I was recently inspired after reading a tweet posted by Hal Grieb, who shared a link to an article on how to communicate a snow event. To ensure the tweet was seen by all potentially interested parties, he included 12 distinct hashtags (#smem, #sm4r, #sm4em, #hsem, #gov20, #egov, #gov2, #crisis, #disaster, #pio, #nws, and then #fatigue). Joking about the number of hashtags he had to include, he then went on to tweet about how the volume of related hashtags now required to identify a tweet might lead to “cylinders of excellence,” (should someone forget to include all of the appropriate tags, thus sending the tweet into unidentified Twitter space), and that a Twitter hashtag user manual or standard operating procedure (SOP) might now be needed to explain how to correctly use hashtags, which ones to choose, and when to include certain ones.

I started thinking – how does someone keep up with so many hashtags, especially as so many are created on the fly? Furthermore, how do we ensure information is discoverable from those who know nothing about hashtags and do not include them in their tweets? Public safety agencies have been dealing with this dilemma for some time – no matter how much work is done prior to a disaster to educate the public about hashtags to use for a certain type of event that might identify their tweets to the corresponding agencies, hashtags always take on a mind of their own – misspellings, random words and phrases, and so many related words leave the “Twitterverse” in a constant state of confusion. Furthermore, with characters and space being so precious within a 140 character limit, hashtags are starting to creep into the messaging of a tweet and making it harder to decipher the meaning of a tweet filled with random tags.

So, how does one make sense of the ever-growing universe of possible hashtags? How does an agency choose which hashtags to monitor for information about their jurisdiction? How does an agency work with its audience to ensure they use the right tags during an event? The uber hashtag – one tag that equals several tags, (related tags to be set up prior to an event by an agency depending on their demographic, hazards, locality, etc.), could be entered into some sort of widget to scan Twitter for one word, and ultimately, all related words. For example, the uber hashtag could be #smem, and the user could then set parameters like #smem=#em, #hsem, #disaster, #crisisdata, #crisis, #disaster, etc… If the uber hashtag is set up prior to an event, an agency can publicize its use within its community – the agency can then add related words or spellings to ensure they are capturing a larger percentage of tweets. During an event, the agency can then add additional tags as they appear, and potentially, the widget could scan for related tags and ask the user if the tag does in fact equal the uber hashtag, for inclusion in future analysis.

The implications of the uber hashtag are huge. First, it helps to solve the problem of shrinking space within a 140 character tweet. Second, it helps to solve the problem of confusion and inability to predict hashtags used during an event. Third, it enables public safety agencies to begin pulling information from Twitter in a standardized way – agencies could share uber hashtags across jurisdictions, but change the related hashtags to account for differences in locality. The uber hashtag serves as a sort of metadata for an ever increasing universe of random tags, and begins to develop a much needed taxonomy of sorts to help clear through the confusion.

To make the uber hashtag a reality, I believe the following must occur:

– Development of a governance framework between entities to assist in standardization of uber hashtags and related hashtags for specific use (e.g., specific disasters, events, use cases, etc.);

– Development of requirements for a widget to enable agencies to set their own parameters when creating or utilizing an uber hashtag;

– Development of data standards, data/relationship model, or taxonomy (e.g., x=y, and identifying potential related hashtags to include within the uber hashtag net); and

– Research into the inclusion and integration of geospatial data so that the uber hashtags can be used for visualization of aggregated data for better decision-making.

Please feel free to comment – what are your thoughts on the concept of a uber hashtag? What is needed for successful deployment? What types of concerns may arise?