I once read that when making a database, rather than splitting up say employees from family into two different table, all are considered people and therefore all people should be put in one table where you then reference back to the table to associate the relations of husband and wife or children.

I am now looking at a database that I want to include companies with the address and phone numbers. I am wondering if it does not make sense than to just say all people and companies are Entities therefore should I just be putting in one table for all Entities. Sort of like the government in the sense that we all pay taxes regardless of whether we are a person or a company.

Part of the reason for asking is, I am trying to create a table with addresses but seems silly to create two tables, one for people and one for companies. So, I am thinking that I should have one table for address but how to deal with two separate table...one for people and one for companies.

Thoughts?

Whenever you are likely to have duplicated fields or data it is better for writing queries and storage space to combine the data into a single structure.

When you are developing a database schema you have to "divorce" yourself from the real world and simply consider the abstract where everything is a "object" with "attributes" .

As "people" and "companies" have many common attributes, names, addresses, contact details etc. You should simply consider them ALL as data entities rather than the arbitrary names we give to everything to identify them.

Agreed, so then to allow the distinction between a person and a company, simply add a field to the table that identifies what that entity is?

Yep, just set a flag that indicates what the entity represents.