Online news video does a poor job of making itself available to viewers and thinking about how they access it. In order to build audiences and grow their business, newsrooms need a better way of structuring video to serve the public, because viewers need trusted, reliable and accurate information from news gatherers in a timely manner that comes to them on their terms.


A brief case study

On the advice of #MozNewsLab participants and their feedback, I asked patrons at a Starbucks in Seattle if they were aware of news video online and whether or not they watched it.

My unscientific sampling confirmed what I had already learned from years of producing video for newsrooms; most people don’t know even to look for news video or those that do have no idea where to find it.

Newspaper websites run into this problem for three reasons

  1. Virtually all news video today is an embedded flv file which web crawlers can’t see inside, only what is around it, such as a title and summary paragraph. HTML5 video will change some of this with more robust SEO, but it still won’t be a good solution, because search engines still won’t be able to scrub the content of the video.
  2. Newspapers chronically forget to include video with the story or only link to a video page, which means even if I search a relevant written article I may not see correlating video. So videos are lost to the void and no combination of search ever finds it again.
  3. Last, search tools on newspaper websites rarely return any link to video. If news organization’s own website can’t find any video, the probability that a second or third party search engine will find it on the web at large won’t be much better.

Virtually all newspapers produce some sort of video and then forget about the last mile and having investments pay off, by not making it easy to find, not monetizing it and sometimes just forgetting to include it on the page with the story.

There needs to be a method of crawling videos so that search engines can find them from here to eternity, that’s where VidScribe come in.

We need a text to video relationship to make it search friendly. We need to map the data to the timeline to allow users access to a better service and open it up to new forms of communication. Imagine clicking on a story, a link, a comment — everything right on the timeline in a non-scrolling simple interface. That’s thinking forward and planning for mobile, ipTV and other display technology yet to come.


Designing a better wheel




VidScribe will be developed over two stages with a final stage of open source support and upgrades for new functionality.

Stage one will be focused on laying the ground work and establishing a standard for newsrooms for implementing metadata, transcription, ensuring accuracy and focusing on search, navigation, social sharing and archiving for future use.

Stage two focuses on making use of the video and the community with deep integration of features such as timeline bookmarking, link curating, story embedding, timeline commenting, social profile integration, branching multitrack audio, heatmaps and scaling to set top boxes and platforms such as ipTV.

Stage three is where it gets interesting, as by this point users will have learned how our framework is structured and start to create their own advancements to the engine. Then much like Firefox, WordPress and other open source projects, VidScribe will focus on supporting this community with education, service, collaboration on experimental projects and support so together we can add functionality and create a better audio video experience online.

This course avoids delayed releases and rushing to release everything without taking time and care to implement a solid framework to build upon. It also gives news producers time to experiment and discover what they’d like added and then collaborate with developers. For users this provides an avenue of becoming familiar with the structure and customizing it to their needs with plug-ins they are encouraged to suggest and create.

Open source for faster, better results

By going open source, a wealth of possibilities I haven’t considered come into play and with many hands coming at the same problem from many different points of views, VidScribe has the opportunity to bring about a paradigm shift in how we create, consume, collaborate and channel online news video.

Within the #MozNewsLab I’ve already found many collaborators: Mark Boas, Nick Doiron, Julien Dorra, Samuel Huron, and many more are working along these same lines and together we’re all sharing ideas and working toward the same goal.

VidScribe can become a toolbox for all these tools to the benefit of everyone.

The big question is how to organize the data with video, to keep it perpetually associated, easily updated globally in all instances and easily transferred, published and accessed from anywhere on the web by news video producers, collaborators and viewers.

The short answer is I won’t know for certain until a prototype is built.

In collaboration with other #MozNewsLab participants, we’ve collectively floated a few theories as how to propose a solution.

My logic right now is to flip the expectation of having video at the top of the hierarchy and instead of having video call to metadata, a metadata script will instruct web browsers how to assemble the video interface by calling to video files and other data sets.

This way the script files become a standard tool set for organizing and presenting video and can be expanded with plug-ins and new lines of code to add more features.

To make it simple and intuitive to use for novice and experienced users alike, a simple upload interface will walk-through step-by-step the files needed for upload to a server and then embed code with a script instructing browsers how to access the files which make up the presentation will be given to users to place on a web page.

I believe that Popcorn.js can act as the puppet master script, but it’s only a hypothesis at the moment. With some support I hope to test my theory soon.

A problem publishers can’t ignore

Video is the future of the web, as Wired illustrated in the below graph, the popularity of video is growing and accounts for nearly half of all web traffic.

It’s only going to grow further as mobile video comes into its own and set top boxes like Apple TV, Google TV, Roku, Tivo and others start to embrace ipTV, and newsrooms should be preparing now by creating a future proof archive that can be easily scaled and adapted for any presentation platform existing or yet to come.

It’s in a publisher’s best interest to invest in a free and open web standard that will be supported for years to come rather than pouring resources into a proprietary system that may not play nice with future technology.

And with embedded metadata which contains copyright and licensing terms, there’s no danger of ever losing control over content distribution, use or sale.

Plus it’s not revolution, but evolution and non-disruptive to newsrooms with little need for training for a change of practice.

Photojournalists already use editing programs like Apple’s Aperture, Adobe Lightroom or Photo Mechanic as the start of their workflow. Here we select our best photos, rename and add metadata to organize, archive and make search-able for users our work for years to come.

VidScribe is a similar starting point for video, but takes it a step further as a web platform for deeper tools and plug-ins to improve productivity and viewership.

  • For producers VidScribe provides automatic transcription to facilitate getting into crafting a story faster, timeline mapping for supplemental content, embedding of written stories, branching curated content and a standardized metadata workflow to speed things up.
  • For publishers, VidScribe incorporates standardized by-line credits, copyrights, archiving and provides the ground work for monetizing video. Also, the process is familiar to newsrooms, thus lowering training costs to zero. Additionally viewers are welcome to add new functionally through plug-ins, providing publishers higher engagement and free upgrades based on user identified needs. And VidScribe is free.
  • For viewers VidScribe allows an avenue for seeking out videos via search engines, social sharing options with timeline bookmarking, closed captioning and commenting to timeline. And a mobile friendly way to consume all branches of a story through one interface quickly and easily.

Further monetizing video with VidScribe

  • Standardized newsroom workflow means quicker and easier production and ability to access for future use, such as in-house to curate, repackage, licence and rebroadcast on future platforms, allowing publishers to obtain more bang for their initial buck.
  • Search friendly video means archives have value for years instead of hours as new viewers come with every search. Also more video views and hits on articles is the first step toward charging ad premiums.
  • Open HTML5 web standards mean contextual advertising can easily be incorporated and changed locally or globally without having to re-export video or pay for an expensive third party tool like BrightCove.
  • VidScribe is free and open source, so it will expand with new tools in the future based on user feedback so you’re never outdated and you’re always allowed to invent your own new solutions.

Call to arms

Collaborators wanted. If you’d like to contribute, comment, collaborate or insult the open source project VidScribe, please comment on my blog below, drop me a line at shaminder.dulai <at> or reach out on twitter at @sdulai.

I’m a hack in search of hackers, but more so I’m looking for compatriots, people with ideas and a desire to see the web experience improve for everyone by getting information to eyes and minds in an easier fashion, on user’s terms so that journalists like myself can focus on telling stories that matter and have a real impact to improve all of society.

As I see it, we’re all working on slightly different angles of the same problem and all our tools can be packaged together to create one awesome toolset.