The Bivings Report - powered by the Bivings Group

Back in January 2009 I started the Master of Information Management program at the University of Maryland, and last semester I took a class on Consumer Health Informatics and another one on Social Computing.  With the permission of both professors, I wrote different versions of a paper for each class about how Wikipedia has maintained accurate health information, and a version of this paper is now available here. The Accuracy of Health Information on Wikipedia (click to access the report) was an interesting paper to write, and I hope that you find it just as interesting to read.

Wikipedia is different from other well known websites with health information in that it allows anyone – subject experts and laypeople – to contribute and edit information that it presents. Given that anyone with access to the site can edit it, scholarly studies have found to the surprise of Wikipedia’s critics that scientific information on the site is as nearly as accurate as that found in Encyclopedia Britannica (read the December 2005 Nature article “Internet encyclopaedias go head to head” that helped publicize this debate). This also goes against the reasoning of established measures of health information accuracy like the standards of the HONcode. Instead of having strict editorial procedures involving subject experts, Wikipedia’s culture fosters and regulates a sizable community of ardent editors who are devoted to protecting the information integrity of the site, and scholars speculate that their motivations are not altruistic. Through the diligence of active Wikipedia users and the establishment of standards, procedures, boundaries of responsibilities, and sanctions, the site has articles with accurate information. To discuss Wikipedia’s accuracy, this paper reviews the scholarly literature about Wikipedia’s accuracy in relation to health information. This discussion will focus on the the following topics: Wikipedia’s contribution to consumer health information sites on the Internet, metrics used to measure the quality of health information on the Internet, its level of accuracy, what aspects of its community culture enable it to maintain remarkably high accuracy levels, and the characteristics and aims of other collaborative websites like Wikipedia.

Interesting TED talk from Reddit founder Alexis Ohanian explaining how Greenpeace’s Mister Splashy Pants became an online phenomenon.

Congressmen Tim Ryan (D-OH) launched his own custom social network yesterday, using the Ning platform.  In and of itself, this isn’t particularly notable.  Lots of politicians have launched their own Ning networks.  However, in every case I know of the focus of the socnet has been on winning elections, and the networks have been launched under the banner of a campaign web program. 

Ryan has done something entirely different here.  His network was built as an extension of his official House web presence, and is a tool for him to communicate openly with his constituents.  Please correct me if I’m mistaken, but I believe Ryan is the first House or Senate member to launch their own social network for the purpose of connecting with their constituents.

Rules for web communications in the House and Senate are notoriously archaic, so the fact that Ryan was able to launch this strikes me as a significant step.  It will be interesting to see how the network does, and if others follow Ryan’s lead. 

As most probably know, Twitter is in the process of launching a version of the widely used retweet on its own platform.  The move has caused some controversy, as the way retweets has been implemented by Twitter is much different from the unofficial protocols that Twitter users developed organically on their own.  Twitter founder Evan Williams explains Twitter’s reasoning here.

I’ve been testing out the new retweet functionality for a few days, and I must say I am not a fan at all.  Sean Bonner has written a great blog post dissecting what he dislikes about Twitter retweets.   In his post, he hits on my two biggest issues.

1. When a retweet appears in your Twitter stream, it shows the avatar of the person who wrote the original tweet instead of the retweeter.  So you have random people’s avatars showing up in your stream.
2. Users can no longer add their own comments to the retweets.

The result of these two protocol changes is a complete lack of context for the retweets.  Bonner summarizes the lost context problem quite well in this paragraph of his post, which focuses on the avatar issue:

Seeing icons and usernames in my stream of people I don’t follow, even with the addition of a little “retweet” icon does not create a richer, fuller experience for me. It instantly makes me assume Twitter is broken and somehow people I don’t follow are showing up in my stream. It’s jarring and uncomfortable. Ev suggests there is no value in having the icon of the person you follow in a retweet but I completely disagree. Seeing the icon of someone I follow, someone I’m familiar with, instantly puts the retweet in context. Is the person regularly sarcastic which might imply the retweet is a joke, is the retweet a link to an article covering a topic this person usually tweets about which would give me an idea of the slant of the article, is the retweet from someone I follow because I respect and trust their opinion or is it a retweet from someone I’m friends with but don’t always agree with or from someone I follow because they constantly opposing my viewpoints and I want to hear their side of the story as well. Seeing the icon of the person I follow tells me a lot about the tweet and why they likely felt the need to retweet it before I ever read it. Seeing the icon of someone I don’t follow, don’t know, and have no context for confuses me.

Not being able to add your own note to the retweet further destroys the context.

Anyway, not a fan so far.  What do you think?

Update: Techcrunch has a very thorough and thoughtful article on this issue.

In the past we have profiled the innovative work of people like Adrian Holovaty who does some exciting things through programming that allow data to tell their own stories.  For instance, he set up the Campaign Tracker for the Washington Post, and he is now running his own site Everyblock – the database driven hyperlocal news site.  Further, in our newspaper studies we have examined how news organizations use the Internet.

While at the CIO Perspectives Forum here in Washington, DC last week, I was wondering about what kind of role CIOs should play at news organizations.  Their organizations may already use them as I am thinking (at least I hope).  However, it is important to note that in an October 2008 white paper titled “The CIO Profession: Leaders of change, drivers of innovation” IBM reports that most organizations mainly view and use CIOs as implementers and not as strategists.  Hopefully, news organizations turn to CIOs for planning and strategy and not just technical execution.  Of course, a CIO with a journalism background could help even more; Adrian Holovaty is so innovatively partly since he has a background in both programming and journalism.

If CIOs are involved in strategic planning, they can help the business and editorial staffers not only understand what is realistic, but they can also introduce them to new ideas that someone without technical expertise would know.  Further, a CIO could also identify potential problems and other issues in advance and help either avoid them or prepare other organizational stakeholders in advance. 

Here are a few examples of how a CIO can help news organizations strategically plan: 

• Broadcast journalists are now going out to the field with recording and editing equipment with fewer producers and camera people accompanying them.  Before such equipment and software is purchased, the CIO could help determine the requirements for such equipment and software and then help identify applicable products.
• A CIO with a decent understanding of database management could help find new application ideas like the ones that Adrian Holovaty has developed.
• During our newspaper studies we look at the website features that news organizations use by examining the different features (i.e. blogs, RSS feeds, video sections, etc.) they harness.  A CIO would have a valuable perspective on helping developing a strategic plan on what the organization’s website should do. Further, he or she she can help evaluate content management systems (CMS), applications, and vendors to determine how well they meet the organization’s needs and desires.
• A CIO can help an organization envision and deploy an API – like NPR’s API — which allows members of the public to use its data in interesting ways that the organizations itself did not image.  This is one way that a CIO can work with the business and editorial people at a news organization to help them find ways to profitably exploit an API.
• The CIO could help find ways to organize content for the public in many different ways.  For instance, tagging enables people to find information in ways other than the traditional divisions like News, Sports, International, etc.  In fact, The Guardian newspaper wanted a tag editor back in October 2007.  This would help make the news organization’s site more easy for the public to use.

These are just a few examples of aspects of journalism in which news organizations can turn to their CIOs for strategic planning, and considering the dire state of media companies, they can use all of the help that they can get.

How else can news organizations harness their CIOs to help strategically plan?

As social networking sites and technologies have flourished over the last few years, there has been much discussion about privacy today.  It is not that uncommon for people to provide updates about their personal lives on their Facebook accounts or Twitter feeds.  They talk about if they are sick, have a crush on somebody, are out partying, etc.

While this information sharing is innocuous at times while concerning at others, there are some social networks that are pushing the limits.  For instance, while at the O’Reilly Gov 2.0 Summit last week, I learned about the site PatientsLikeMe.  This is a fascinating site in which people with medical conditions come to connect with other people suffering from the same condition.  However, there is so much more than providing moral support and answering questions, people are expected to build detailed profiles about their bodies and health histories.  When they undergo treatment, they are encourage to share their experiences to it.  Does it give them gas?  Do they get headaches?  Is their sex life affected?  Of course, does the treatment actually work?

Now, it is one thing for me to announce on Facebook that I’m going to have Papa John’s Pizza for dinner tonight, but it is another to share personal side effects of a medication that I’m currently taking.  PatientsLikeMe does acknowledge privacy, but expounds upon the importance of openness on the site by stating: “You see, we believe sharing your healthcare experiences and outcomes is good…for a greater purpose: speeding up the pace of research and fixing a broken healthcare system.”  By sharing detailed health information about yourself, you help others understand how medical conditions and the procedures used to treat them work.

So, do you think that openly sharing your health information on the Internet is worth the potential it can to help others with their health?

I attended the O’Reilly Gov 2.0 Summit today and enjoyed listening to many of the insightful speakers.  One of them was Clay Shirky; who is popular on the tech conference circuit.

He talked about how organizations have tried to harness their community members to use the data and services that they provide turning them to them to the public to develop applications and new uses for their services.  Shirky cited Apps for Democracy as a success story in this area; over four dozen applications have been developed to harness government data.  On the other hand, the Los Angeles Times tried to invite its readers to help write editorials through a wiki; this was a failure since the space was used to post information that was either not useful or vulgar (ie links to porn).

Shirky argues that when organizations invite others to participate in their work, they need to have a “social contract” that is complete enough to provide a purpose for the applications and new uses but not complete enough that it stymies participants from coming up with new ideas that the organization that did not think of.  Further, by limiting creativity, this chips away from the participants’ motivation to work.  According to Shirky, the Apps for Democracy contract was sufficient, and the LA Times contract was too complete.  The newspaper already had an idea of what its users would produce, and it was embarrassed that the experiment didn’t turn out well after advertising it.

Shirky said that organizations that undertake such initiatives need to keep three things in mind:

1. Expect surprises; allow and trust people to be creative
2. Don’t take credit or advertise the initiative until the results are understood
3. Allow people to use their own motivations

On the heels of an overhaul of their print product back in October, the Los Angeles Times launched a redesign of their website a week or so ago.  I like the new www.latimes.com a great deal from a usability standpoint.  The new sites prominent, simplified horizontal nav bar is a huge improvement over the more complicated content structure present on the old site.  And while I’m sure some will find the black and white design unexciting, I think the palette evokes the print product and makes it easy for the eye to focus on the content.  This is a nice improvement.

Check out the screenshots below to see how the site has changed.  Click on the images for full before and after shots.