The Hashtag

The “hashtag” is a commonly spelled short word, often abbreviated and “smushed”, that is used to categorize online comments with a keyword and is preceded by the “#” symbol. They are given many names by different people: topics, tags, keywords, hashes, trends, channels, search terms, subjects, filters. They are born quickly, often burning out just a fast. At other times, they seem eternal. In some cases, their use is so ubiquitous for a specific topic that they will forever be associated to it. They can be created by a single person, edited and replaced by someone else, and officially sanctioned by a larger group consensus. They usually work well, but sometimes become co-opted, cross-pollinated, or ignored. Love them or hate them (or have no clue what I’m talking about?), in some form or another, they are here to stay. They have become cultural shorthand.


By far, the most prevalent use of the hashtag has been on Twitter, but they were never included as a core feature. Twitter didn’t invent or even lead the way with hashtags. They were the organic creation and adoption by users after years of tagging blog posts, indexing, and categorizing data elsewhere the web. They made Twitter better (though not everyone agrees). Some estimates suggest that 20-30% of all tweets now contain them. As Twitter puts it…

“It was created organically by Twitter users as a way to categorize messages.”

The acknowledged godfather of the hashtag is Chris Messina who on August 23, 2007 posted:

3 Rough Ideas:

Here are three roughly sketched ideas for the evolution of the hashtag.


Currently, most tweets are geolocatable by either knowing the author’s location or by inclusion of geodata itself. What I would like to see goes a bit beyond that. It would allow me to join a remote localized conversation, to add my thoughts to a spot on the map in a more efficient manner akin to the ability to throw my voice far and wide. This is a new method of creating a single word that contains more than the sum of its letters. In this case I see value in combining the keyword itself with baked-in geodata. I understand this sounds like magic in light of today’s social tools, and it might be — in my mind it’s a word that glows and crackles with letters that conceal a greater source of content than the alphabet allows. The only simplistic example I can think of could be called the “markup of the hashtag” — imbuing the word with extra data which adds to it and is machine readable. Maybe it’s not so magic after all?

Moving the hashtag from a variable to an object.


This centers around viewing a conversation and filtering without the false positives. Methods like these might use Bayesian interpretation to look at my location, my contacts, my previous tweets, the content and meta-data of the posts themselves and give me a better view of the topic. This could provide insights into a conversation using techniques and code that is transparent, yet configurable or removable by the end user. Of course, you’re saying, “Hey Steve — Lots of stuff out there exists to do just this. Have you seen” Yes — but I’m thinking about what comes next and can become an integral part of the platform itself, available as a layer to all clients, API access, and tools.

Intelligence baked in.


Platform-wide official adoption by Facebook, Google, and (insert name of hot new social network) of hash tags by these players would change everything. They might even provide a conduit through some of the “walled garden” data problems facing the internet. Some services have adopted the @ symbol to refer to names, using autocomplete to ease the pain of remembering and finding your subject. A new topic-focused method using this technique could be standardized and rolled into numerous social platforms.

Make it easy.

Last Thoughts:

In many ways this discussion of the tagging of content has been going on for years. The subject seems to have lost a bit of its luster, but it’s probably never been as important. And let’s face it, the phone number is past its prime. It’s time for a better use of the # symbol. It’s already on 7 billion old school phone keypads worldwide in the bottom right corner.

Post sources and further reading: