Hi,

I've been teased by the Relative Object Identifier on a more philosophic
level.
(ITU-T Rec. X.680 (1997)/Amd.1 (1999 E))

Actually what is bothering me is that there is no syntax for linking
RelativeObjectIdentifier with whatever is the oid-root.
It seems to be something the author must write in comments.
(And in principle, whatever is in comments is not of interest to a
compiler!)

Ideally I believe a decoder generated by an ASN.1 compiler should be
able to detect any protocol violation without human interaction.
(I.e. without any additional program code or compiler specific
modification of the specification etc.).

For the Relative Object Identifier I believe this validation according
to a protocol is not possible, as there are no way the compiler can
determine which oid-root to use.

I believe it would be possible to make it 'compiler safe' by introducing
a notation similar to the notation used for 'Table constraints' in
X.682.

I know from the book of Olivier Dubuisson that a notation for ROOT has
been proposed and did not become a part of the standard. Actually that
would not have solved the problem, because, as I see it, the linking
between RelativeObjectIdentifier and the ObjectIdentifierRoot was still
missing. 

Best regards
Egon Andersen
-- 
* Talura ApS      * Phone: +45 43 52 50 00 *
* Baldersh�j 24 B * mailto:[EMAIL PROTECTED]  *
* DK-2635  Ish�j  * http://www.talura.dk   *

Reply via email to