[Date Prev][Date Next] [Chronological] [Thread] [Top]

Re: protocol-19 comments



>>> Hallvard B Furuseth <h.b.furuseth@usit.uio.no> 12/4/03 10:45:56 AM >>>
>[Protocol] says:
>
>> 4.1.7. Attribute and PartialAttribute
>>
>> Attributes and partial attributes consist of an attribute description
>> and values of that attribute description.
>
>This isn't quite correct. The values are the values of the attribute,
>not of the attribute description. (The value of the attribute
>description is the attribute description:-)
I can change it to say ... and attribute values.

>> Each attribute value is distinct in the set (no duplicates).
>
>...when comparing values with the attribute description's equality
>matching rule, or octet by octet when there is no equality matching
>rule.
>
>[Is that poorly worded? I don't quite know how to put it.]
I can say that no two attribute values are equivalent as described by Section 2.3 of [Models].

>> 4.2. Bind Operation
>>
>> - authentication: information used to authenticate the name, if any,
>> (...)
>> resultCode field of the BindResponse.
>
>I suggest a blank line here, to clarify that the rest of the paragraph
>only applies to simple auth.
I agree.
<snip>

>Why was section 6 (Implementation Guidelines) removed,
 
Two of the guidelines are [Models] material, and the other one (regarding referrals) was moved to the related section.

>and this sentence in section 4.12 (Extended Operation):
>
>Each request MUST have a unique OBJECT IDENTIFIER assigned to it.
In improving the flow of this section, this sentence was removed because it was redundant.
 
The current text says:
"The requestName is a dotted-decimal representation of the unique OBJECT IDENTIFIER corresponding to the request."
and later in the instructions for extended operation specifications, it states that a specification assigns the OBJECT IDENTIFIER.
 
Is this insufficient?
 
Jim