Where does service autonomy start and stop when concerning data management ?

[Matthieu Siggen]

Hello,

I'm really confused about how to define services when concerning data
management. I'm going to take an example to show where is my problem.

If I'm developping an application with two main business process. One
is an agenda planning service (where I could enter meetings) and the
second one is a work task manager (where I could enter tasks to do and
when to do them). Until now this seems to be two pretty cool services.
One for the agenda and the other one for the tasks management.

Now, my problem is that both of them have to access a calendar storing
when I'm occupied and when I'm free. If I wasn't in an SOA I will
store my availability in a database table. But the problem then is
that two different services access to the same table. So how to solve
this problem ? Should I make a third service for period availability
management and both other services will use this new service ? But in
this case, my two services will no more be totally autonomous. Or will
it be the job from the orchestration layer to call first my
availability service to check date availability then my agenda service
to enter the meeting and last one more time my availability service to
store this slot of time is taken?

I hope my question is understandable. Thanks in advance for your help.

Matthieu