[802SEC] Final Clause 19 Report regarding P802.11k to RevCom
Dear Geoff,
Stuart is presently out of town and has asked me to follow up on your comments. He will be back in his office on Monday, February 25th.
Thank you for your comments. You will find our responses along with the email message you sent below:
<Geoff>
I am troubled by the responses to the MIB issues, to wit:
Cl 11k-D8. SC D P 127 L
Comment Type T
LB96#18-Aboba: The modified IEEE 802.11 MIB, including all the changes, does not appear to have been run through a MIB compiler to test whether it will compile.
SuggestedRemedy
Issue a MIB file including all of the changes, then run the updated MIB through a MIB compiler, correcting the errors.
REJECT.
This comment will be addressed after D9.0.
appears to claim that the problem will be fixed "after D9.0" and this appears to be a ballot report for D12.0. Thus it looks like it is time to have this problem fixed. Either you were supposed to have it done by now and the comment thus should have been removed or ???. If it was done, then why hasn't the rebuttal been updated?
<Response>
The Aboba comment was a carryover from the final Working Group Letter Ballot. Both of our MIB participants have been reassigned by their companies and no longer actively participates the Task Group and we have therefore had no MIB experts in the Task Group to run the compiler on the MIBs. We have, this week, received a MIB compiled from one of those experts and have the compile errors in hand.
There are two issues here that have been raised:
1. The MIB is presently not the same as was represented in the Aboba comment, i.e. it has been separated into two MIBs; a required MIB in Annex D, and an optional MIB in Annex Q;
2. Both MIBs have now been compiled and errors identified. There will be an errata list prepared by the TGk editor to address the changes needed for the IEEE editors. There will be explicit editorial instructions to unambiguously correct the identified errors.
<Geoff>
Further,
Cl 11k-D12 SC Q P 181 L 47
Comment Type TR
"FALSE::= { dot11RRMRequestEntry 36 }" This is the only place where the attribute is explicitly set to FALSE. No other place is this done.
SuggestedRemedy
"::= { dot11RRMRequestEntry 36 }" to delete the FALSE REJECT.
We deem this comment to be editorial. This editorial change and any other editorial MIB
changes required for error free compilation will be made by the editor prior to publication.
seems to indicate that compiler error checking will be made post-balloting. That is, they will be made at a time when technical errors may not be corrected and the inclusion of ANY corrections is subject to the whim of the SA editorial staff. That seems like a bad idea. It is not clear to me that all errors pointed out by a compiler will be (a) editorial and (b) will require a change that is so obvious that it should not require checking by the balloting group.
<Response>
Now that we have run the MIB compiler and reviewed the errors that the compiler has pointed out, we, as the Comment Resolution Committee, believe that the changes are editorial and are so obvious that they can be unambiguously and precisely corrected by the SA editorial staff per the errata list instructions.
<Geoff>
Why is running the MIB through a compiler for error checking stubbornly being held off until after balloting?
<Response>
Our MIB compiler expert experienced a company acquisition which prevented him from addressing this issue until now.
If we can be of further assistance, please don't hesitate contacting either Stuart or myself.
Best regards and on behalf of Stuart Kerry,
Harry Worstell
802.11 Working Group Vice Chair
Principal Member of the Technical Staff -
Research Communications Technology Research
AT&T Labs - Shannon Laboratory
Phone: +1 (973) 236-6915
EMAIL: hworstell@research.att.com
----------
This email is sent from the 802 Executive Committee email reflector. This list is maintained by Listserv.