The best way to look at your AP naming convention is to understand how the people who work with an AP day to day process information. If your helpdesk person saw that someone is associated to an AP with name X, would they know where that AP is physically just by looking at it? The people who most need to know that are probably the helpdesk people.
Some people put effort into putting the IDF name, physical port, mac address, etc in the name, but that can easily be switched by switching the hardware or moving a cable. The physical mac address can be obtained by looking at the controller. When an AP is moved or replaced, many times people will not update the IDF, wired mac, and physical port, and the system becomes broken. It is worth it, if people are diligent in updating the information if it changes, but a hassle if not. It is more important that helpdesk people who use the information daily have an idea where it is.
Also, consider the places where you would see the names of your devices -- they will be in the controller, in Airwave perhaps, and if you sort them, you want them to somehow be sorted correctly by region, facility, building, etc, so a heirarchical approach to naming would making sorting names in a management system or a spreadsheet easier to tolerate.
Again, these are just ideas--- What you do on a day to day basis should drive how your naming convention works...