An American Optimist

This preview of the HBO documentary series 24/7 (see below or use this link) as it focuses on the NHL’s “Winter Classic” (i.e. a game played outdoors in a large stadium) captures nicely the importance of storyline as an essential component to creating drama around a sporting event. While you nor I probably cared much about the long-standing rivalry between the professional ice hockey teams from New York and Philadelphia… after watching this 12 minute preview, chances are pretty good most of us are eager to see how the story ends. After watching, we are now part of the collective experience surrounding this event. This is also one of the primary objectives of Podium…. storyline = drama = powerful collective experiences.

When thinking about structure, I keep coming back to the concept of an “open-data dashboard.” I think the reason is that Podium will most probably end up being a structural hybrid and dashboard is the easiest way to capture this idea. Structurally this dashboard will have different sections (I’m completely smitten by the “tile concept” now being championed by Windows Mobile). The sections will include: 

1. WIKI: A collaborative open-source wiki built on top of a semantic NoSQL database. This Wiki will hold factual and objective information about athletes, teams, venues, events, results, statistics, etc. Think of this as a “Wikipedia” focused just on the Olympic community. There is no point in going down the traditonal MySQL wiki route. Eventually, we’ll arrive at a semantic web and it just makes sense to begin that process for the Olympic community now. Collaboration is critical. This database will eventually become the external communal brain for the community. Everyone should be able easily add data.
2. READER: An open-source HTML5 reader (think: Flipboard) that can filter and display articles and blog posts in a graphically attractive interface. There are thousands of media companies covering the Games and Olympic sports… there should be a single place to aggregate this content. This section may eventually be the one that drives the monetization strategy the most. I have some specific ideas on how we can be a new type of “advertising” network that will sit inside this “Reader.” One of the main reasons I’m so excited about building this “open data dashboard” for Sochi 2014 is that the “vertical” of Olympic sports is already a multi-billion dollar a year industry with lots and lots of sponsors, advertisers, and corporations participating already involved.
3. STREAMING TICKER: A streaming data ticker (think: Stocktwits… Howard Lindzon is pioneering something really special in this niche) that search, filter, and display real-time data flows from the myriad of social networks (Twitter, Facebook, YouTube, Foursquare, etc.), not to mention, the streaming data from the Olympic Organizing Committee itself. This channel will be the primary way that on-site spectators receive information updates and announcements. Streaming data is becoming more and more central to the experience of sporting events, yet… the tools are still in their infancy. This section will be one of the most technologically challenging (try building a team of Hadoop wizards) but also one of the most critical to the success of Podium.
4. APPS: An 3rd party app store and hosting platform for the dozens of individual apps on-site and remote spectators will want to use while watching events.
5. OPEN DATA PLATFORM: A federated open data platform, built on top of NoSQL (Document Object Database… DODB), which enables various Olympic entities to upload and share/license data sets (complete) to other members of the Olympic community.
6. COMMUNICATIONS HUB: Inteeroprobility using OAuth for users/members to link the profiles with already established social network profiles on other platforms (Facebook, Twitter, LinkedIn, etc.). 

Note:  I have purposely LEFT OUT any live video streaming section. I view Podium as something that enhances one’s viewing experience, whether s/he be at the event in-person or watching remotely at home or in a bar. Podium will be everyone’s 2nd screen in their multiscreen viewing experience. We intend to drive more traffic to live broadcast viewshipship and increase ticket sales. At the end of the day… (as I’ve written before), sports is a social activity that depends on drama. The better information spectators have about the storyline, the richer the experience will be for them. Thus, in 10 years if we sit down in our stadium seats or in front of the TV without an addiitonal digital screen at our fingertips, we’ll feel naked. Podium will be central to the experience of Olympic sports.

So… what I’d like to do is to start with the Drupal 7 CMS as the base-layer but to use either MongoDB or CouchDB, instead of the standard installation of MySQL. There is, apparently, a plug-in for MongoDB. One of the issues is whether Drupal 7 is HTML5 compatible at this point. I am very eager to stay away from having to build this “dashboard” for each and every platform flavor in the game today. Thus, while there may be some initial short-term performance sacrifices, I’m leaning towards HTML5. Perhaps, the “responsive HTML5” Omega Drupal framework will solve this issue.

I’m not sure. Should I be looking at the custom Web App approach, instead of Drupal? The nice thing about Drupal is that it’s already an accepted open-source tool in the open data world. Why try to recreate the wheel? The other nice thing is that with Drupal I can use the open-source data portal CKAN software. There’s a number of cities/governments already using CKAN package for their open data initiatives. The only problem is that I suspect CKAN using MySQL and I’m doing my best to be “leading edge” with the data.

For the wiki I can use Mediawiki with the Semantic-MediaWiki extension, which is also open source but I suspect not “NoSQL.” I don’t know… it’s nuts. There’s all this pieces and I’ve got to try to find a way to get them all to be:

• Open-source technologies
• Semantic data (i.e. RDF)
• NoSQL (i.e.Document Object Database)
• HTML5 framework

You got any ideas? Please let me know… because as this project grows and starts engaging more and more people, we will soon be locked into our choices. Better to fiddle around now and make the right choices then to be sorry later. Drupal 7, HTML5, Hadoop, CouchDB, MongoDB, CKAN, RDF, MediaWiki, Socrata, and on and on. Lots of technologies to choose from. How would you build it structurally?

Ok… here’s the thing. My vision of where I see Podium going isn’t possible with today’s standard off the shelf tools. It's going to take some customization.

I like the aspect of wikis (like MediaWiki) whereby users can, themselves, add content. The problem is that it’s not "flat" text content we’re after but rather, it's a combination of data and structured content. When we look over at the standard open data platform being developed with Socrata or CKAN on top of Drupal 7, the issue we face is that the model is more of an upload and share. Implicit in this standard open data model is the idea that a data set is somewhat complete when it gets uploaded. That’s not how I see Podium.

Rather, I would like for Podium to be a place where people can both nominate the data sets, themselves, (i.e. female figure skaters) but also… contribute data on a piecemeal basis. In other words, we may know the names of the women figure skaters from the 1950;s, for example, but lack certain details on a majority of them. For instance, height and weight at the time of competition. Or even photos (headshots) of the competitors. I’d therefore, like Podium to empower individual users to add pieces of data. There may be a great grand-daughter of a French women’s figure skating Silver medalist who has scanned photos of her great grandmother. That user should be able to upload that piece of data into Podium. Think of it like a fill-in the blank data set. Over time… the Olympic community will make the data better and more complete.

But in addition to “data sets” we also have something that we’re tentatively now calling “content streams.” Content streams are a completely different type of information. Basically, this new data stream is unstructured. Think of all the data that flows through Twitter on a daily basis concerning the Olympic athletes. Just today, for example, there were thousands of “mentions” of athletes who compete in Winter Olympic sports. This ambient “chatter” is actually quite valuable data, albeit quite different than traditional data sets. Think of this ambient chatter as a sort of color commentary on these people and sports. We have “color commentators” in the broadcast booths of sporting events… thus, why shouldn’t we have ambient chatter flowing through various data streams.

In order to capture, store, organize, and share these new data streams we’re turning to new types of databases, known colloquially as “NoSQL” is a play on words with the traditional MySQL relational databases. Other people are calling these new unstructured databases “Document Object Databases,” which seems to me to be a better name for them. There’s a number of flavors of DODBs out there such as MongoDB and CouchDB.

What I’d like to see is a structured wiki that is built upon a DODB (NoSQL database). We also have even mentioned yet that 800 pound elephant in the corner of the computer lab… semantic data and RDF. Nonetheless, it’s pretty obvious that we’re all headed in that direction over the next couple of years. So, as you can see, the choices available on how to structure Podium are not exactly the right fit.

The more I think about Podium, the more I see it as a merging of data set, content wikis, and content streams. In other words, Podium is going to be a massive information portal. Now I know the word “portal” has been thoroughly denigrated over the past decade but there’s something right and true about the word that should make us give it another look. Podium will be like a giant community brain. It will be both a starting point and an end point depending on the type of user. Journalists will use Podium as a tool for research, spectators will use Podium as an entertainment medium, and developers will use Podium as data resource.

I can see Podium playing quite nicely with the myriad of new HTML5 readers (Flipboard, Pulse, et al) that are sure to evolve and become the dominant way people consume aggragted content. People will create customized RSS/JSON feeds (that won’t need to know the technical details) which can then be fed into their readers or other apps and services.

The point is that if you stop and think of the amount of data and content associated directly with the Olympic Games and their constituent sports (50+ sports), it’s staggering. Actually, it’s so large and so disparate that no one company could ever hope to organize it. Fortunately, we have seen what the community has done with Wikipedia, Ushahidi, or Open Streets Maps. Podium will stand upon the shoulders of these early crowd sourcing projects.

Therfore, Podium must be structured as an open source project. I will begin reaching out to various media companies, technology companies, and most importantly… the IOC, itself. In five years, there will be dozens of other open source hybrid data/content portals in lots of other verticals. I just happen to really like sports (rather than… art, medicine, automobiles, hunting, etc.) and the Sochi 2014 Winter Games are the perfect venue to launch Podium to the world.

I’m doing some research today for my new Podium.io (the top-level domain is .io) project and I wanted to capture some ideas and thoughts here (very rough). I will use this blog as a notebook and scratchpad for my thoughts. But first, let me give you a brief introduction of Podium so we’re all on the same page… more or less (feel free to give input). Podium.io aims to eventually become the first open data platform (think: “open data.gov” for an analog) for the global Olympic sports community… both Winter (7 sports) and Summer (26 sports). The project will launch its beta in February of this year and will intitially focus on building valuable open data sets for the Sochi 2014 Winter Games. 

Podium is the final iteration of what started as an idea for a digital travel website that then morphed into a mobile application, and finally grew into a what I was calling a HTML5 digital dashboard. The basic premise of Podium was that there was (and still is) a lack of high quality relavent information focused on the 2014 Winter Games in Sochi, Russia. As an American expat who actually lives here in the same region where the Games will be held, the lack of information was glaringly obvious and annoying. So I took it upon myself to fill this information void. Unfortuantely, as I began to look into building a website and/or a mobile application (iOS, webOS, Android, Mobile Windows) to address this information void it quickly became apparent that the Olympics are a data beast like no other. I started to get nervous ; ))

The Olympic Games are not simply the largest events in the world (sporting or otherwise). In fact, the Olympic Games is a *massive* global community of more than 15,000 athletes, close to 200 National Olympic Committees, dozens of International Sports Federations, hundreds of actual Olympic Teams, tens of thousands of staff and officials, hundreds of thousands of live spectators, and millions of fans following the action from around the world. The Olympic Games are basically an industry unto themself. As such, it quickly became clear to me that no single start-up could EVER hope to capture all that data and information, by itself. It’s not just data sets but also, all the data streams as well (think: Twitter, Facebook, blogs, etc.). The Olympic community is so massive that it’s now fractured into hundreds of data silos that never “talk” to one another. 

Once I digested the enormity of the data challenge I was then confronted with the reality that there would hundreds of media outlets around the world, which would be building their own individual mobile apps. Media companies like NBC, Eurosport, The New York Times, RIA Novosti, ANO Panorama, The Guardian, not to mention, application developers like Foursquare, Uber, Zagats, Facebook, etc. would all be launching apps for the 2014 Olympic Games. I suspect London 2012 will be remembered as the last “broadcast-centric” Olympic Games. For me to try to compete with all of these corporate giants would be a fools errand. So… now I was not only nervous but also, stuck!!

I was just about to throw in the towell when I had a key insight that changed everything. The insight is that all these companies would need two things in order to be successful. First, the average spectator wouldn’t want to download 25 separate applications for a single event, even one as large as the Olympic Games. There would need to be a single platform upon which all these various apps could be discovered and hosted. Think of it as an “app store” for the Olympic community.

Secondly, in order to build these digital applications and websites, all of those global media companies would need fresh, accurate, and plentiful data. Data is the lifeblood of sports. Without data… one can’t create the storyline needed to build the tension and thus, the requisite drama.

Sports are, at their most basic level, a way for groups of us to share in a collective experience. It’s these collective experiences that we all seek out and which, really give true meaning to our lives. We are most alive and happy when we are part of a larger community that has its own common purpose. Whether that purpose is fighting a war or cheering on our home team is immaterial. What is critical is that there be drama (i.e. an unknown ending) and an opportunity to self-identify with others from our “tribe” (jerseys, face-paint, team colors). That’s why sports are such an important part of our lives. Sports are not, as commonly assumed, simply brief sojourns from the reality of everyday life but instead, sports are reality, itself. The magic of the Olympics is that they take this drama to a level not normally achievable with local city-based sports.

As a technology geek it occured to me that I *should be* focused on building the open data platform upon which all these global media companies would build their respective applications. This insight changed Podium from being a product to being a service. There are dozens of examples of open data platforms built by governments around the world… so I asked myself, why not do the same for the Olympic community. Actually, the IOC is like a government for the world of sports. Like the open data platforms in government, Podium.io will be built by the community, itself. Instead of agencies and ministries, Podium will recieve data from media companies, NOCs, IFs, Olympics teams, and if we are really successful, from the sporting public, itself. 

Tomorrow I’ll start to examine different ways we might structure this open data platform. Do we want to use MySQL or NoSQL? Can we find a database “wiki” that uses CouchDB? Do we want to use tools like Mashery or Socrata? I don’t have the answers yet. We are entering uncharted territory.

What I do know is that we need to build a platform that is open and transparent, so that the majority of the Olympic community feels inclined to contribute (crowd source) in helping to build something that is bigger and better than what any single company could build. We pretty much know that the old broadcast model has reached its logical limit. Podium.io seeks to push the technical boundaries of how we can blend the roles of participants and spectators in Olympic sports. The journey has just begun. We’ve got two years to build this baby. We will do it together!