It depends which you mean.
If you're treating it as a pure LDAP directory there's little difference between AD as an LDAP directory and, for example, an OpenLDAP directory, at least in terms of talking to the directory and returning data for known attributes from it.
However, having "known attributes" is the key: Many of the attributes defined in an LDAP schema are likely to differ across different directory implementations. A few standard (RFC defined) attributes do exist, but they're pretty basic; there's no need for someone else to follow the schema MS have defined for AD.
Functions written for AD itself are may fail because they are likely to depend on both LDAP and the underlying schema. An example of that would be retrieving group membership for a user. In AD that will always be the member / memberOf pair, but no rule enforces that outside of MS AD, it could be anything.
Ultimately it's extremely difficult to write any kind of code for a system you know nothing about. AD has the advantage that it's very well documented (in part because of MS, and in part because of the enormous community surrounding it).
Most of the systems you see that integrate with multiple directories present the end-user with options to define the names of common attributes. Including the attribute that defines a username (sAMAccountName, uid, etc), group membership, etc, etc.
Chris
