There’s a reason that the mass media became so, well, mass. It’s a communication model that works.
Communication is at once quite simple and elegantly complex. Consider Claude Shannon’s (1948) general communication system.
This is a nice model of traditional mediated communication: the message goes one way.
Notice the noise source at the bottom of the model. Noise is thought to be introduced when the signal is in some way compressed (he was writing in the Bell System Technical Journal, after all).
Engineers could concentrate on reducing the noise introduced as a function of compression and decompression.
Social media, however, introduce an entirely different source of noise. You see, one-to-one communication can scale. One-to-many communication can scale. Many-to-one communication can scale a little bit. Many-to-many communication, however, cannot scale.
Very quickly, many-to-many communication simply engulfs everyone involved.
It’s deterministic. It cannot be avoided.
On Jan. 27 one of my favorite Twitterers, @TheBloggess, captured this in a tweet to Twitter co-founder @ev:
Dear @ev, I don’t have enough time to twitter *and* keep a full-time job so I’m going to quit and twitter full time.
As I have followed more and more people on Twitter, I have found this to be the case. When I was following 400 and fewer, I could actually read all of the tweets. Now it approaches 1,100, and I have no chance.
When the Twitterverse is really cranking, they come in faster than I can read … even if this is all I do.
Some of it is noise, and some of it is really valuable information that I hate to miss. And although I have used TweetDeck to sort out some especially relevant people, every one of my followees posts updates than can add value to my life.
One of my favorite Tweeps, @dannybrown brought this up in a comment to this post.
Accepting countless thousands of followers and then saying it’s too noisy is a little like playing rock music when you’re a Chopin guy…
And he’s right. You can choose to follow just a select few. But then it’s not really social media. If you’re simply communicating with the same 30 people you e-mail anyway, then it’s just a one-to-one communication medium where others can eavesdrop.
The very brilliant part of social media — especially Twitter — is that I met Danny Brown, which I surely never would have done otherwise.
And the many-to-many scaling problem is not one, I think, that can be solved. It can be coped with, but I don’t think it can be solved.
This frustrates me tremendously, because I enjoy continuing to learn from you on Twitter. And I certainly will not abandon Twitter because of noise. There’s a very good wheat-to-chaff ratio.
But returning to the theme with which we began, there is a reason that the mass media have dominated for so long. We may abandon one delivery medium (newsprint) for another (broadband), but it’s mass nonetheless.
Mass simply works as a communications model. Engineering, not monetization, although that may be the case, too. Right, Twitter?
We’re social critters. But we’re social as a dyad or a small group. You can speak to millions, but you can listen only to handfuls.
Social media opinion leader Chris Brogan has talked about going down trying to scale. And as much as I admire his social media practices, every attempt to maintain 1:1 contact as the audience grows will indeed go down.
What are your thoughts on this social media quandary?
I'm a cognitive scientist and communication scholar who manages a psychophysiology lab at Texas Tech. I teach courses about the cognitive processing of media messages and research methods.
{ 1 trackback }
{ 23 comments… read them below or add one }
Interesting questions.
With regards the “is it really social media”, if it’s people that you don’t normally email, then yes, I’d say it’s still social – just talking to one “stranger” is being social. I very rarely email the people I tweet to – I find it’s a different contact list (with some exceptions).
But you’re right in that as the audience grows, the actual contact goes down. At least in the one-to-one sense. However, if you look at it from another point of view:
1 follower – 1 conversation
10 followers – 10 conversations
100 followers – 15 conversations (argument’s sake)
1000 followers – 50 conversations
10000 followers – 200 conversations
So although the one-to-one approach dies a natural death, you’re still conversing with more people because of the sheer numbers. Yes, you’re missing out on a helluva lot more, but that extra 10,20,50 or 100 conversations that you now have? Aren’t these social interactions?
It’s a tough one to call. Though I still believe that creating the noise, or allowing it to be created, and then complaining about it is an ironic (and probably weak) standpoint.
Danny Brown’s last blog post..Food (And Drink) For Thought
PS – Ooops! Thank you for the shout – it’s been a pleasure to connect with you
Danny Brown’s last blog post..Food (And Drink) For Thought
Danny — You have set me nerd heart aflutter. Your follower to conversation ratio is an empirical question.
Man I wish that I were a better coder. I need a Web crawler to ascertain the follower-to-conversation ratio.
It would be quite awesome if it’s a rather predictable function.
Danny Brown’s model is actually very close to the only “working model” you can get to when scaling social media.
It is almost logarithmic in nature and if we define a “maximum” of about 450 contacts, we are still within “tribe models” of personal contact (see “rules of 150 plus periphery” as researched by Richerson and Boyd, if I remember correctly, who expanded on Dunbar’s number).
Would love to get into a discussion about this via @markusvonroder.
Markus Roder’s last blog post..MarkusvonRoder: @tim_heckhausen Thankx for the help!
Markus — Stay tuned. I need to do some research to continue this discussion!
This might be a little far-fetched, but I thought about the invention of the printing press in the 15th century when I read this interesting post: The printing press was the starting point for one-to-many communication by making information available for much more people than before. But to be actually able to make a use of that, people had to learn to read, which obviously took some time.
Social media makes many-to-many communication possible, but its still young. So maybe we need some time to learn a new kind of “literacy”, to really make use of the amount of communication which is possible now (and not just be overwhelmed). Maybe one day we’ll deal with this just as easily as we deal now with books. Well, maybe not.
Daniel’s last blog post..Politikwerbung 2.0 in der Schweiz
Daniel makes a good point. I am an “old brain” kind of guy. The human brain — in largely its current form — has been around a whole lot longer than the first mediated communication (cave drawings?).
But social media are forcing us into a new era. I like that sociologists call it “ambient awareness,” according to this great NY Times piece by Clive Thompson.
We cannot know if the same depth with social media if we are to have any appreciable breadth.
To get a better understanding of how the networks function from a scalability perspective we read the book “Linked” by Barabasi. It’s a general audience book and might not be at your level as you’re a professor but it helped to understand the nature of social networks, power law distributions, etc.
Chris — Thanks for the suggestion. I will add this to the reading list. I’m a quantitative nerd, so it sounds great.
I believe social media is transactional communication on a massive level. Nothing better than being able to reach out to so many, and have them return in kind. That’s what communication is all about.
Jennifer Larson’s last blog post..Keying In To The Proper Keywords For Your Blog
Jennifer — I love the conversation, but it seems as if we must choose breadth or depth. Or unemployment and full-time social media!
I’ve only got around 700 followers, which puts me totally out of Chris Brogan’s league, but 700 conversations would be impossible, even at 140 characters. I find I “speak” frequently with the same people. The bots, those following to get followers, the wheeler-dealers, etc., don’t interact in the same way. And some people (for whatever reason) find my tweets interesting, but tweet rarely themselves. And there are some I follow (@StephenFry) and don’t really expect 1:1 interaction with. And that’s okay. There are even times when I find Twitter boring (!) because of the lack of conversation.
For me, this mix works. At least for now.
The limit begins to really stretch at about 700. And assuming any normal growth curve, the limits will continue to be tested. And the more you converse, the more you’re likely to draw new followers, right?
I’m giving a talk tomorrow about social media to the faculty and graduate students here at the College of Mass Communications at Texas Tech tomorrow. We’ll be talking about this.
And I think that Chris Brogan is secretly triplets or something in order to keep up with his volume.
The “Ambient Awareness” Argument is a good one – I had totally forgotten about the NYTimes article (evtn though I had cited it myself a couple of times before!!).
So yes, it is totally possible that with a lower depth of “communication awareness”, we could stretch the “old limits’ set by various versions of the Dunbar’s number idea.
Markus Roder’s last blog post..MarkusvonRoder: @dhatfield Thx, you are an angel
. Getting to statistically relevant terrotory slowly…
Markus — This will definitely be a fun project if I can ever get a few spare minutes to do some computations.
Upon an initial reading, my training does not suggest that the isocortex should be the limiting function, but I will have to read Dunbar’s actual papers.
The outer cortex gets way too much credit in my book.
What a great debate! May I approach from a slightly different angle? I think there are other use cases at work here than those surrounding conversations.
I don’t have hundreds or thousands of followees, so may not be fully qualified to chip in on this, but then all of us can only imagine what it must be like to be someone like Stephen Fry, who only last week passed 50,000 followers and is now approaching 120,000 as I write this. Up until a few weeks ago he would follow everyone who follows him, but he has had to stop that at around 32,000 followees – clearly he doesn’t have enough time on the day and hasn’t automated the process.
However, I can imagine that some other dynamic takes over at the point you reach those numbers. I can imagine that the medium becomes one of response rather than engagement. Stephen says something and the world feeds back in a stream of responses. Perhaps one of those responses randomly piques his interest enough to follow up with a direct message chat, but the vast majority are ripples on a surface responding to his tossed pebble.
Nonetheless, that kind of feedback engine is probably extremely useful.
Furthermore, Sam, I noticed that you referred to followees as followers at one point in your article, (the paragraph that begins “Some of it is noise…”). It’s not your followers that produce noise, but your followees, and I think this is an important distinction to make, not least because there is significant value in the difference between the two sets of people, as this is the transmission dynamic for retweeted information.
And what about that set of people who overlap – those you follow and who follow you back? They are a particular tribe but it is currently difficult to see who they are.
There are then also questions about the extent to which you prune and weed your followees, (and followers as well, but apart from blocking the obvious spambots and recruitment agencies I currently let people follow at will). And whether followees should be grouped? Is there an advantage to mixing friends, with companies, with celebrities, with pundits, with notifications? Or is it simply more convenient to keep the streams apart and therefore should grouping be a native feature of Twitter or a wider standard rather than just a feature of certain clients?
Finally, there are currently unsupported use cases. What would happen for example if a behaviour developed whereby people categorised their tweets in some standard way. I don’t mean hashtags, (although perhaps a subset of hashtags), but a set of tokens that indicate to a twitter client what the tweet is about. Is it a response to someone – part of a public conversation with one or more particular people? Is it a question to one’s followers? Is it a joke? A complaint? A call for help? What would adding a mood to a tweet be like? What if the twitter client supported a keystroke that caused the software to look at you through your webcam and analyse your facial expression while you hold down the key? Tweet, press, happy, send. Tweet, press, fed up, send.
Might any of these use cases make many-to-many comms more feasible than we might think?
Chris — These are some great thoughts. Too much, I think, to get lost in the bottom of the comments. I will try to respond to your thoughts — with proper attribution — in an upcoming post.
In the meantime, I fixed the follower/followee typo.
Briefly, what you describe reminds me of “chunking.” If you look at the perceptual limits of humans, one of the most oft-cited papers is “The Magical Number 7 Plus or Minus Two.”
Most people can silently rehearse a 7 digit number but not a 10 digit number, unless there is a “chunk” with separate meaning, such as an area code, which really then represents one chunk rather than three digits.
It’s funny because using TweetDeck, I could have all of these groups. However, I am reticent to have more than 4 columns, because the other columns will fade off screen, and, theoretically, out of consciousness.
Sorry for the long rambling post – once I started thinking I found it hard to stop!
I do think chunking is definately one strategy to reduce cognitive load, but another is to let go of the load and accept a patchy input stream.
I wonder if people had the same problem with Television when more channels started to appear…all that interesting stuff they were missing when they weren’t watching! (Or did radio teach them to let go earlier than that…?)
And I agree with you on TweetDeck – I like having a mixed stream, but it is also useful to have a short friends group to quickly catch up on the ‘closest’ tribe.
Look forward to future posts (I wonder if anyone has tried to describe a full set of emerging twitter use cases…?)
I think that there is room for all. Mass media has it’s place, but if you go to a party, you don’t want one person talking at you. You want to mingle and learn about new people and opinions. The key, as in all areas of life is balance.
Susan/Together We Flourish’s last blog post..D.O.V.E., Program for Young Women and Teens R.O.O.T.S.
It seems to me that the vast majority of platforms are really M-to-M; magazines and the super bowl are cases where multiple advertisers attempt to reach many people. Those platforms are designed to provide a bit of exclusivity for a period of time (a page or a 30 sec spot) so that the first M can be effectively reduced to ‘1′. The content is produced in such a way to maximize the second M. If successful, the business model allows for money to change hands.
Things like twitter are in the connection business so there is no effort to provide exclusivity or an audience. It is likely that the lack of emphasis will limit scalability. The result is a noisy Some-to-Some model.
Anthony Power’s last blog post..Twitter as A Broadcast Platform
Hi Sam,
Sorry for being 10 months late to this post – I just discovered your blog today! Interesting post and comments. I want to push back a little on your premise that many-to-many communication is not feasible or scalable.
I think it feasible and scalable – but I think that we need to think about the communication process in a different way. You don’t communicate and interact in 1:1 conversations in the same way that you interact in small groups. To me, M:M is a scaled small group conversation.
For example, when I participate in a Twitter chat – I am having a many-to-many conversation with several other chatters. There is usually a topic and a moderator. The moderator’s role is to kick off the conversation and keep the conversation going (like a guide). The participants direct the conversation in the direction that they want. Participants can then communicate in the following ways:
> Respond or pose questions to moderator
> Pose Questions or Make Statements to the Entire Group.
> Respond or Ask Questions to individuals
What happens is that you end up with a very fast moving group dialogue where people are responding to each other, the group and the moderator. Since, there are 500 people talking at once about the same topic, you cannot listen to every comment with equal interest. So, you tend to concentrate on and discuss the elements that are most interesting(important) to you. Other participants will retweet the comments that are most useful to the entire group.
My favorite part of these M:M digital conversations is the ability to archive them AND Go Back and Replay the dialogue. This helps me see what I missed, capture new ideas, new insights and reinforce learning.
So, those are my two cents. Thanks for starting the conversation.
- Sam
(Note: I didn’t look at all of your posts – so if you blogged about some of these points later – I apologize)
Sam Smith´s last blog ..What if You Could Google the Minds of Your Attendees
Other Sam: These are great points, and you managed to cover exactly the primary objection (not really the right word) I would raise: in that kind of volume, details get lost.
So it becomes somewhat like watching the Matrix fly by. I am likely to raise a question already asked and answered by the sheer volume. So when I have engaged in large chats, they tend to devolve into several smaller sub-chats more in keeping with the idea of Dunbar’s number.
However, you have inspired me to think more about this, as I have not really followed up on this topic.
Thanks for stopping by!
Hi Samuel,
Great debate. The constant struggle for control is very recognizable. I don’t know about you, but to me it comes in waves. At times I get frustrated and start to weed out my sources of information. But on other occasions, I kind of get into a mindful mood where I trust my skills as a surfer to stay on top of the waves and allow them to take me to where they are taking me while I’m enjoying the ride.
I’d like to know your stand on this.
There’s another thought I had while reading Chris Brogan’s post you referred to which may add another perspective to the discussion. Brogan talks about how he will go down trying to maintain a one-to-one conversation ratio while becoming overwhelmed by followers.
But what does he mean by that? Does he mean that within one of his many networks, he will fail to substantially engage in enough conversations to actually contribute something to a) the people he talks to and b) the people who listen in, all in favor of the quality of one-to-one conversations in another one of his networks (say, his family)?
Or does it mean that there is some kind of magical number of meaningful conversations a person can have per day/week/month which gets so dispersed over several connections that Brogan gets disconnected from the nature of those core-types of conversations and becomes someone who utters discrete words which only make sense if you are able to combine all interactions and conversations (which by then are so numerous it is impossible to grasp them)?
Another thing I’m wondering about when reading the post and the comments is this: how come our body can make sense out of so many stimuli which are processed seemingly simultaneously? And why wouldn’t that be the clue to deal with the many-to-many-nature of the social media environment? Because information in the form of written language has to be processed consciously? What if that isn’t true? What if the interactions do speak much louder than words? What if our minds and our desire for control get in the way of reading what’s actually being communicated within social media networks? What if we’re like surfers thinking they have to figure out every atom of the sea, the sky and the earth first in order to be able to read the waves and enjoy a great ride?