show and tell for hackers in DC.

Round 41: Polka Sprockets

16 Feb 2017

Chris Nguyen - @uncompiled

Credibility in the Age of Alternative Facts

https://github.com/factcheckthenews/

Made this thing: https://www.factcheckthe.news/

Alternative facts (total bullshit) have real consequences – see the Comet pizza shooter.

There are many browser plugins, but most people don’t use those – what if there was a website that you could plug in the URL and get a summary of whether this is a fake website?

This uses a “Credibility Checklist” using red flags to figure out whether that web page is fake or not, using things like HTTPS, deceptive domains, Open Sources, Web of Trust.

Open Sources is a curated list of online sources: http://www.opensources.co/

How many words are in all capital letters? Are there excessive exclamation points? These are signs that it’s probably fake news.

How about BuzzFeed? Breitbart? They score around 80% likely to be credible; they have regular readers who find their articles credible, so that makes sense.

Heuristics: https://github.com/factcheckthenews/heuristics

Alexandra Ulsh - @AlexUlsh

A quick personal security checklist to go through for securing your devices and accounts.

https://github.com/alulsh/personal-security-checklist

She knows her stuff - she’s Information Security Engineer at MapBox!

This is a great resource for taking care of your personal digital security.

Quick concise format of the most important things that you need to do – aimed at regular people, not developers specifically.

Submit issues and pull requests!

If you’re trying to get your folks doing better security practices, and they only have patience for one thing, get them on a password manager.

P.S. also check out

https://github.com/alulsh/intro-to-security-for-developers

Aaron Schumacher - @planarrowspace

I’m making a Go program! It is not as good as AlphaGo. I call it OmegaGo! I hope to have Monte Carlo Tree Search implemented in time to present.

https://github.com/ajschumacher/omegago

Go is a game that is thought to be difficult for computers to beat humans, but that’s changing – Google’s AlphaGo recently beat 4-1 the best human

Algorithms used: Reinforcement learning…, Thompson sampling…., Monte Carlo tree search: after we make this choice, how well did we do? by then sampling and trying things that have been successful in the past, you ideally see what happens when you play against an intelligent player and extend the tree gradually.

Lizzie Ellis - @lizmeister321

Film Club - bringing the book club into the 21st century, but with movies. I’m starting a film club designed for remote participation – pick a movie, watch a movie, discuss a movie, using web tools. Will present on what i’m using to start and what my long-term goals are. Looking for group feedback as well!

Remote component so anyone can join! Why not invite all our friends to participate?

How it works: Movie selection email goes out on Friday with film links Two weeks to watch the movie In-person discussion with remote participation available; Online comment board open simultaenously Pick a new movie as a group online Repeat!

Movie recommendation: Footlight Parade (the dance numbers, not the rest of the movie so much)

You can sign up too!

https://www.tinyletter.com/lizzie_ellis

https://github.com/lizmeister321/filmclub

Jim Webb - @jimwebb

Making a Donald Trump Swear Jar with Google’s VoiceRecognition API

https://jimwebb.com/swearjar

https://jimwebb.com/transcribe

Uses the Google Voice Recognition API, which isn’t keyed to Jim’s voice, but also doesn’t get any better at Jim’s voice.

In the transcription, it guesses what you’re saying, but can figure out better based on additional context.

Listens for a list of pre-determined swear words like Sean Spicer and Kellyanne Conway, then adds money into your “swear jar.” Once you have at least $20 in your jar, it creates a button where you can “cash out” and donate to the Council on American-Islamic Relations

Automated bingo? Or drinking game? :beers:

Travis Hoppe - @metasemantic

Miniprez – python software designed to turn text into beautiful & mobile-friendly presentations!

This is a python library that takes in text and compiles it into super parsed-down HTML and Markdown.

It creates beautiful presentations with lots of functionality and features, like emoji, looping background animations, math formula support, unsplash support for background images.

https://github.com/thoppe/miniprez