JNDI: Naming Service vs Directory Service

Question

As far as Java/JNDI verbiage goes, are "directory services" just more complex versions of "naming services"?

If you read the JNDI glossary, it basically defines these two as follows:

• naming service - An API that allows you to perform operations on different contexts
• context - Sounds an awful lot like a HashMap
• directory service - An API that allows you to perform operations on directory objects
• directory object - an object that is in a directory....(?!?!)

So, unless I am mistaken, it sounds like both types of services allow CRUD-like operations on different types of objects, either contexts (in the case of a naming service) or directory objects (in the case of directory services).

So my question is: whats the difference?!? Aren't both of these context/directory object types really just hashmaps/hashtables under the hood?

I'm choking on these differences because I'm trying to get an understanding of when something is a naming service, and when it is a directory service, and these definitions are difficult for me to gauge.

Answer 1

I think directory service is referenced in this acronym only to recall naming services with directory in the acronym, like LDAP.

LDAP, DNS and so on are naming services, because directory services and naming services are synonyms.

From wikipedia:

In software engineering, a directory is a map between names and values.

Note that in wikipedia, Name Service redirects to the same article.

Answer 2

Aren't both of these context/directory object types really just hashmaps/hashtables under the hood?

No, they are almost certainly database tables under the hood, and they have all kinds of semantics such as attribute lists, the ability to create child contexts, etc, that hashmaps/hashtables don't have.