Re: objectIdentifierMatch on ambiguous name

[Hallvard B Furuseth <h.b.furuseth@usit.uio.no>]

Kurt D. Zeilenga writes:
>At 12:40 PM 2/10/2003, Hallvard B Furuseth wrote:
>>Kurt D. Zeilenga writes:
>>> Hence, I suggest replacing the PARAGRAPH with:
>>> 
>>>         While the <descr> form is generally preferred, the <numericoid>
>>>         form should be used when an unambiguous short name (descriptor)
>>>         is not available.  See Section 6.2 for additional discussion
>>>         of Short Names (descriptors).
>>
>>I think something about which context the oid has is necessary.
>>Naming context as well as as attribute type vs. object class.
>>Otherwise people will be likely to think that if they only define
>>one OID with name 'foo', the server doesn't need to know that
>>'foo' is an attribute type and not an object class.
>>Sorry, it's too late now for me to come up with a good wording.
> 
> Isn't this more appropriately covered in 6.2?

Maybe.  But wouldn't you just end up restating and expanding "the
<numericod> form should be used..." part?

Something like

"Short names may only be used in contexts where the server can
unambiguously translate them to numeric OIDs and knows what kind of
entity the OID represents (i.e. whether it represents an object class,
an attribute type, or something else.)"

> (And as a follow-up: Is this covered appropriately in 6.2?)

I can't see that it's covered at all.

-- 
Hallvard