OSI-DS 40

[Colin Robbins <[e-mail]>]

------------------------------ Start of body part 1

Paul,

I have not read this paper too closely yet, but do have one comment to
make.  (Infact it can be taken as more general point that applies to a
number of the recent OSI-DS and related documents I have seen).

When defining the schema you define attributes such as

     IR       
     FTPDir
     ArchFile

these all appear to be caseIgnoreStrings (can't find the actual
definitions in the text, I assume IR is for iRName etc).
These seem to be used to name entries in the DIT, i.e., they are used
in the RDN. 
(I think section 3 on Schema needs much more text to define what the
attributes etc are to be used for, some are obvious, but not all).

Why define new attributes for naming?  I claim using commonName would
be better.

If I use a DUA that does not have any knowledge of the newly defined
schema (as is often the case with newly appearing schema), it is
unlikely that I will be able to establish any information about these
nodes -> the DUA will not know which attribute syntax is used by the
value, so will not know how to present it.  Many DUAs will choose to
"hide" entries named with unknown attributes.

If common name is used, at least my DUA is likely to present the RDN
in a reasonable fashion.  I will then see something like

     cn=FTP Archive

I can make a reasonable guess as to what is then contained below that
part of the tree, and may be able to extract some useful information.
At least I know that part of the DIT exists, and may then go about
obtaining the schema definitions.

This is a general comment, and does not just relate to OSI-DS 40:
Unless there is a real reason to use some different attribute for
naming, why not use commonName for the RDN?


Colin


------------------------------ Start of body part 2

Please be aware that NeXor was previously known as X-Tel Services Ltd.

------------------------------ End of body part 2