Even though this thread is old, since this is the site/thread that came up in a web search I'd like to share my experience in reply to some of the points made here.
With a job share, the individual teachers are out of the class as much as a single teacher is (meetings/other). So instead of 2 teachers, we have 2 + relief = minimum 3. Just as with 1 teacher we would have 1 + relief = 2 minimum. The minimum is largely unattainable seeing as there isn't a specific person employed for relief. When one of the job sharing teachers goes off sick for any prolonged period of time, the other doesn't take over - there is a reason why they are job sharing after all. So now the 3 teachers are 4... minimum.
I think it's important to be realistic with the figures. All the negatives applied to having one constant teacher here can be applied to job sharing where there is potential for them to have a greater negative effect. I read over and over about two good teachers being better than one bad one. This is clearly something that will happen only in the minority of cases so it is as irrelevant to this discussion as the possibility of having two terrible teachers is.
Far from being unreasonable, these concerns are common and are taken very seriously by educators. The fact is that it works for some (most?) children, but not others.
Remembering that on a bad year, 1 can equal more than 2 but 2 can equal more than 4... plus equal numbers of support staff. As has happened with us.
There are clearly steps that can be taken in individual circumstances to maximise continuity, which is something that some of the replies address regarding support staff for example. The first stage towards implementing an effective system that takes on board the needs of the staff and the children is acknowledging potential issues.
And there are, clearly, potential issues which need to be acknowledged and discussed openly and honestly. Nobody wants to take away the opportunity for teachers to job share, even though in some other industries it isn't an option because of the nature of the jobs, so I'd like to see more of a willingness to engage from some of the teachers posting here.
Can any of the professional educators here post links to scientific research that demonstrates the innocuous or beneficial nature of job sharing in the early years of education?