The Incubator is most appropriate for projects which are building strong development communities in an attempt to grow to TLP status. Sqoop has been under active development for a year and while it is gaining a user base, a diverse developer base does not seem to be materializing of its own accord. Thus I don't think that going through the process hurdles of incubation are necessarily appropriate at this time.
My goal is to perform the necessary technical transitions, then get Sqoop toward a state where it can produce releases rapidly and regularly, independent of the heavier-weight Hadoop release process.
Within this framework I intend to improve Sqoop's integration with other systems (namely Hive) which could not be done within the context of the MapReduce project itself.
According to http://incubator.apache.org/incubation/Incubation_Policy.html the requirements for graduating from incubator to TLP status include:
The project is not highly dependent on any single contributor (there are at least 3 legally independent committers and there is no single company or entity that is vital to the success of the project)
I don't think it's appropriate to enter the incubator at this time, when I don't have a foreseeable plan to overcome this hurdle. Hadoop is also moving away from subprojects, so it doesn't seem reasonable to add yet another one there.
Nothing prevents Sqoop from being brought into the Incubator at a later time, if a diverse set of developers emerge to improve Sqoop. But an incubator project that has a single committer seems overwrought.