable its rendering goodness. The latter package was originally constructed as a means to generate good-looking “2D and 3D hillshaded maps of …
Update: this post now has a follow-up with some further details.
Throughout the past year or so, both R and Python focused data science practioners have picked up on a paper by Friedman and Popescu from 2008 entitled Predictive Learning Via Rule Ensembles. Interestingly enough, the “meat” of the paper …
It’s time to setup a new laptop, so once again I’m jotting down a list of apps I normally install right away, slightly updated again.
Skype is avoided, use web version once again.
I think that Age of Empires II is probably the game I’ve been playing on or off for the longest amount of time. In fact, it is this game which pushed me to write a custom tool to prevent DirectDraw palette issues more than eight years ago, before custom …
Getting a reliable, real-time screen of an Android device is more challenging than one might think, e.g. for the purpose of training a reinforcement learning agent.
This post describes a basic approach:
adb shell screencap -p
Sadly, adb mangles binary output and has a tendency to convert \n to …
A while ago I was interested in performing some HTTPS MITM analysis on an Android app, without having to rely on a real device but instead using an Android Virtual Device, or in other words: an emulated device. It turns out that this is surprisingly tricky with a lot of …
Say you have video such as this one on Facebook. Most of these come with hardcoded subtitles, which we’d like to extract (for whatever reason).
Opening the page in Chrome and inspecting the network requests quickly uncovers the media URL:
https://video-bru2-1.xx.fbcdn.net/v/t42.9040-2/10000000_703970666457205_4416008794480312320_n …