On Feb 10, 2011, at 1:38 PM, Tom Lane wrote:
> I don't deny that there are risks here. But I think the value of being
> able to move an extension when it is safe outweighs the difficulty that
> sometimes it isn't safe. I think we can leave making it safer as a
> topic for future investigation.
>
> Dimitri did suggest treating an extension as nonrelocatable if there is
> any other extension installed that depends on it. But that seems like
> more of a kluge than a nice solution, primarily because it does nothing
> for the loose-object risks. I'd rather just document that moving an
> extension post-installation might break things, and leave it at that for
> now.
+1
David