This is a meme that seems to be replicating its way around the university at quite a speed. It is a common concept in the IT industry, but entered our usage during a recent discussion about the relationship between innovation, information gathering, agile development and ITIL managed service delivery. As a way of accomodating dynamic and responsive innovation within a managed and highly accountable service delivery environment, it is of great interest to people beyond IT.
So what then is an 'incubator'? Consider the egg analogy (which almost works). An egg contains a growing organism that may develop into something relatively independent and strong. During the growing stage it is safely stored in its egg, with a local plan and store of nutrition to keep it going until it is ready to hatch (remember that eggs have to be given sufficient nutrition from the start). We can also place the eggs within an incubator, which controls the environment in which the develop and which prevents them from being smashed. Hopefully, we will find that at least some of our eggs hatch out into chicks, and we can release those chicks into the outside world to roam free and happy.
But why did we remove the egg from its mother? Surely she had a lovely nest in which it could grow? (This is stretching the analogy now). Perhaps it was the case that she couldn't cope with this new development, so we took it away for a short time so that it could grow without affecting her directly (definitely streching the analogy too far).
OK, back to reality. Imagine the mother hen to be a service delivery organisation (there are several within an institution like a university). Often such services are so well tuned to what they currently do that they cannot easily deal with innovations. These innovations may even be disruptive technologies that threaten some of the basic principles around which the service is organised. The service organisation knows that it should investigate the innovation, to discover if it should invest time, money and change energy into it. But that investigation itself would require too much of a commitment. For example, the service would not want users to think that it endorses the innovation and incorporates it into its service catalogue (and hence provides support for it through its helpdesk). This is where the incubator becomes useful.
The service may create a parallel 'investigative service' to make the innovation available, whilst not incorporating it into its central operation and not having to permanently commit to support people in using the innovation. This parallel operation is the incubator. It acts to get people to try an innovation, help develop and fine tune it, and gather information about its effects (especially useful when a new technology may be disruptive). Note that the most important thing is that incubator projects are always clearly identified to end users as such, so that they know that it is not fully supported and guaranteed.
My conjecture is that any organisation that aims to be efficient and scalable, but at the same time innovative, should have an incubator.