Normally at the end of the mock
you should be able to review the questions/answers and see what's wrong and find the correct answer, but I'd rather discuss about why some of your answer are probably wrong.
For instance :
7) Entity bean for Catalog retrieval
The catalog is probably something huge, with a lot of products, so I think it is fair to assume that each customer will get only o very small part of it at a time, based on some selection criteria. Basically it will be a small and static view of a very limited part of the catalog, with the only possibility of picking a product at a time and adding it to the shopping cart. So the client won't modify or alter in any other way the catalog itself, it will have just a snapshot of some data. So, no transactions involving the catalog itself, no persistence to update, no concurrency between different clients to be concerned about - each one has its static, read-only list of products, eventuallly some concurrency on the product level might be useful, but not on the catalog level.
On the other hand it looks more like a service call, gimme a list of so-and-so-products, service provided by the store to actual and potential clients, service that doesn't look client related, just a short, one-time query and retrieval of the list. Hence it looks like a stateless session bean to me.
Billy
