A web browser stands in-between the user and the webdesigner. The latter can define the browser’s window, while, since CSS2, the former can override the css style with his own. More generally, a standard web page is really in-between.
Here below, what would look the Reghardware at the Review section. The top navigation menu is embedded in the browser’s window, that’s notified by a bit of the chrome at the right-hand side. There are there a black dot to disable the function, and a double-arrows to collapse/expand the menu.
To get this state or to inform the user, the simplest way is to modify the pointer appearance; it could be automatically done as a browser option or define as default by the webdesigner. Anyway, the options give the choice to the user: automatic; ask to me; show to me with the pointer; never.
Now one has chosen the first article and this one goes through several pages. The list of pages can be an element to place at will by the user:
Or the same list has been defined by the webdesigner:
All that seems to works fine with a top navigation menu, what’s happen with a left/right navigation menu. Well this is the same principle as shown below, with both a top and right menus:
One retrieves the same buttons as previously, plus some up and down buttons (it’s up to the user if he/she rather like a scroll bar, since that is handled by the application)
One can argue that this is useless since styles and scripts can do the same. Yes, but one can say that the less the browser’s engine has to perform, the best it is. That is less work for the webdesigner too, though that doesn’t prevent a fallback, but it’s up to the designer. Moreover, a few website are written with the accessibility in mind, so this purpose is let to the browser and that’s an other good reason.