>Folks,
> I am a new user of the mere mortals framework (okay just getting to that stage..) and have a question..
>
>The current state of the main application we have here in my office, is well, a combination of Foxpro Dos code, and some VFP forms and a lot of data tables across many directories on our network..
>
>We are lookin normalizing the database and creating year independent data tables and a database container for the application.
>
>My question is, if I am getting data from one or more tables, should the access of this data be through a business object or just a view (remote in the future, local for time being)?
>
>Why I am asking is, we have a ton of code that is called, that builds these business services to return data in a cursor to forms and reports, but in using mere mortals we are trying to move to a more OOP enviroment, where forms and reports have their own data sessions and such..
>
>Any thoughts here??
>
>Thanks!
>
>Tony Miller
>Seattle City Light

Mere Mortals works best if all your tables are in databases, and your business objects are based on views. You will run into quirky behavior if you attempt to define views across multiple databases.

BTW, most business objects have a primary cursor; that is, a cursor that is acted upon. You will find switching between local and client/server a lot easier if your primary cursor is based on a view.

Sounds to me like you need to evaluate your current system, do some data normalization, and translate your business logic into business objects. Read the topic about business objects in the manual carefully. It may help define a solid direction for you, and don't hesitate to ask questions when needed.

Jim Weil