2009/4/17 Todd Benson <[email protected]>:
> Thing -> relates_to <- Thing
> ...is absolute folly.

I'm not saying it's pleasant by any means. It's simply unavoidable and
perhaps even "better" (from a normalization point of view) in some
cases.

Using "complex joins" as a reason, no excuse, for not allowing this
king of relationship seems silly to me.

In any form of hierarchy (eg human genealogy, comments on comments or
any other form of "same item" relates to "same item" in some way)
you'll end up either using thing -> thing or end up with unnormalized
data, or even worse, not being able to deliver the required feature (
"I'm sorry you can only comment once on any given comment you cannot
comment on a comment on a comment" ).

Citing insert or update complexity is just silly, you write that code
once, then use it in an abstracted model where you don't care.

> Nice, succinct post by the way.

Thanks :)

--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups 
"sequel-talk" group.
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to 
[email protected]
For more options, visit this group at 
http://groups.google.com/group/sequel-talk?hl=en
-~----------~----~----~----~------~----~------~--~---

Reply via email to