I have been searching but could not find a thread that explains this:

I need to perform validation so that a calculation field is unique.

This is the situation, we work in a field where our contacts switch companies a lot. If someone tries to enter a name in company B, I want validation to check to make sure that that contact doesn't already exist in company A or C. If it does, it should warn them so they can decide to switch companies for that contact or create a new contact with the same name.

Please link me to a thread if you know where it is!

Also, I found a website that recommended a self join based on the calculation that concatenates the first and last names, then have a validation calc that counts the records. But it was for filemaker 3 and i couldn't get it to work anyways.

So basically is it the layout deciding which validation to turn to...

http://www.databasepros.com/FMPro?-DB=resources.fp5&-lay=cgi&-format=list.html&-FIND=+&resource_id=DBPros000489

But we would need to know a bit more about the relational structure? A random association leads me to think in super-type/subtype direction. Only then - might the above suggestion have some point.

Please elaborate!

--sd