Yes, I could have clarified that. Here's an attempt:
"Relational" data is data whose value stems from its relationships with other data. For instance, if it is statistical data that is viewed in aggregate rather than as individual rows. Or if you need to answer the question "how many rows look like this row?"
"Object" data is data that is useful in and of itself, and is largely self contained. A pretty good example is a blog post: each post has a bunch of metadata, including possibly a string of comments. But you seldom if ever run queries across batches of blog posts (other than indexing them by date).
It's always bugged me that blog entries -- the staple of the ORM tutorial -- have little to no relational value, which is why they work so well in ORM.
Splitting blog posts by date is hardly something thats rare. See also, all blog posts by tag, all posts by author, all posts containing the word "javascript", latest comments by a particular user etc.
Well organised blogs have a reasonable amount of relational data.
"Relational" data is data whose value stems from its relationships with other data. For instance, if it is statistical data that is viewed in aggregate rather than as individual rows. Or if you need to answer the question "how many rows look like this row?"
"Object" data is data that is useful in and of itself, and is largely self contained. A pretty good example is a blog post: each post has a bunch of metadata, including possibly a string of comments. But you seldom if ever run queries across batches of blog posts (other than indexing them by date).
It's always bugged me that blog entries -- the staple of the ORM tutorial -- have little to no relational value, which is why they work so well in ORM.