Subscribe / Unsubscribe Enewsletters | Login | Register

Pencil Banner

Twitter needs to grow up and take responsibility, some say

Brandon Butler | May 6, 2013
Twitter, the increasingly popular microblogging service, has come under recent criticism for allowing easy spread of false information.

Google already has an optional two-factor system, but Behrens says there's a careful line. Usability is the biggest question, Behrens said; Twitter still wants to keep it easy for Tweeters to use especially non-technical savvy ones, which is why he believes an opt-in approach would likely be best. Behrens wonders if Facebook and LinkedIn follow in Twitter's and Google's footsteps in exploring two-factor authentication?

Alternative suggestions

Implementing two-factor authentication will not solve all of the problems that have cropped up around Twitter in recent weeks, though. It may help prevent the AP's account from being hacked, but it would do nothing to prevent false rumors from spreading like wildfire.

Wired reporter Mat Honan offers a solution in arecent article, noting that he regrettably tweeted incorrect information linking Brown University missing student Sunil Tripathi to the Boston Marathon bombings. When Honan tweeted it out, some number of people retweeted it, sharing it with their followers; some number of their followers may have retweeted it as well. Even if Honan had issued some corrective alert to his followers that the tweet was incorrect, it would not guarantee that everyone else down that chain of retweets would see it as well.

Honan, therefore, proposes a way to mark tweets as having knowingly false information, or showing them having been edited. Twitter does allow the ability to delete Tweets, which also deletes the post from any users who have retweeted it, but there is no post-facto editing of Tweets currently. Others have proposed some sort of upvoting and downvoting system, a common feature of many social media sharing sites like Reddit.

Paul Gillen, a social media expert in the Boston area, says all of these issues amount to growing pains for Twitter as a platform. Twitter is being used in ways that its creator Jack Dorsey likely did not originally envision when he created it in 2006.

Gillen is optimistic that between steps Twitter will take, such as implementing two-factor authentication, and by the general Twitter user base learning who to trust and who not to, that the platform will improve. He cautions against throwing out crowd-sourcing platforms altogether, though. Wikipedia, for example, after last month's bombings very quickly compiled a well-sourced posting on the incident. Don't dismiss tools because of some bad experiences, he says.  

 

Previous Page  1  2 

Sign up for MIS Asia eNewsletters.