RE: When to not deref aliases

["Jim Sermersheim" <JIMSE@novell.com>]

I think there is sufficient agreement that SHOULD NOT is the best (or least-evil) in regard to bind. I also feel that people agree that modDN can be worded like the other update operations (...server will not...). I don't know where people stand regarding compare. 

The X.500 compare operation deref's aliases (unless told not to). This sways one to believe the server should deref (ala section 3.3 of RFC 2251). Implementors likely don't deref on a compare. There is no way in LDAP to prevent deref on a compare, so following that type of pattern, one may be led to believe the server shouldn't deref.

Requesting feedback so I can get the bunch out of my panties (or knot out of my knickers) and move on.

Jim

>>> "Kurt D. Zeilenga" <Kurt@OpenLDAP.org> 1/24/01 6:42:04 PM >>>
At 12:10 PM 1/25/01 +1100, Ramsay, Ron wrote:
>In the meantime, I don't think we should hold our breath and SHOULD NOT
>seems a workable solution.

I concur.