Transforming Media Consumers into Media Producers

Think about how often you read text (hint: you're doing it right now).  Consider often you write text, and yes, email counts.  Text is the de rigeur agent of interaction on the Internet.  Even with the rise of digital multimedia, text is still the regent of digital communication.

Now think about how often you view video (including television).  Again, it's probably quite frequently.  But consider how often you create video.  Even among those that work in the digital multimedia domain, the number of people that create video is minute.

Yet, both capturing and sharing video has never been easier.  Video capture cameras are becoming increasingly common in cell phones.  Nokia is the world's largest digital camera manufacturer, eclipsing even single-purpose digital cameras.  Additionally, Internet content-distribution systems like YouTube and GoogleVideo make sharing media with others simple and virtually cost-free.  Despite all of this, the gulf between content consumers and content authors is as expansive as ever.  Why are so few of us inclined to create media, especially video?

Marc Davis of the Garage Cinema research group at UC-Berkeley asks this question in an IEEE Multimedia article regarding the relative failure of video editing software to address the needs of the amateur content author.  Simply put, contemporary video creation software tools are based on the professional media creation workflow.  The problem with that is the professional workflow relies on skills of dozens of specially trained individuals.  Only the most dedicated of hobbyists are willing to take on the onerous task of performing all this work singly.  To make media creation more accessible, we need new tools and workflows that are better suited to amateurs, and they needs to be more sophisticated than just a simplification of professional tools.

We decided to take Davis' conclusions and examine what this might mean for mobile content authors.  How can we leverage the rich data that exists on a mobile phone while still allowing content creation to be accessible?  We decided to synthesize two different techniques for improving media creation on mobile devices.  Creating a system called Scheherazade (named after the heroine storyteller from The Book of 1001 Nights), we are exploring how to better facilitate those interested in the spontaneous and serendipitous media capture opportunities allowed by a ubiquitous, multi-purpose capture device like a mobile phone.

The first aspect of Scheherazade is utilizing the various context-based information available on a mobile phone to create better metadata for media on that device.  GPS/GSM can be used to determine the location a video was captured, who captured it, when, whether it was inside or outside, what the weather was like and with some simple annotations from the user, what the video portrays.  All this information can be associated with the media, allowing for better archiving and searching of media.

The other facet of Scheherazade is endeavoring to encourage the reuse and repurposing of media artifacts by allowing for collaborative narrative creation.  When creating a narrative, why create capture all the footage yourself when it's entirely possible that someone else has captured some of the material that you desire?  When combined with the above metadata, we would like it to be very easy to search for media by various criteria.  If one was creating a narrative about Vancouver for friends/family that live elsewhere, one could search for existing content about Stanley Park, Granville Island, downtown, etc.  Using this, they can supplement and personalize the content, repurposing media that others have created.  Content they create will be similarly annotated and made available for others to reuse (if the author desires to share it, of course).

This work was inspired by an event that Nokia hosted and arranged at the Come Out and Play festival, called Manhattan Story Mashup. We wanted to expand this system to video content, as well as making use of the contextual information available on capture devices. 

Scheherazade is still in prototype development, but we will continue to update our progress.  Hopefully Scheherazade will provide insight into how mobile media creation can be improved and MAGIC can utilize this information to enable the content consumer's metamorphosis into a content author.

Author: Nels Anderson, MAGIC