When you examine an object, there are several ways to handle it in your code. If it's scenery, you can use a match statement in the locations{} block or the on_command{} block. If it's an object, you can use a match statement in the on_command{} block. If it's scenery or an object that doesn't require any conditional logic, you can use examine_message in the objects{} block.

This is probably just a question of style, but is there any "best practice" for locating code for things like examine? I'm also curious about which is more efficient. I'm guessing that placing it in the examine_message is more efficient as it requires less code. The thing that I'm uncomfortable about is that command handling is scattered about in different places. This makes it harder to maintain your code. Any thoughts on this?

I too find the code a bit unwieldy when examine commands are in different places, as when a game project gets big I can never remember where I've put things (I end up keeping a notepad file with the entire code open so that I can Ctrl+F).  As such, what I've been trying to do is keep everything under on_command and in alphabetical order, despite the fact that it would be a lot easier just to use examine_message (this also helps if I later decide that a particular object IS going to require conditional logic).

It is a matter of style but I'd personally favour one of three approaches

1. Put EVERYTHING in on_command {} block. All examine logic for objects AND general location logic goes in here, nowhere else.
2. Add a : gosub "examine_messages" ; to the on_command{} block and put everything in a new subroutine called "examine_messages". All examine logic for objects AND general location logic goes in here, nowhere else.
3. Put all examine per-object examine logic in the objects definitions themselves, and put all per-location examine logic in the locations themselves.

   lamp : object "a lamp" {
      start_at = my_location
      on_examine {
         : print "Hello" ;
         : done ;
      }
   }

Location-based examines sounds like not a good thing to do, though there are exceptions of course. I can speak only for myself but I'm usually working with a different system to develop my games, which gets then imported in Adventuron. There I'm adding an examine message that is based on the fact whether the object is present at the location where you're invoking the code or not. So my examine commands are mostly object based. Sometimes they are location based because there is no object but maybe something in the room description that wants to be examined. But even in that particular case I have the code all at one place. That didn't change in Adventuron btw. but as far as I understand my code is mapped and created different to standard Adventuron code.