Re: a heretical question about LDAP in general

[David J N Begley <david@avarice.nepean.uws.edu.au>]

On Sat, 13 Nov 1999, Tom Spindler wrote:

> Have there ever been any proposals to simplify LDAP? That is, to make the
> actual protocol octet-based rather than relying on ASN.1 and BER?

Many protocols use some form of encoding to make heterogeneous communication
more consistent - for example, NFS atop XDR atop RPC atop UDP atop IP (here,
XDR is the encoding).  LDAP (having been derived from X.519 DAP, but using a
more restrictive form of BER) is not alone in using ASN.1 - see SNMP, for
example (ASN.1 is from whence all those OIDs derive).

> And possibly have a standard simplified API as well? (I ask, because in all
> honesty, I only see about 10 API calls that are ever used in any LDAP-based
> programs.

The LDAPv2 API mapped to C is published as IETF RFC 1823;  the LDAPext working
group in the IETF is working on updating this for LDAPv3, and also providing a
mapping to Java.  Additionally, there are other federated naming APIs that can
provide some form of access to information stored in LDAP directories (ADSI
for Microsoft Windows, JNDI for Java, etc.).

> I found it "interesting" that the hardcover LDAP book by the ex- umich
> folks almost solely talks about the API rather than the protocol itself -
> but who really WANTS to talk about ASN.1?

Which book - "LDAP: Programming Directory-Enabled Applications With
Lightweight Directory Access Protocol" by Howes and Smith?  Given that it's
about programming applications (as opposed to an engineering dissection of the
protocol) I'm not surprised it concentrates on the API!  ;-)

As for ASN.1 - see above;  other standards also use ASN.1.

Cheers..


dave