The kanji to either side are Label objects on ReactiveLabel objects so that a click will give you access to information - can you see at a glance which match our mnemonic panels ?
Using the interior of architecture is the classic approach since at least the Greeks and Romans - but for that we may have to wait for Google Interiors. I may be able to get permission to do a set in the interior of Government House - those might be a rather 'honorific' set of characters.
If you are not comfortable with the URL to a JPG, here is a smaller version:
And here is the applet during testing of re-sizing constraints:
