That is, the problem stems from the type of people involved, not the network itself.

I’ve written a bit about how communities process innovations, how they propagate and how they are adopted. Each community has its innovators, its early adopters, the early and late middle and its laggards. In most communities, the relative numbers of each of these 5 groups follows a bell curve. Roughly 16% are early adopters and innovators, 16% are laggards and the majority is in the middle.

The key measure of the majority is that they will usually only adopt a new innovation when told to do so by someone, from the community, that they trust and respect. They do not like to be the first one in a community to adopt an innovation. They are comfortable with what currently works.

The innovator group, on the other hand, have a large number of connections outside the community. They bring in the odd ideas, the weird bits of information that can generate new ideas and innovations.

They are the ones who say “Hey, my Uncle Bob heard about someone who fixed the a similar problem, only he used this really weird algorithm.” Innovators love to solve problems and search the world for data that can help them learn.

Now, innovators are generally not held in very high esteem by the community. They are disruptive and as often have ideas that are useless as they do ones that are useful. They love new stuff because it is new, not always because it is useful. They are seldom seen as community leaders and often have greater freedom, either financial or situational, that allows them to pursue the novelties they love. Because of their extensive outside connections, if their work is not supported by the community, they can often leave to find those communities that will support them.

A lack of innovators means that fresh, creative ideas are not easily brought into the community.

Early adopters are the important filters here. They often have enough outside contacts to be able to understand where the innovator is coming from. They are very good at figuring out which of the many ideas that the innovator tosses out are actually useful to the group. They’re the interface between the community and innovators.

Early adopters are usually community leaders. They are the ones that promulgate the great ideas that the innovators come up with to the rest of the community. By being right, by helping the community, they gain a lot of power and prestige.

So, the majority looks to the early adopters to push innovation and change, not the innovators. The latter are just too disruptive to the clean, stable processes preferred by the majority in the middle.

A lack of early adopters means that innovators are not easily supported by the community and that useful new ideas have a much harder time getting the notice of the majority.

There are not enough filters to properly present useful ideas to the community. Innovators simply appear disruptive. Useful new ideas do not traverse the community because there are not enough trusted people presenting creative ideas.

I would suggest that the problem is not the vast number of connections amongst the groups, that the problem is not the internet. It is that these online groups, may have coalesced in ways that diminish the power of this 5 group adoption curve. In most real life communities, at least the successful ones, the innovators and early adopters number about 16%. About 65% make up the early and late middle.

Perhaps these online communities have very different makeups. Perhaps the percentage of the middle is much larger, since it is now so easy to connect, and the middle feels much more comfortable connecting with those that already think like them.

In describing these networks, the author makes the point that they mainly connect to each other. This sounds exactly like a group of early and late majority. If a community is made up of mainly people like this, say 80%, then the lack of enough early adopters could have a strong effect on the adoption of innovation.

The early adopters are the gatekeepers for novel and useful ideas in the community. If the number of early adopters is lower than normal, the number of new ideas that can traverse the community is greatly decreased. Consequently, there will be less support for the innovators, who may very well go find other communities that they can innovate with.

The ease with which the Internet allows connections to be made means that innovators will have many more routes available to them. In real life, they can not easily move beyond the community they inhabit. On the Internet, it is easy. So they may leave to greener pastures.

This may also pull along some early adopters, who, after all, like to be the ones who act as filters and to gain the community respect that comes from helping to disperse new ideas. This could result in a positive feedback loop that greatly decreases the numbers of innovators and early adopters, leaving a community of mainly the middle. This would seem to fit the description of the article.

This would explain the lack of outside connections, as early adopters and innovators have the majority of these. Without early adopters to funnel their ideas, innovators will leave for greener pastures.

On the flip side, if there were enough early adopters, their ability to pull in innovators who have ideas that would help the community, the key aspect of an early adopter, would allow the flow of innovative ideas into the community.

So how to increase the number of early adopters, which will then attract innovators and permit novel ideas to traverse the community? Well, one could advertise on Craig’s list, I suppose. Far easier would be to find a way to take the early adopter’s in the community already and find a way to increase their power, to artificially raise their numbers.

Many of the ideas suggested in the article, such as skunk work projects, are really just ways to isolate a group from the community. This would tend to increase the relative numbers of innovators and early adopters. They will be drawn to new things like ants to honey.

These are ways to prime the pump, to create a situation in the community where the early adopters have a much larger impact with higher representation than they do in the general community.

But this is somewhat indirect. Why not utilize the metrics available in the network to identify who falls into each group? Some companies are already doing this, because the way an early adopter appears in a network is different than a late majority.

Making a greater effort to identify and accumulate early adopters in the community by using the Internet itself would be very informative. Increasing the impact of early adopters would attract and support more innovators, providing more ideas to the community. If the level of early adopters is less than expected, say under 10% then efforts must be made to increase this percentage, either actually or relatively.

To bring in more early adopters would require a campaign of some sort to attract people with the right connections. Initially, this may not be easy. Better to artificially increase the relative numbers of early adopters.

So, take the early adopters that are already present and create a ‘new’ community, an artificial one, where their numbers would be much higher. Put them together, along with some innovators and let them go at it.

Again, this is kind of what is suggested in the article, but with much less discussion about why it might work. In the real world, early adopters and innovators are usually kept separate from the main group by putting them in places like Research. A difficulty with online networks is that there is not often a defined process to isolate these people and thus increasing their numbers to the point where their talents are actually useful.

In the large, efficient networks that are possible using Internet technology, the early adopters and innovators may get swamped out, becoming too small a percentage to actually affect change.

The solution is to find ways to identify and isolate them from the community but in ways that use their important attributes to help the group.