From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-yw1-x1135.google.com (mail-yw1-x1135.google.com [IPv6:2607:f8b0:4864:20::1135]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by lists.bufferbloat.net (Postfix) with ESMTPS id 143853B29E; Sun, 19 Mar 2023 20:26:08 -0400 (EDT) Received: by mail-yw1-x1135.google.com with SMTP id 00721157ae682-5416b0ab0ecso197314837b3.6; Sun, 19 Mar 2023 17:26:08 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20210112; t=1679271967; h=cc:to:subject:message-id:date:from:in-reply-to:references :mime-version:from:to:cc:subject:date:message-id:reply-to; bh=5hYaluKaJurUTuCw7mgyZ7OSUc+4xUgmrqfI9JZo8cw=; b=ZHxOU5Q59gc8kmtuwpVs8arQ06GeEV28Nifwh/ZUqdH7lib7QmUIaUfOjDSgIZyydz s2WmSfYqMT/SBXD3xAVOshFxNCzDz+slMzoLCl7q3PlEQlW+Nf3h+eGybT9N3kLvxP7K cPmTdZABs4/ZziIF5VhQ1pY2XQ/33tAMuFXjPtbdF+ncDGHgc+wnvOAsqPLA74PLn1LL t8FxGVMLd45GwXHktaCN0S9CNlCw+wzQCsHN/8Q67sZWI8dyJMSiJ+/NiVfC7EDD6ohq LYeNekbrtu6gqlr27UAcErAtBUk9g9STVt8GqLueAv/beGXdWgXO/mfe69l8erzzSqV2 3KwA== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20210112; t=1679271967; h=cc:to:subject:message-id:date:from:in-reply-to:references :mime-version:x-gm-message-state:from:to:cc:subject:date:message-id :reply-to; bh=5hYaluKaJurUTuCw7mgyZ7OSUc+4xUgmrqfI9JZo8cw=; b=lvgsbgX2ReRM+ZW3dXfvpmSnOcdOU1PXD1fnau6oEP8wHdu6esUHpvW4rBOprkJQ4B o83P118QyDNB8byIfYk2ifdcq7fMl+S7UvQSyudvzJut0V7eqHBOw49AG/fXikXtQcC6 2XXDK8tikXE8LTDdz/1jkvTfN6+b1gfXveKMsHuFViuY6KwNTcJ3Tr6F5+6mKj0BIIYe yeouM+9AlhuLNcH5eqK95+rnnFWJiuH658p6jZ+SNtBWil2xj/jgbNawEbrsalg8X/KU BZaWnukGlNAHbaTBPCLaPvdD1N2uuE0M2SckwU8KJZ/8jma3sUjw2Dau4dIBwi58kZ9f FTSw== X-Gm-Message-State: AO0yUKUtZJGFJKjbkqetRZlrEtE34Z8eOAC/g8a0KKWg7u8H9/Z/nS13 RiMW2Tnd3ZV/QlVUTQaZByOHMkqe0qxDQpMzOd/xctSx X-Google-Smtp-Source: AK7set9qdy2FBYPltlgPUo4V5RiG3+2gsUBdIbANabkIpFPr5TL4MavMD2tTIVcVO12r9wWBZqJapfGx609qxFBRay4= X-Received: by 2002:a05:690c:731:b0:544:8405:e1f9 with SMTP id bt17-20020a05690c073100b005448405e1f9mr8774369ywb.1.1679271967226; Sun, 19 Mar 2023 17:26:07 -0700 (PDT) Received: from 1064022179695 named unknown by gmailapi.google.com with HTTPREST; Sun, 19 Mar 2023 17:26:06 -0700 Received: from 1064022179695 named unknown by gmailapi.google.com with HTTPREST; Sun, 19 Mar 2023 17:26:04 -0700 Mime-Version: 1.0 (Mimestream 0.41.5) References: <1672786712.106922180@apps.rackspace.com> <77CCAD19-07E0-4F9E-88C1-D207CF7BF376@cable.comcast.com> <83ffc0dad19e3343e49271889369cefc@rjmcmahon.com> <3CD0B9E6-0B2A-4A70-8F53-ED0822DF77A6@gmx.de> <13DE6E53-665F-4C20-BBE2-70E685421E9D@gmx.de> <22C819FA-DDD7-4B9B-8C09-8008D4273287@gmx.de> <5e7fac51071bdbb20837e72e7eedfc7c@rjmcmahon.com> <3f45d2a0b6e46d7b2775fb801e805f93@rjmcmahon.com> <70F71290-C6CB-4D19-8A88-F0F17C0BDDA2@gmx.de> <5e0cd693c4749d128dbb48d6c1129071@rjmcmahon.com> <2ab2983d-6beb-49cb-8c35-e481cbfdc7a3@Spark> <8F56CCA3-61C4-475F-975D-99D851C6A7CF@gmx.de> <1d6c10c9a692bb3f2869fb1b40fa449a@rjmcmahon.com> <005d1e7e3e1d19bce308436e46a3ec5e@rjmcmahon.com> <569691b3e7dfc57bbf98c4fc168fc6cf@rjmcmahon.com> <2885829.1679221616@dyas> <0649f7dc2b6e7ca1634b675dc6f9d953@rjmcmahon.com> In-Reply-To: <0649f7dc2b6e7ca1634b675dc6f9d953@rjmcmahon.com> From: dan Date: Sun, 19 Mar 2023 17:26:06 -0700 Message-ID: To: rjmcmahon Cc: Rpm , libreqos , Dave Taht via Starlink , bloat , Michael Richardson Content-Type: multipart/alternative; boundary="00000000000016843605f749f97b" Subject: Re: [Rpm] On metrics X-BeenThere: rpm@lists.bufferbloat.net X-Mailman-Version: 2.1.20 Precedence: list List-Id: revolutions per minute - a new metric for measuring responsiveness List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 20 Mar 2023 00:26:08 -0000 --00000000000016843605f749f97b Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable On Mar 19, 2023 at 3:00:35 PM, rjmcmahon wrote: > Hi All, > > It seems getting the metrics right is critical. Our industry can't be > reporting things that mislead or misassign blame. The medical community > doesn't treat people for cancer without having a high degree they've > gotten the diagnostics correct as an example. > > An initial metric, per this group, would be geared towards > responsiveness or the speed of causality. Here, we may need to include > linear distance, the power required to achieve a responsiveness and to > take account of Pareto efficiencies, where one device's better > responsiveness can't make another's worse. > > An example per a possible FiWi new & comprehensive metric: A rating > could be something like 10K responses per second at 1Km terrestrial > (fiber) cable / 6m radius free space range / 5W total / 0-impact to > others. If consumers can learn to read nutrition labels they can also > learn to read these. > > Maybe a device produces a scan code qr based upon its e3e measurement > and the scan code qr loads a page with human interpretable analysis? > Similar to how we now pull up menus on our mobile phones listing the > food items and the nutrition information that's available to seat at a > table. Then, in a perfect world, there is a rating per each link hop or > better, network jurisdiction. Each jurisdiction could decide if they > want to participate or not, similar to connecting up an autonomous > system or not. I think measurements of network jurisdictions without > prior agreements are unfair. The lack of measurement capability is > likely enough pressure needed to motivate actions. > > Bob > > PS. As a side note, and a shameless plug, iperf 2 now supports > bounceback and a big issue has been clock sync for one way delays (OWD.) > Per a comment from Jean Tourrhiles > https://sourceforge.net/p/iperf2/tickets/242/ I added some unsync > detections in the bounceback measurements. Contact me directly if your > engineering team needs more information on iperf 2. > A food nutrition label is actually a great example of bad information in consumer hands. Since adding those, Americans weights have ballooned. I= =E2=80=99m not saying they are in direct correlation, but that information has definitely not caused an improvement in health by any measure at all. Definitely not a model to pursue. There needs to be a clear distinction between what=E2=80=99s valuable to th= e consumer and what=E2=80=99s valuable to the ISP to improve services. These = are dramatically different pieces of data. For the consumer, information that directions their choice of product is important. Details about various points in the process are useless to them. How many hops has no value to them, only the latenc, jitter, throughput, and probably some rating on slow start or other things that are part of the ISP equation but entirely for the purpose of making better choices for their needs. 10k responses in x seconds is meaningless to a consumer. I have never, and I=E2=80=99m not exaggerating, EVER had a home user or IT guy or point person for an MSP ever ask about packet rates or any of the stats that keep getting brought up. This is a solution looking for a problem. Consumers really need things like published performance specs so they can assemble their needs like an a la carte menu. What do you do, what=E2=80= =99s important to you, what details support that need, and they need that in a simple way. Like a little app that says =E2=80=9Chow many 1080p TVs or 4K= TVs, how many gaming consoles, do you take zoom calls or VoIP/phone calls. Do you send large emails, videos, or pictures.=E2=80=9D Put another way, all the specs are like telling a soccer mom the torque curve of their minivan. If =E2=80=98we=E2=80=99 the industry make a nutrition type label that has n= umbers on it that are not useful in the context of a consumer decision making process in a direct x gives you y way, it creates data that will get misinterpreted. These stats should be made available for the providers pushing data so that they can make sure they are meeting the human readable and useful data. I care about the AS path and the latency on my upstream services to various providers, I can try to make better choices on buying lumen or hurricane and how I will route those things because I understand how they affect the service. Those are really useful numbers for me and any ISP that wants to have higher =E2=80=9CA+ for gaming because latency and jitter are low and b= andwidth is adequate=E2=80=9D ratings. --00000000000016843605f749f97b Content-Type: text/html; charset="UTF-8" Content-Transfer-Encoding: quoted-printable





On Mar 19, 2023 at 3:00:35 PM, rj= mcmahon <rjmcmahon@rjmcmahon.= com> wrote:
=20
Hi All,

It seems getting the metrics right is critical. Our indu= stry can't be
reporting things that mislead or misassign blame. The= medical community
doesn't treat people for cancer without having a= high degree they've
gotten the diagnostics correct as an example.<= br>
An initial metric, per this group, would be geared towards
respo= nsiveness or the speed of causality. Here, we may need to include
linea= r distance, the power required to achieve a responsiveness and to
take = account of Pareto efficiencies, where one device's better
responsiv= eness can't make another's worse.

An example per a possible = FiWi new & comprehensive metric: A rating
could be something like 1= 0K responses per second at 1Km terrestrial
(fiber) cable / 6m radius fr= ee space range / 5W total / 0-impact to
others. If consumers can learn = to read nutrition labels they can also
learn to read these.

Mayb= e a device produces a scan code qr based upon its e3e measurement
and t= he scan code qr loads a page with human interpretable analysis?
Similar= to how we now pull up menus on our mobile phones listing the
food item= s and the nutrition information that's available to seat at a
table= . Then, in a perfect world, there is a rating per each link hop or
bett= er, network jurisdiction. Each jurisdiction could decide if they
want t= o participate or not, similar to connecting up an autonomous
system or = not. I think measurements of network jurisdictions without
prior agreem= ents are unfair. The lack of measurement capability is
likely enough pr= essure needed to motivate actions.

Bob

PS. As a side note, an= d a shameless plug, iperf 2 now supports
bounceback and a big issue has= been clock sync for one way delays (OWD.)
Per a comment from Jean Tour= rhiles
https:= //sourceforge.net/p/iperf2/tickets/242/ I added some unsync
detecti= ons in the bounceback measurements. Contact me directly if your
enginee= ring team needs more information on iperf 2.

A food nutrition label is actu= ally a great example of bad information in consumer hands.=C2=A0 Since addi= ng those, Americans weights have ballooned.=C2=A0 I=E2=80=99m not saying th= ey are in direct correlation, but that information has definitely not cause= d an improvement in health by any measure at all.=C2=A0 Definitely not a mo= del to pursue.

There needs to be a clear distinction betw= een what=E2=80=99s valuable to the consumer and what=E2=80=99s valuable to = the ISP to improve services. These are dramatically different pieces of dat= a. =C2=A0 For the consumer, information that directions their choice of pro= duct is important.=C2=A0 Details about various points in the process are us= eless to them.=C2=A0 How many hops has no value to them, only the latenc, j= itter, throughput, and probably some rating on slow start or other things t= hat are part of the ISP equation but entirely for the purpose of making bet= ter choices for their needs. =C2=A010k responses in x seconds is meaningles= s to a consumer.=C2=A0 I have never, and I=E2=80=99m not exaggerating, EVER= had a home user or IT guy or point person for an MSP ever ask about packet= rates or any of the stats that keep getting brought up.=C2=A0 This is a so= lution looking for a problem.
<= br>
Consumers really need thing= s like published performance specs so they can assemble their needs like an= a la carte menu.=C2=A0 What do you do, what=E2=80=99s important to you, wh= at details support that need, and they need that in a simple way. =C2=A0 Li= ke a little app that says =E2=80=9Chow many 1080p TVs or 4K TVs, how many g= aming consoles, do you take zoom calls or VoIP/phone calls.=C2=A0 Do you se= nd large emails, videos, or pictures.=E2=80=9D =C2=A0

Put= another way, all the specs are like telling a soccer mom the torque curve = of their minivan. =C2=A0

If =E2=80=98we=E2=80=99 the indu= stry make a nutrition type label that has numbers on it that are not useful= in the context of a consumer decision making process in a direct x gives y= ou y way, it creates data that will get misinterpreted.

T= hese stats should be made available for the providers pushing data so that = they can make sure they are meeting the human readable and useful data.=C2= =A0 I care about the AS path and the latency on my upstream services to var= ious providers, I can try to make better choices on buying lumen or hurrica= ne and how I will route those things because I understand how they affect t= he service.=C2=A0 Those are really useful numbers for me and any ISP that w= ants to have higher =E2=80=9CA+ for gaming because latency and jitter are l= ow and bandwidth is adequate=E2=80=9D ratings. =C2=A0

--00000000000016843605f749f97b--