Hey everyone;

just wondering if it is Good practice or choice whether to use a function like "UUID" or just teh plain built-in filemaker

serial number option?

thanks,

-i

I would love to hear people’s ideas on this as well. For me, I have use all three with the following “framework” for why:

• Built in record ID—I use this when I need a database to identify table specific records. One use would be a script that tracks which record the user is on so that it can return there after performing some task. Another would be to identify all duplicate records.
  
• A calculated serial number (record ID)—I use these when I need to track records in a table after exporting/reimporting for instance (as the built in record ID is different)
  
• UUIDs—I use UUIDs when ever I need to uniquely identify records across or outside of tables. This might happened whenever I move “records" from one table to another or even when I export the data for a collaborator.
  

What are your thoughts?

I use a UUID custom function for all my tables. I don't see myself switching back to serial numbers.

I find this technique useful for a few reasons:

I don't have to worry about dealing with duplicates when importing

I can have "helper" tables, like Notes or Files that can relate to *any* parent table ( Customers, Invoices, Payments, etc ) based on one pair of fields.

I can copy and paste tables without worrying about resetting serial numbers (not really necessary, but hard to stop myself).

Here's a good discussion of different UUID custom functions: FM Standards: Key values.