I don't know how Confluence handles web requests under the covers, but some of our users have asked whether this feature exists in Confluence (we just moved from Trac). I have since found that it doesn't. In Trac, if you have a page hierarchy like so:

• Page1
  • Page1Child
• Page2
  • Page2Child
    • Page2ChildsChild

You could access Page 2 Child's Child like http://trac/Page2/Page2Child/Page2ChildsChild and Page 1's Child like http://trac/Page1/Page1Child and so on and so forth. This is very user friendly because the URL actually means something to them.

However, in Confluence the page hierarchy as far as URLs is completely flat. The major reason we want this is because we have multiple projects per space to cut down on the sheer number of spaces we would need. For example, if we have test reports we'd like to be able to have multiple pages called 'Introduction' in the same space, as long as there wasn't already a page with that name under that level of the hierarchy (i.e. the same parent). Having to name my page 'Project 1.4.6 Test Report Introduction' is a bit extreme, not to mention ugly when you export the page hierarchy to PDF for external consumption, as the table of contents is derived from the page titles.

Now that I think about it, this may be two feature requests...

1. The ability to access a page using its hierarchy in the URL and
2. The ability to have multiple pages of the same title in the same space, enforced by parentid.