The former for the most part (it's an example to show what *can* be
done).
Also, while it is true that for the most part we have document ==
aggregate root if you want to use DDD terminology, sometimes this is
impractical and just because you have split the data storage up
doesn't mean it's still not an aggregate root.
On Dec 21, 1:10 pm, Wayne Douglas <
codingvi...@googlemail.com> wrote:
> Hi
>
> A while ago i asked about the structure of my domain - and whether or not we
> should look at putting potentially large child collections as their own
> document and link using live projections or such. The response was that
> unless these collections will make the doc size > 1Mb not to worry too much.
>
> Reading the excellent article at:
http://codeofrob.com/archive/2010/12/16/ravendb-stackoverflow-style-v...