> About This Site
> Tree Rearrangement
Prose Explanation (Old)This essay has been superseded by a newer version.
It turns out that a prose explanation of how the table of contents works is more desirable than I imagined it would be, and not quite as tedious.
Let me start by repeating what I said in the overview: the pages in this site are arranged in a tree. The simple word “tree” has many associations, all of which I will use freely, so you should probably read about trees and look at my examples of trees, if for no other reason than to get in the right frame of mind.
Now, notice that some of the items in the table of contents are marked with the symbol “>”. If you start from the top and read off the marked items, you get the path from the root node to the current node. For example, here's the path for this essay.
urticator.net/Essays/About This Site/Navigation/Prose Explanation
You may notice that the path looks like a plausible URL. It would actually be the URL, but for two things: first, the elements “http://www.” and “.html” are missing; second, I actually use a completely different form of URL, so that I can rearrange the tree without breaking bookmarks. See the design notes for more information.
Next, notice that there are blank lines in the table of contents, and that these lines divide the items into groups. These groups are groups of siblings, i.e., nodes with the same parent, just like files in the same directory. The parent of each group of siblings is also displayed—it is the first marked item above the start of the group. For example, the nodes Essays, Actors, and Works are siblings, all children of the root node urticator.net.
In short, the table of contents displays all the nodes from the root to the current node, along with all their siblings and children. Each item in the table of contents is a link that takes you to the corresponding node; the current node, however, is not a link, because you're already there.
Implicit in the above is something worth making explicit: the table of contents does not display the entire tree, just the parts close (in some sense) to the current node.
As you move around in the tree, you will find numerous pages that seem to contain no information. This is intentional—such pages represent nodes that are purely structural, i.e., that exist only to contain other nodes as children. The children of the root node, for example, are structural nodes. It would be appropriate to make use of the amusing phrase
This page intentionally left blank.
(which, incidentally, I can't determine the source of) but I prefer to leave the pages actually blank.
Finally, although I hope the behavior is so natural as to be unnoticeable, I should admit that the table of contents does not always behave exactly as I have described here, and that the pages are not even always arranged exactly in a tree. For complete details, see the essay Exceptions to the Tree Structure.
@ May (2000)