jump to navigation

Football GOOOOOOL: Describing Every Action or Each Event 2011/08/30

Posted by nydawg in Digital Archives, Information Technology (IT), Intellectual Property, Media.
Tags: , , , ,
trackback

When I was a kid, I used to play soccer and I loved playing and watching the game.  To me, the sport was all about teamwork, continuous movement and evolving strategies.  Now, years later, I still like the sport, but never play and only occasionally watch it on TV.   Looking at it from an archivists’ perspective, there’s a few points which merit mentioning.
The first point is the distinction between soccer and football.  In the US, we call it soccer, but almost everywhere else in the world knows it as football (not “American football”).   So this got me to thinking about the way a culture names a sport, location (“English” channel ) or even art movement (Dada?).  If digital humanities specialists were trying to research the development of the sport in this country and around the world, it might be a more difficult proposition because the terms used are different. . . . and may have changed over the years.

Another interesting difference between soccer and football on TV is the way it is described by the commentators.  It is so infuriating to me to watch it on American TV because the English-speaking commentators are usually so caught up telling the dramatic (or financial) backstory of the players, that they neglect to accurately describe the action on the field as it happens.  Although I’m not fluent in Spanish, I much prefer watching soccer in Spanish because they are telling a story in real-time. . . . and then when someone shoots and scores? GOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOLLLLLL!   The announcers are excited as the plot develops.  The point i’m trying to make is that although the “content” of the game (the action, the game, the video or moving images) is the same, but the Spanish-language commodity (w/ the play-by-play) makes it, in my opinion, a more valuable commodity because it is more descriptive (even though I don’t speak the language), especially for those who do not know (or care about) Manchester United or the football clubs in the UK (or elsewhere).   I’m not interested in learning that David Beckham used to play with so-and-so, but if someone describes the six passes leading to a shot on goal, please do tell!  Like other soccer fans,  I am interested in the process and how the game develops, or specifically, in knowing that the left wing crosses the ball to the forward who is guarded by the fullback (or whatever). . . .   If the people at FIFA or the World Cup wanted to build up an audience, I think they should really consider using Spanish-language commentators for the play-by-play, and then hiring translators to put it in English.  (Yes, I’m sure that could probably be automated.)  Let’s give soccer it’s due.

And this brings me to the point about the game, and why I loved it so much, how it’s different from baseball (or American football), and how it relates to archiving.   In baseball, tennis and American football, the whole game is set up as one event (pitch, serve or down) after another, and each event is described (and archived) as something with a beginning middle and end.  (“The pitcher winds up, releases and throws, called strike one.”)    In soccer though, because it is so dependent on the uninterrupted passage of time, the whole game is about process and the subtle changes that happen as players work together to accomplish a “goal”.   I find this fascinating, because as in archiving, parameters need to be set in order to determine best practices.  For example, professional games usually run 90 minutes, but only the referees keep accurate time, so even if the clock expires, the game continues until the referee says it is finished.   It’s not over-time, they keep track of how much “real” playing time is “owed” from injuries, fouls and other circumstances that arise in play.

In some ways, soccer is like basketball, but even that sport is set up and described as a series of events, and it is at such a fast pace, that play-by-play announcers don’t have to announce every pass, so they can usually focus on the end result “He shoots and scores!”  I guess what I’m trying to say is that baseball and those other sports can be described mechanistically, or the descriptions can be automated.

A few years ago, I worked at a video indexing retrieval software company called Virage (later bought by Autonomy recently bought by HP), and worked in a small department in New Jersey that ” captured” (encoded, digitized, etc.) and cataloged every Major League Baseball game from the 2001 season from Opening Pitch of the first game to the last out of the last game of the World Series.   Since I was a video technician, I was mostly responsible for making sure the video feeds (content) were captured, saved, renamed and uploaded so they would be accessible (and fully searchable) online within an hour after the game was concluded, but we also had stringers in the ballparks who were keeping score of the games, and electronically sending the data to our computers in New Jersey.  At the end of the game, the data and footage would be combined so a fan or any subscribers could search for any event (e.g. every double Derek Jeter hit in the month of May, [but not every pitch]), and create a highlights reel to be streamed or (maybe) saved locally.   MLB still does it, probably using the same software, but now they catalog every pitch and every commercial, and rather than simply rely on the software to combine the data, they hire seasonal workers to describe everything.  What’s interesting to me is that the original project was simplified and stripped down and provided an excellent way to standardize scoring and create a functional system.

Working with these brilliant software engineers (some now work for Google), I learned a lot about how structured data can be used to create a more valuable commodity.   If the software is created to match the scoring with the video feeds, then it becomes a mathematical exercise in which an event like a home run will take about 45 seconds, where as a strikeout (third strike) may take less than 1 second.  By using data in this way, Virage figured out a way to quickly and efficiently make every baseball game fully searchable with a very small group of people.   Today, it’s gotten a little more descriptive, but almost to the point of too descriptive, and not really functional anymore (in my opinion).  Who would ever search for ““ground out,” “from knees,” “last out,” “premier plays,” instead of “no-hitter”!
I suppose it’s a testament to the original idea that MLB is using the software on a larger scale with more (internal) stakeholders than simply fans creating on demand highlight reels, but I’m still left wondering why, and how would any logger want to tag everything:  “It is not only the game action that is tagged. If a squirrel runs onto the field, the play will be tagged with “animal.” If there is a shot of a man sipping a beer, there is a “drinking” option under the “fans” category.  I suppose if you’re not a professional archivist or cataloger, it might seem “cool” to be able to do that, but as an archivist, I’m left thinking, “Are some taggers wasting their time tagging 1000s of events in a 3 hour game?
But, I guess if there’s an app for that, there must be some function somewhere at some time.

And of course, there’s some other very cool things that MLB is able to do with their content in order to maximize subscriptions, but really, do you need to hire some person to point out that a squirrel ran on the field some day.

Advertisements

Comments»

No comments yet — be the first.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: