Hi! I'm designing the data model of an application. I have a table "people" who have a lot of attributes. Some of them are: "phone number", "AOL Nickname", "Yahoo Nickname", "MSN Nickname", "e-mail 1" and "e-mail 2" I have to options to model this: 1) Creating a table: contact_informations(aol, yahoo, msn, email2, email2) and then doing: Person has_one :contact_information Or... 2) Just putting those attributes in the table "people"? I will probably want to query the people table to get a list of e-mails of a given set of people. Will I have a significant loss of performance if I use the first option? In other words: anyone knows what's the compromise in performance when using has_one instead of just storing the value as an attribute? Cheers, Federico F.
on 2007-04-13 01:15
on 2007-04-13 01:39
What do you perceive as advantages to the first option? I don't really see any reason for it. Performance isn't an issue here, but surely the second option is just simpler, and that's a pretty good reason to go with it. -Nathan
on 2007-04-13 02:30
unknown wrote: > What do you perceive as advantages to the first option? I don't really > see > any reason for it. Performance isn't an issue here, but surely the > second > option is just simpler, and that's a pretty good reason to go with it. > -Nathan I just feel that it's prettier.. but you are right, the other way is simpler, and prettier :d