It’s become routine for me to “tune in” to REBarcamps using Tweetgrid.com/irc. I set up the feed in the morning and let it run all day. I’ll check in once in a while during the day to see if anything of interest catches my eye, but what is of greater interest to me is who the major communicators and what the big topics of conversation were. I’m trying to look at signal vs noise.
I put together some Wordle-based illustrations after REBarcamp Charlotte and they were well received, but I decided to do something a bit different with the data collected for REBarcamp St. Louis. The video below was triggered by two things. First, I lost most of the morning data by accidentally closng the Tweetgrid.com/irc window, so much of the data was lost for the day. Second, so few people were sharing information and so few tweets were created that it made any valuable analysis impossible. The truth, we need better listening tools.
The Prezi used for this video can be found here: #rebcstl prezi
What this app does is generate a “trending topics” list from JUST the people you follow on Twitter. You can sort for words or hashtags by timeframe. If you find a tweet you want to respond to or retweet, you can do that right from that app as well. This is all Friend Trend does. That’s it. And I like it.
But all night long that headline haunted me. I tossed and turned most of the night and went in and out of coherent thought. And I woke up with a moment of clarity. I rolled over and hugged my wife and said, “We need to change Mothers Fighting For Others.” She bit. “How?”
So, although we were running a 100% Macintosh office and had no worries at all about our systems being impacted by Y2K, the newspapers were very concerned about the power company and required that we rent a LARGE generator to kick in when the lights went out at 12:01 am on January 1, 2000. Of course, that never happened. And, truth be told, none of us at AdOut ever thought it would.
Some very well meaning, very smart people are being distracted by the lure of shiny objects. And they are distracting others in the process.
![Reblog this post [with Zemanta]](http://img.zemanta.com/reblog_e.png?x-id=c6b76e88-34a7-49b7-a333-0081bf3c701e)
![Reblog this post [with Zemanta]](http://img.zemanta.com/reblog_e.png?x-id=15bf8a7b-3e44-4cc1-8b0e-80eeb0140ea6)
![Reblog this post [with Zemanta]](http://img.zemanta.com/reblog_e.png?x-id=106268e5-6d41-45f9-9d8a-c3e213bd6e70)
.. 
.. 
.. 
.. 
.. 
Signal VS Noise: A Look At The REBCNASH Twitter Stream
I wasn’t going to write this post. I obviously changed my mind.
This is a post about REBarcamp Nashville, but it’s not really about REBarcamp Nashville. Nothing I’m about to say has anything to do with the quality of the 51 sessions that took place during the day at REBCNASH. From everything that I’ve heard from people who were actually in attendance, Brian Copeland ran an excellent REBarcamp, filled with great information and attended by many who were new to the Web 2.0 space. One of the speakers I spoke with mentioned that he was pleasantly surprised by how attentive the audience was. He said that many were taking notes on actual paper, with actual pens.
Signal vs. Noise
I decided to monitor the Twitter Stream for REBCNASH based on a conversation I had the previous day about the volume of noise that was coming from conferences and how hard it was to find valuable content in what was being shared via Twitter. The claim was that these conferences were becoming polluted with noise. They were echoing the feelings Matt Stigliano had while trying to listen to the content being generated on Twitter at SXSW. And I remembered clearly watching his cry for people at SXSW to do more than just broadcast their Foursquare data. You can read about it here: Two Weeks of Social Media Hell.
This is no scientific study, but I did want to be as accurate as possible. So, I cross referenced my main monitoring, using Tweetgrid.com/irc, with Twazzup and Twitter Search,. Luckily, the Twitter gods were kind and the search stream was consistent between the three tools. In total, there were 184 tweets that used the hashtag #rebcnash that day. Those tweets were generated by 77 different people. I don’t know how many were in attendance. It doesn’t matter. What matters is that 77 people decided that they wanted to use the hashtag on Twitter to share something about the event with the rest of the world.
To analyze the content, I brought the tweet stream for REBCNASH into a spreadsheet and categorized each broadly. A tweet was either “signal” or “noise.” Since REBarcamps are learning conferences, I defined signal as any tweet that gave a piece of information that contributed to learning, or a provided a link to something that might. Everything else was considered noise.
Out of the 184 tweets, I only considered 8 to be signal. And when you see the 8, I think you’ll agree that I’m being generous. Here are the 8 “signal” tweets:
Again, I think I am being VERY generous here. Example, I included Jeremy Helton’s tweet because it might cause me to go take a look at Social Fusion. So, I counted it as signal. I could debate the “signal worthiness” of several of the others, but this should give you a sense of how low I set the signal bar. Retweets of these signal tweets (only a few) were not counted as signal.
A Closer Look At The Noise
So, the math is pretty simple. If only 8 tweets were signal, 176 were noise. Example: “no sweet tea here at #rebcnash yet but always hope. Had some awesome sweet tea the other day though. Must have more
” Which is a perfectly fine tweet, (I’ve said similar things on twitter while at a barcamp) just not signal by my definition. There were, in fact, almost as many tweets about tea, 6, as there were tweets that contained any real content.
Praise: these were tweets that simply praised some aspect of the conference without really providing any insight. An example of a praise tweet: “Can’t wait to line up the rest of the afternoon at #rebcnash.” These tweets contained the most used word at REBCNASH, which was “great.”
Questions: these were tweets that were predominantly coming in from outside of the barcamp itself. There were 17 questions asked. Only two of them were answered using the hashtag, one of them by me. An example of a question tweet: “Which is the best Twitter app for a Palm Pre? #rebcnash.” This was never answered.
Statements: these were tweets that simply made a statement, often seemingly random. An example of a statement tweet: “Learning more about twitter at #rebcnash” and “Is hanging and sponsoring #rebcnash today. Loving ‘Love is the Killer Ap dude’s jacket!” The last one could easily have been put into praise or even location as well.
Location: these were tweets that simply let people know where someone was while they were at REBCNASH. An example of a location tweet: “Second half of #rebcnash has started! (@ REBarCamp Nashville w/ 10 others) http://4sq.com/ai3HWT” The majority of these were not Foursquare posts, however, just people letting us know what session they were in.
Photos: these were tweets that contained photos. An example of a photo tweet: “#REBCNASH Schedule is Revealed! http://post.ly/dmq4″ Many of these also contained praise or a location or both, but were only counted in the photo category.
How Do We Increase The Signal To Noise Ratio?
I’m not here to debate why so little content was placed into the Twitter stream during this REBarcamp. There was no WiFi at the event, so a livestream was not possible and computer access was limited. I get it. And once again, just to be clear, those who actually attended are saying emphatically that the information shared in the sessions was excellent. Clearly, however, desire plays a role. First and foremost, you have to want to create valuable content or want to consume valuable content to make any of this work. And you certainly have to be able to identify what valuable content looks like in either case.
Personally, I’d like to do a better job of sharing valuable information. So, for those who have the desire and the ability to recognize or create good content, how do we make it easier to get more signal into the stream and get more signal out of it as well. @jazzychad has done a good job with Tweetgrid.com/irc and an even better job with Madch.at, but even those miss the mark on many levels. In this specific case, if you were interested in gleaning some knowledge from the REBCNASH stream, having the very best listening tool in the world would still have only netted you, at best, 8 potential nuggets.
And having the best tool for sharing great content only works if people actually share. From my own experience, I know I am more diligent to present quality information if I know it has some legs. It’s one of the reasons why we’re creating the Live Blog app. When I know the information I’m tweeting at an event is going to live as content on my blog, I’m more careful to make sure it’s good content.
Some Questions
I’m not sure I have the right answers for most of those questions, but one thing I do know for sure is this - I’m personally going to give more thought to the content I’m sharing at the next event I attend. I’m going to shoot for more signal and less noise. I think everyone will benefit. Including me.
Tags: noise, rebcnash, signal, Social Media, tools, Twitter
Posted in Blog, Commentary, Social Media | 46 Comments »