Hashtags Are Evil

Some Evils Are Necessary

Before Tweetworks, hashtags were the only way to categorize a tweet. And if I have anything to say about it (and I do), hashtags will be a thing of the past in the near future.

Hashtags are evil little things that suck up precious space in the already tight confines of a 140 character post window. But, since they’ve been the only mechanism a Twitter user had to add some measure of order to their tweeting chaos, people adopted them without a fight. The problem is, they don’t really work that good and they come with a heavy cost.

The Limited Utility of Hashtags

• Typos or Incorrect Tag – In October 13, 2008 there was a fun meme going around with the hashtag #Scariest3Words. A funny thing happened though as the day wore on, some people started using #3ScariestWords instead of the original hashtag. Of course this is a harmless example but imagine if this happened with topic you really cared about. Beyond the dislexia shown here, hashtags are wide open to typos since there is no validation process when they are entered. Enter #Scarcest3Words and there will be no warning message telling you that you’ve entered an invalid name.
• Duplicate Tags Pollution – There is nothing stopping someone for using a hashtag that you set up to track your conversation for something completely irrelevant. Worse yet, someone could seek to hijack your hashtag for their own shady purposes. Don’t laugh, the spammers and bots that are hitting Twitter are crafty. You don’t think someone might notice a big dog like Aaron Strout or Gary Vaynerchuk starting a hashtagged tweet stream and try to lace it with their “visit my site for cheap Viagra” messages? C’mon! But, if we assume that the most likely scenario is that some other group of people decide to use the same hashtag you are using, we can easily see that things can get confusing come query time.
• Lack of Depth – South By South West is a pretty big deal. Using the #SXSW09 (or is it #SXSW9? or just #SXSW?) tag only gets you so far in describing what you are talking about. In order to bring it down to something more specific you need to add another tag like #Music but then you might want to toss another one like #Rock and then #Band. Pretty soon you have chewed up half your 140 characters with hashtags.
• Who Started It? – I’m pretty sure that my friend Lyell Petersen started the #Scariest3Words thread but I can’t back that up with paperwork. And I am certainly not going to wade through the dozens of pages of Twitter search results to figure it out. You might not always care who started a conversation but you might care how or why the conversation got started. Every once in a while it is good to check back to the source to make sure the conversation is still on track. Once a hashtag is in the tweetstream however, God knows what will happen to it over time.

The Cost of Hashtags

• Time – It takes time to type a hashtag. It takes time to write a search query and sift through all of the responses and evaluate what you are seeing. While the hashtag may help to categorize a specific tweet as being about some topic in general it does little to put it in the context of a conversation as mentioned above. Therefore, a user has to invest quite a bit of time piecing together the individual components of a conversation.
• Missing Participants – Newbies and novice users have no idea what hashtags are in most instances. I didn’t know what they were when I first started using Twitter and I like to think of myself as reasonably plugged in. If you are a seasoned social media pro, be honest you’ve been asked “what’s a hashtag?” dozens of times. Think of how many amazing comments, answers and insights we are missing because not everyone is familiar with this rather “inside baseball” convention. Or perhaps, some people just can’t be bothered to use a hashtag since they themselves will never do a Twitter search.
• Processing Power and Bandwidth – Initiating a query from a third party application to run against the Twitter database uses some serious horsepower. There’s a reason why Twitter imposes API limits. Constantly hitting the API looking for tweets with a certain hashtag so you you can pull it onto your system and display it a certain way creates a lot of two way traffic and requires a lot of processing power on both sides. This isn’t counting the volume of storage space needed for all the extra characters unnessarily tweeted on a daily basis just for hashtags.

Tweetworks Groups and Threads Are Holy Water

The entire Tweetworks application is based on one simple concept, capture and associate a post to a specific conversation on the way in and present it as a comprehensible conversation to anyone who sees it in the future.

Groups – If we continue the South By South West example from above, the SXSW09 group on Tweetworks would replace the #SXSW09 hashtag. A user interested in tweeting about the March 2009 festival/conference would only need to locate the group, click in and start tweeting. No need to type a hashtag each time he wants to make a comment about SXSW09.

Threads – Adding a conversation about a live music event at SXSW09 is as simple as tweeting “I’m at Flamingo Cantina where Beach House is about to hit the stage.” within the SXSW09 group. Anyone who’s interested in tweeting it up about the show can jump right in by replying to the thread. Or if you don’t like the way the conversation is going, start a new thread.

With this system we avoid all of the pitfalls and costs of using hashtags:

• Typos and misnomers don’t happen when selecting a topic is point and click.
• The pollution effect is highly unlikely since each group not only has a unique name but in most cases the person who set it up also gave it a fitting description.
• It’s easy to see who established the group and who initiated a thread.
• There’s little time wasted in finding, participating and reviewing conversations that meet your interests.
• Any user, from the social media super star to the new guy, can log in and be active on day one.
• Tweetworks performs like a relational database driven application should. Each piece of information is stored and indexed appropriately. If the data went in organized it will come out organized. Hint hint.

Hashtags your days are numbered… You’ve been warned.