Mark Boas

Trint

---

Trint was our first attempt to apply Hyperaudio technology to a commercial project. Gathering the same team that had worked on Hyperaud.io we started work on a tool for journalists that would allow them to edit machine generated transcripts while maintaining those all important word-level timings.

Maintaining word timings meant we could layer a number of useful tools on top of the editor, including: timing of highlighted text, automatic caption generation, playback of highlights and Interactive Transcript export.

As Chief Technical Officer, I was responsible for selecting the technology platform (React, Node.js, MongoDB) as well as communication and logistical platforms for running a business and a remote development team. Risk assessment, Security and Research and Development were also part of my responsibilities.

As Product Manager I was deeply involved in establishing the value-adding features for the Minimum Viable Product, defining the User Experience, writing Use Cases / User Stories and rolling out a beta that we could be confident charging people to use.

Trint went from conception to paid product in approximately 18 months - along the way we were invited to take place in BBC Worldwide Labs and were successful in applying for a Google DNI grant.

Blog posts

• The Road to Trint
• Trint — Lessons Learned

• Activity: Start-up
• Date: Oct 2014 - Mar 2017
• Role: Co-founder, CTO, Product Manager

Close

Palestine Remix

---

Having been exposed to timed-transcripts via the various interactive news pieces we created for Al Jazeera, we were asked to take the Hyperaudio concept further and promote some of the great documentary pieces on Palestine in the Al Jazeera archive.

We used code from the Hyperaud.io, repurposed and enhanced so that 'Remixes' could be completely created within the browser without the need to be stored in a database. We added both Twitter and Facebook 'snippet' sharing and created transcripts for 28 documentaries in 4 languages, including Arabic. In an attempt to make Palestine Remix mobile compatible we also created Hyperaudio Lite.

The project has seen 4,517 remixes so far! Palestine Remix also won a Webby, was Adobe Muse Site of the day and was winner of the Technical Innovation category at The Drum Online Media Awards.

Blog posts

• Remixing Palestine
• An Interview with Rawan Damen of Palestine Remix

• Activity: Content Awareness
• Date: Sep-Nov 2014
• Service: Project Planning, Development

Close

Hyperaud.io

---

Funded by the Knight Foundation, Hyperaudio Inc was created as a non-profit technology platform that explored the application of timed-transcripts as a way of increasing accessibility and promoting media literacy through use of the remix. All code released under open source licenses.

Sponsored by the Mozilla Ignite initiative, Hyperaud.io was selected to take part in the Gigabit Challenge and part of Mozilla's WebFWD incubator.

We worked closely with Chattanooga Public Library to apply Hyperaud.io to oral history content and encourage kids to tell their own stories through remixing.

Work continues on The Hyperaudio Project in collaboration with the Studs Terkel Radio Archive as we bring 1000 hours of audio content to the Hyperaud.io platform.

Blog posts

• What is Hyperaudio?
• The Hyperaudio Pad — Next Steps and Media Literacy
• Hyperaudio Inc — Why We’re a Nonprofit
• The Kids Are Alright
• Opportunity Knocks
• Why Hyperaudio is Good For Journalism
• Make it Your Own
• Remixing Palestine
• Kickstarting the Studs Terkel Radio Archive
• The Road to Trint

• Activity: Technology showcase & promotion
• Date: Apr 2013 - Present Day
• Role: Co-founder, Project Manager, Developer

Close

OpenNews

---

In 2011 The Knight Foundation and Mozilla got together to create the OpenNews fellowship; a new initiative intended to encourage developers to venture into newsrooms and apply their unique set of skills to journalism.

Interested in how new forms of media on the web could be utilised by news organisations, I applied and was one of five chosen from 250 applicants to become a 2012 OpenNews Fellow. I had the good fortune to be placed with Al Jazeera English; based in Doha and spent 10 months working with them — both on and off site — to create some unique types of interactive pieces to be published on aljazeera.com.

I had the opportunity to experiment with a new form of medium called Interactive Transcripts; timed-transcripts that allow readers to navigate, search and share spoken word content. The result of this experimentation was a number of interactive presidential debates which we published prior to the 2012 US elections. Certainly a rare opportunity to be able to try your work out on so many eyeballs.

I retained a close relationship with Al Jazeera after the fellowship came to a close and introduced two fellows to the Interactive Team in Doha. Together we worked on further interactive projects involving word-timed transcripts and brought the first Data Journalism course to Al Jazeera.

I also had the privilege of visiting the AJ+ headquarters in San Francisco as that initiative was just getting started.

Blog posts

• First Steps with Al Jazeera
• New Ways to Consume Video
• Drupal and Al Jazeera
• Newsrooms and Shipping
• Hack ON!
• News and Technology - An Intoxicating Mix
• Media - Doing it Live
• Shipping and Nipples
• Second Screen at the Mozilla Festival
• Bringing Data Journalism to Al Jazeera
• Reflections on OpenNews - Technology and Journalism
• Do it again, but better!

• Activity: Fellowship
• Date: Feb 2012 - Feb 2013
• Role: Developer, Journalist

Close

Breaking Out

---

Breaking Out was a project I worked on together with my colleague Mark Panaghiston and BBC R&D as part of their Perceptive Media experiments.

The brief: to create a radio story that would contain differing content depending on the listener's circumstances. Taking into consideration inputs such as location, weather, cinema listings and even social media logins we adapted the dialog to create a unique experience for each listener.

This project involved a great deal of real-time audio manipulation of both content and effects. Taking advantage of the newly minted Web Audio API (of which I was part of the W3C Working Group) we were able to "compose" audio on the fly. We used libraries such as speech.js to create speech from text and various "convolutions" to add ambience to the acoustics. We even managed to pull in, manipulate and mix BBC Radio Streams.

The result was an ambitious concept successfully brought to life and demoed around the UK as an example of how media can be manipulated to fit your circumstances. It also served as a reminder of the power of aggregated information.

Blog posts

• Breaking Out – Web Audio and Perceptive Media
• Breaking Out - An Audio Experiment
• Breaking Out – The Making Of

• Activity: Research and Development
• Date: Mar 2012 - May 2012
• Role: Researcher, Project Manager

Close

jPlayer

---

It was while researching an idea for an audio-based web-service that the idea for jPlayer came to me. It was 2009 so there was no web-native audio element. There were a few audio libraries that used Flash, but none that would allow me to style the player with CSS and HTML. I knew that we'd done a lot of work with Flash-based audio in the past and so I asked my colleague Mark Panaghiston whether he thought it was possible to use Flash just for the audio part and to some how set up and command the player through JavaScript. It was and jPlayer was born.

I remember the early releases with fondness; I once created a quick post on Reddit.com asking for feedback and had gone out. I came back later that evening to find a huge Reddit thread and jPlayer going viral on Twitter. It was quite an evening.

Over the years jPlayer became a leading audio (and later) video library. As soon as web-native audio was available we switched to using it, with Flash being used as a fallback.

Mark P was fanatical about support, often writing code for developers that were struggling - the virtuous feedback loop we created on the jPlayer Google Group (now 7000 strong) helped us keep up with the latest browser quirks. In total we calculated that jPlayer was downloaded well over a million times and used by leading audio companies such as Pandora. We had engineers from browser manufacturers following the community to see how people were using audio on the web.

jPlayer was very much a community and open source activity which consumed a huge amount of time and energy. In the end, due to financial obligations we had less and less time to spend on it. It's only now that we are returning to jPlayer and a new React based plugin. The hope is to make the project sustainable by selling skins and themes on jPlayer.it

Blog posts

• jPlayer: the CSS styleable jQuery audio player plugin

• Activity: Open Source Library and Community
• Date: Jan 2009 - Present Day
• Role: Project Coordinator

Close

MDN

---

After running an HTML5 Audio and Video course for the W3C, teaching Masters students Web Media at the University of Florence and being asked to put together a course for Frontendmasters.com I found out that I quite like writing course material and — dare I say it — documentation.

So when the opportunity came along to write for the Mozilla Developer Network, I jumped at the chance. Tackling a series of difficult subjects such as game audio, buffering and streaming media.

Writing clear documentation with instructive examples can be challenging but it's a process that I've always found rewarding. If you're lucky you'll get to write the definitive resource on a subject as I did on audio for HTML5Doctor back in 2009.

Blog posts

• Media buffering, seeking, and time ranges
• Web Audio playbackRate explained
• Live streaming web audio and video
• Cross-browser audio basics
• Audio for Web Games
• HTML5 Audio — The State of Play
• Native Audio in the browser

• Activity: Documentation and Courses
• Date: Jan 2009 - Present Day
• Role: Writer and Presenter

Close

Radiolab

---

Soon after attending the very first Mozilla festival and by initial experimentation with timed transcripts, I was put in contact with Henrik Moltke - a journalist working with Mozilla who had been doing something similar under the name of "Hyperaudio".

We were given the opportunity to work with Henrik and the Mozilla community to create a showcase demo for one of WNYC's Radiolab programmes, which I view as Podcasts before Podcasts were really cool and often done live in the studio.

Uniquely we used Soundcloud as our Content Management System and liaised with Soundcloud developers on the design of the demo. Again my colleague Mark Panaghiston was instrumental in developing the core code that would make things tick.

Blog posts

• Further Experimentation with Hyper Audio
• Activity: Demo / Proof of Concept
• Date: Feb - May 2011
• Role: Project Manager and Developer

Close

University of Florence

---

Up until very recently the University of Florence ran a Masters of Multimedia course at RAI headquarters in Florence. I was asked to put together and present a module on Web Based Media and thought it would be a great experience. Which it was.

I found that it was a lot of hard work, but ultimately satisfying to put together a course module for students that were genuinely excited by the subject.

I learned so much from the experience and the students that when I was asked back to do give the same module the following year I was extremely happy to do so.

• Activity: University Course Module
• Date: June 2013 - June 2014
• Role: Development and presentation of course material

Close