Flipboard

VAST Video Ad Service

Videos are more ubiquitous than ever before: there are virtually no boundaries on how, when and where people can interact with video content. Flipboard as a curation platform brings multi-format content to peoples around their personal interests and passions. There has been tremendous opportunity for video ads to perform efficiently on our platform. At Flipboard, we have been building video inventory since 2013 in the form of transcoding, storing, and serving proprietary format video ads on an in-house platform. As we see high-quality beautiful video ads satisfy advertisers, we also face the challenge of scaling the platform to meet the increasing demands of our video ad inventory. To bring Flipboard’s video ad inventory to a larger group of advertisers, we need a programmatic selling channel that supports standard video ad formats, provides easy performance measurement and flexible video behavior control. This is where VAST can help.

Detecting Trustworthy Domains

High quality, truthful, diverse and informative content is Flipboard’s #1 priority. Hand-picking trusted sources guarantees quality, but is very time consuming, and can potentially miss out on the multitude of excellent but smaller publishers. To address this problem, we’ve developed a machine learning (ML) system called the Domain Ranker. Its goal is to automatically distinguish authoritative domains from plagiarists, spammers and other low quality sources. It learns to predict how our editorial team would label a domain by analyzing the content and the signals generated by our users. The Domain Ranker scales our editorial thinking to a much larger amount of content than we could handle manually, ensuring high quality across all topics.

Clustering Similar Stories Using LDA

There is more to a story than meets the eye, and some stories deserve to be presented from more than just one perspective. With Flipboard 4.0, we have released story roundups, a new feature that adds coverage from multiple sources to a story and provides you with a fuller picture of an event.

Introducing PSync

Here on the Android team at Flipboard, we have a lot of settings for users to adjust their experience. If you throw in internal settings, we have about 100 total preferences to manage. This is a lot of boilerplate to maintain, because preferences in Android have no built-in synchronization (unlike Resources). Our 1 PreferenceFragment class has a couple hundred lines of boilerplate fields at the top where we keep in-code mirrors to these preference values. This design is tedious, brittle, and requires a lot of overhead to keep in sync with XML. We developed a Gradle plugin called PSync to solve this problem.

Presenting BottomSheet

At Flipboard, we love building visually stunning and highly interactive UIs. When building these UIs, we tend to build them as fairly stand alone components. This makes it very easy for our developers to implement a similar interaction model and aesthetic across the whole product while working in parallel. BottomSheet is a UI component we developed to facilitate a new interaction model for saving an article to one of your magazines (otherwise known as “The Flip UI”).

Image Scaling using Deep Convolutional Neural Networks

This past summer I interned at Flipboard in Palo Alto, California. I worked on machine learning based problems, one of which was Image Upscaling. This post will show some preliminary results, discuss our model and its possible applications to Flipboard’s products.

NSUserDefaults Performance Boost

Since iOS 8 was released we’ve noticed some sluggishness when using Flipboard in the simulator. When taking a trace with Instruments in normal use we noticed a significant amount of time was being spent in 1 CFPreferences.

Introducing GoldenGate

You might not know it, but both Flipboard for iOS and Flipboard for Android make heavy use of web views. We use web views so we can ensure consistent designs for our partners’ articles across all platforms. Communication between native code and the JavaScript code running in a web view is something that is both tedious to implement as well as very bug prone as it’s mostly just string concatenation. Today we are releasing a library to make this task easier when developing Android applications which use web views.

60fps on the mobile web

Flipboard launched during the dawn of the smartphone and tablet as a mobile-first experience, allowing us to rethink content layout principles from the web for a more elegant user experience on a variety of touchscreen form factors.