Describe an agreed-upon semantics of the namespace stuff.

[hanjoosten]

          The problem is that we have no agreed-upon semantics of the namespace stuff. So how are we going to build it first-time-right?

Originally posted by @stefjoosten in #1467 (comment)

[hanjoosten]

The groundwork for namespaces has been done. In the Haskell code, we have the Name type, that is used now all over the place. Semantically hardly anything has changed: There are some restrictions on names:

• A name is restricted by the characters that may be used in it (Only alphanumerical, digits and underscore).
• A name consists of a non-empty list of nameparts, separated by dots.
• Each namepart must contain at least one character

The last namepart of a name is called the localname. The casing of the first character of a localname is according of the casing of names that we are all used to (Uppercase for Concepts, Lowercase for Relations)

[hanjoosten]

Now it is time to add semantics for the namespace. There have been several discussions about this in the past. My gut feeling is that @stefjoosten and @Michiel-s have slightly different ideas. Regardless if these ideas differ or not, it is good to agree upon a well defined semantic, so I (or someone else 😃) can build this one-time-right.

So please, @stefjoosten and @Michiel-s , feel free to come up with your description