@RobinSloan posted an article on Twitter Media last week that piqued my curiosity, The Art of the Hashtag. In it, he described how media outlets such as MTV are carefully crafting Twitter hashtags and promoting them on TV to encourage participation and achieve trending status. I thought this was interesting for two reasons. Firstly, this use of hashtags goes beyond the original purpose of categorizing Tweets, and secondly, because the author seemed to be encouraging the use of hashtags in this way, by explaining how to create hashtags that “perform well.” I started thinking about both the origin of the hashtag as a “poor man’s metadata” and the many ways in which it is used, wondering what implications this new type of usage might have for search and semantic processing of real-time data.
Twitter’s support page for hashtags says that hashtags are a community creation and that there are no formal rules for using them. It’s not surprising then that the meaning of hashtags is often quite obscure. Sites like twubs and hashtags.org attempt to provide some explanations of hashtags and why they are trending, but these sites are not always helpful for understanding some of the more obscure examples.
I haven’t made any formal attempt to categorize all the ways in which hashtags are used, but I’ve observed the following examples:
1) keyword metadata: hashtags used as keywords to categorize tweets and make it easier to find or group them via search:
2) Call to action: hashtags designed to encourage participation and become trending topics, such as those described in The Art of the Hashtag:
3) Aside: hashtags used as an aside to express extra information such as emotion or feeling about the tweeted text:
AdvertisingAge also ran an article about hashtags as entertainment this week but didn’t seem to make the connection that many of these tags are created by the entertainment industry specifically to encourage people to participate. They’re not just obscure but clever tweets randomly propelled to trending status.
So while the Twitter community is spontaneously adding a variety of information that is enormously helpful for search and metadata analysis, this positive behavior is diluted by the entertainment industry’s gaming of the hashtag to create trends of meaningless tweets. The latter is admittedly an interesting phenomenon in and of itself, but because these hashtags are promoted on separate media, mostly TV, in many cases it may be virtually impossible to identify them as such after the fact, unless of course people are already keeping track of them.
So what are the implications of evolving hashtag usage for inferring trends in news, culture and politics via analysis of Twitter data? Are studies such the visualization of U.S. Mood Throughout the Day as inferred from Twitter affected by the media industry’s manipulation of behavior via hashtags? I’m not aware of any articles that address this phenomenon and it’s implications. If you are, or if you have additional examples of hashtag use, let me know in the comments.