RE: LDAP Observation

[Alan Lloyd <Alan.Lloyd@OpenDirectory.com.au>]

Lief, the debate got skewed, hence my comment - and my response was in
the light that at least a number of X.500 vendors attempted to get
interoperable replication to work and that over time - like DSP and LDAP
access, etc the implementations will improve and become interoperable as
per smtp, X.400, TCP/IP, etc.

The point of the original observation is that LDAP has produced roughly
the same pile of paper as all of X.500, but specified at best 10% of
X.500 and the easy bit at that. LDAP (development and the system
issues)are now faced with a system model issue including: distribution,
access controls and mutual authentication and all the issues of
replication. 


Therefore, it does not matter what the standard is called - there will
always be interoperability problems - the LDAP efforts will not be any
different - because its called LDAP. In fact my point is that LDAP has
now introduced heaps of vaguaries in dealing with big or small
directories, has a mass of schema/information issues, and a mass of
scaling issues with high operational costs...

If it has take three years to reinvent DAP from a standard that is ten
years old and this only covers 10% of the directory system requirements,
then how long do we debate to reinvent the rest of it. 

If there was more effort towards implementing a single, but complete
standard, then then the testing would be more effective and
longstanding.
Re inventing wheels - just means re inventing the tests and watching the
same errors occur.

I for one believing building distributed information systems based
distributed information system standards. That way a user gets utility
and capability. I cannot do that with just "a lightweight protocol" that
can be used in so many ways. 

regards alan 

----------
From: Leif Johansson
To: ietf-ldapext@netscape.com
Sent: 10/8/98 9:06:20 PM
Subject: Re: LDAP Observation 


Dear all,

> To cite a test of a protocol that did not work too well between
> companies a year or so ago as the basis that the system engineering is
> wrong is not very useful.

Why? If the standard in question has been around for a long while
and products implementing that standard are unable to pass inter-
operability tests then surely this has to say _something_ about the 
system engineering...


	Cheers,

Leif Johansson				Phone: +46 8 164541		
Department of Mathematics		Fax  : +46 8 6126717		
Stockholm University 			email: leifj@matematik.su.se 	

    <This space is left blank for quotational and disclamatory
purposes.>