Monday, 14 March 2011

Session 5: Social knowledge production and services - Getting news from Japan

(Apologies for the lateness of this post. I've been frantically trying to get in touch with people I know in Japan and have not been able to focus. This will not happen again.) 

News 2.0 
By now, we have all heard about the earthquake and subsequent tsunami that devastated the northeastern side of Japan last Thursday. Though the tsunami warning for Hawaii had me on edge last week, I was far more worried about friends and former colleagues and their safety, since the epicenter was very near the city I once called home.

While having dinner with a few friends, an ominous silence suddenly overtook the restaurant. All eyes were glued to the grim news on the televisions around the room: One of the largest earthquakes in the history of Japan had just occurred. Deciding that it would be too dangerous to stay out that night and beside myself with worry over the state of my friends in Japan, I quickly excused myself. CNN provided the most coverage, but much of the information was second-hand from NHK and prone to a lot of errors. Thanks to friends in Osaka that were online at the time, I was able to get a link to a live stream of NHK reports via Ustream. Many other friends either Tweeted links to Ustream or posted it in their Facebook statuses. 

Ustream is interesting because you can see streaming traditional television news stations contrasted with live updates provided by users logged into the "Social Stream" on the right-hand side of the viewer window. 

NHK Feed via Ustream
with commentary on the right

Ustream allows users to stream their own content across the internet. These users range from a person broadcasting their baby's first birthday, to major news organizations like NHK and CNN. On the side of video feeds available on Ustream, one is able to log in to a running commentary sidebar that functions much like a regular chat room or a Twitter feed. Users may log in through a variety of social networks and can interact with other users watching the Ustream feed. 
Much like Amazon or Netflix, Ustream
appears to use collaborative filtering to suggest feeds
Ustream features a page devoted to news that not only streams the most popular news feed at the moment, but it suggests feeds also being viewed by a large number of viewers. As you can see from the screenshot above, the most popular feeds are (understandably) those with news about the situation in Japan. The article by Lerman, with its mention of "collaborative filtering," came to mind when I saw this page. Many different friends were slowly relaying the link to Ustream over their Twitter and Facebook accounts, filling up my news feeds. I too reposted the link which other friends also reposted from my Facebook wall to their own. 

Unsourced reports from viewers
contrasts with live reports in the window to the left
It was strange at times to see the news feed playing as people were making comments. Sometimes users would post information that was contradictory to what was being said on NHK, and this would set off a wave of speculation that rippled through the conversation for a while. Unscientific observations of the feeds at the time showed that while trollish behavior was still present, a surprising majority of those interacting with others on the social section of the news feed were not participating in anti-social behavior. Much of what I saw were messages of support for Japan, suggestions for where people could go for help, or information being given out in response to questions. 



A combination of news/information sharing,
reactions, and calls to action
Some of the information was correct, and links to donation sites actually led to the appropriate place (mostly the Red Cross), but there were also incidents where well-intentioned people were giving out misinformation. There were people that had heard from the American branch of CNN that a nuclear reactor had already exploded, while live footage from Japan showed that the reactor was still intact. Users that could not speak Japanese were then confused, and there was a small panic in the comment, which was eventually quelled when someone translated the live feed and said that the reactor did not yet explode. CNN later changed the story as they were updated, so accurate information eventually circulated in the social section of that particular Ustream feed. 

What Liebenluft said in his article, "...the problem isn't just that Yahoo!'s site helps ninth-graders cheat on their homework. It's that a lot of the time, it doesn't help them cheat all that well," reminded me of the little exchanges I saw on the Ustream social networking feed. For those users that didn't stick around after reading the misinformation, it is a lot like they were trying to cheat on their homework but getting answers that would make the cheating all but meaningless, since they were wrong anyway. 

Regardless of the incorrect information being spread, it was always relatively quickly corrected. Why? I believe it was because of the presence of the live news stream alongside the social comment section. Incorrect information by users was refuted by the live news feeds and in general, attempts to propagate the misinformation were stamped out by a majority of users after the video feed contradicted it. 

The positive aspects of the social comment section included accurate links to places where people could send donations to Japan, first-hand reports from people near the areas affected (mostly in Japanese with some interaction from English speakers living in Japan), links to pictures from places being discussed in the news feed, and general information provided with links to things like the schedule for rolling blackouts and modified train service. Also quite valuable for users that don't speak Japanese were the live translations/summaries of what was being said on the news feed by users in the social comment section. 

I suppose the point I'm trying to make is that a more 'traditional format' like regular network news, can be enhanced by the presence of the online community with added-value like the translations and links mentioned above. The live video feeds also provided a good grounding source to prevent the misinformation that can often run rampant through internet discussions. I'm quite impressed with how the system has been working and will keep an eye on it in the future. 

Finding out through Facebook: Status updates from those affected by the quake

Not knowing a lot about the situation at the time and impatient to hear if my friends were OK, I tried calling their phones; both cellular and home lines. No one answered. I had friends in unaffected parts of Japan try and call as well. Once again, nobody answered the phone. I watched the news, as many did, and saw the City of Kesennuma, the city I once worked in, ablaze.

My stomach dropped and I scoured the internet for any and all scraps of information I could find about the situation in the city. I have yet to find out about many of my Japanese friends, as only some have been able to Email me regarding their safety, but I am keeping an eye out on the missing persons database Google has created in response to this situation. So far I know one of my friends is still missing, thanks to this database and am still hoping to hear he is safe. 

Online communities have also impressed me with how information--in this case the well-being of my friends--can be shared with a speed that would have been difficult before sites like Facebook. Here are a few screenshots:

An update from the sister of my friend
on his Facebook wall
A clip from the Facebook wall of my friend Rachel

Through this posting I was made aware that JET program officials had posted lists of their teachers that had reported they were safe and was able to find a few more names of friends in other parts of north eastern Japan. Beside updates on the statuses of my friends, I saw on Facebook an outpouring of support and links to donation sites for the disaster. 

Social filtering at work: A large majority of friends
in Japan posted earthquake relief and links to information
during roughly the same time. 
A large proportion of my friends on Facebook either work in, or are from Japan. Because of this, the status updates of friends that posted things unrelated to Japan seems to have been ignored by the status update algorithm Facebook uses in favor of the updates posted by my friends in Japan. I thought this was a relevant, personal example of social filtering at work.

Throughout this tragedy, I have never been more aware of the role social networks have come to play in how I acquire my news. Because of Facebook and Twitter updates I found out that some of my friends were safe, that there was a link to Ustream for me to get updated information from, and that there were places I could go to donate money to the earthquake relief effort. I can't state how grateful I am for these resources and will simply close with a plea for everyone to please remember the people of Japan during this difficult time. 

Japan Relief Links:




Sources:

Lerman, Kristina (2007). Social Networks and Social Information Filtering on Digg. Proceedings of Int. Conf. on Weblogs and Social Media, Boulder, CO. http://arxiv.org/pdf/cs/0612046v1

Leibenluft, Jacob (2007).  A Librarian's Worst Nightmare: Yahoo! Answers, where 120 million users can be wrong. Slate, 7 December 2007.  http://www.slate.com/id/2179393/fr/rss/ 

Saturday, 12 March 2011

Interesting to Me

I have friends in Kesennuma City in Miyagi that frankly, I would not have known were safe without Facebook. The Huffingtonpost put up an interesting article about social networking and the response to the disaster in Japan: http://www.huffingtonpost.com/2011/03/11/twitter-facebook-become-v_n_834767.html?ir=Technology