Re: [rmcat] Opsdir last call review of draft-ietf-rmcat-wireless-tests-08

["Xiaoqing Zhu (xiaoqzhu)" <[e-mail]>]

Hi Fred, 

Many thanks for helping to review this draft and sharing your thoughts. We've uploaded a revised version (-09) and have accommodated your review comments. 

https://tools.ietf.org/html/draft-ietf-rmcat-wireless-tests-09

Please find our responses below  in line. Any further suggestions are, of course, always welcome. 

Best,
Xiaoqing, on behalf of all authors


On 1/13/20, 11:32 AM, "Fred Baker via Datatracker" <[e-mail]> wrote:

    Reviewer: Fred Baker
    Review result: Has Nits
    
    Reviewer: Fred Baker
    Review Result: I have a few comments
    
    I have reviewed this document as part of the Operational directorate's ongoing
    effort to review all IETF documents being processed by the IESG. These comments
    were written with the intent of improving the operational aspects of the IETF
    drafts. Comments that are not addressed in last call may be included in AD
    reviews during the IESG review. Document editors and WG chairs should treat
    these comments just like any other last call comments.
    
    My first comment will probably be addressed by the RFC Editor, but I'll bring
    it up. In several places, the language of the document is awkward at best. One
    that sticks in my mind is the description of a network containing an 802.11b
    domain as containing an "abnomaly". I'm pretty sure it's a typo that a standard
    spell checker would have complained about, but Google thinks it's also a skin
    care product. The document would benefit from a spell check and an editing pass
    by a native English speaker.


[authors] Our apologies for the typos and thanks for catching it. The updated draft has
gone through one pass of editorial changes in an effort to improve its language usage.
Additional revision suggestions from the RFC Editor shall be very much appreciated.

    
    The second relates to a specific recommendation in the document concerning
    802.11b networks. The issue is that the dynamic range of 802.11 network speeds
    is very wide - 802.11b is theoretically capable of 11 MBPS, but typically
    achieves something on the order of 2-5 MBPS, while 802.11ac is theoretically
    capable of 1.3 GBPS. The difference in speed an introduce issues in network
    behavior. Where 802.11b is relevant, the document suggests that "additional
    test cases can be added" to cover the case. I suspect that the real issue isn't
    802.11b, although the paper cited refers to it; the issue is a network
    containing a mix of speeds with a broad range, with the slower ones perturbing
    the behavior of the faster ones. I'd suggest that the authors think about the
    fundamental issue, and make specific recommendations appropriate to the case
    (such as defining effective ranges for link speeds).


[authors] You are right that the fundamental issue is not 802.11b per se, but rather the effect of]
mixing heterogeneous link speeds over a wide range. So we’ve revised the sentence in
Sec.4 as:

	“Presence of legacy devices (e.g., ones operating only in IEEE 802.11b at a much lower
	PHY-layer link rate) can significantly slow down the rest of a modern Wi-Fi network. As
	discussed in [Heusse2003], the main reason for such anomaly is that legacy devices take
	much longer to transmit the same packet over a slower link than over a faster link, thereby
	consuming substantial portion of the transmission opportunities over the air.”

[authors ]The text of Sec. 4.3.2 (Effects of presence of legacy devices) has also been revised to emphasis
on the presence of devices with operating at heterogeneously link rates in these additional test cases,
instead of tying them with 802.11b.

    
    (for the record, I sent this to ops-dir@, but don't seem to be able to link to
    the post.)