14 September 2022


Alex_Dawson, annette_g, AramZS, brwalder, cel, Chris_Needham, Dan_Druta, gendler, igarashi, ivan, Jake_Holland, Jay, jgraham, Judy, Mike_Prorock, mprorock, Nick_Doty, npd, Phil_Archer, phila, tantek, Tantek_Celik, Tatsuya_Igarashi, tidoust, Tim_Frick, Tzivya_Siegman, valerie_young, Yoav_Weiss
Nick_Doty, Tantek_Çelik
cpn, yoav

Meeting minutes

<tzviya> Date: 2022-09-14

Nick: important ans timely topic
… background in privacy, but this is an urgent topic. Want to learn and see how we can apply to web standards
… outlined goals we could work on as a community
… ientified areas for potential standard work and cover ongoing work
… considering metrics that would enable other measures to take effects
… Also measure the impact of the W3C meetings themseleves
… wanted to talk about working mode about how we talk about sustainability in W3C

maybe horizontal reviews
… A topic that has come up at multiple meetings and TPACs

tantek: one of Mozilla env champions
… been 3 years since TPAC 2019, some discussions there
… had an AC meeting session this past spring

<npd> yes! remote meetings has been such a promising way to mitigate our direct carbon emissions impacts

tantek: virtual session in 2021
… The TAG has included sustainability as a core principle
… Moving forward with the assumption of sustainability being a principle
… If folks want to argue, there's the TAG issue
… Created sustainability CG to create a culture and process of horizontal review
… In addition to horizontal review, want to allow folks to talk about our biggest impact
… Need prioritization, so wanted to ask the group

jholland: multicast CG which has sustainability implications
… one of the motivating factors in adopting multicast
… BBC report about network consumption
… Can address a lot of that with multicast
… Would be great for it to get more attention
… large events take 10Ks servers to serve, and protocols can cut that down significantly
… RE the notion of tracking our own impact
… worthwhile to build models, but the impact of the web can be more effective
… don't want to discourage people from flying here, if it solves the broader problem
… need to be explicit about that

Nick_Doty: Our own emissions is not a substitute to other work

jholland: need to not be distracted

judy: few quick comments: was a session on sustainability at the W3C AC meeting in Q2
… we had a strnger response from membership than any other AC committee
… interest distributed across different topics
… own emissions wasn't covered by the AC
… The hybrid meeting format can be used and help other orgs reduce their emissions
… efficiency of standards - heard impressive stats of CO2 that can be saved

<AramZS> Alex looks like he wants to queue but I don't remember how to queue someone else?

yoav: I'm co-chair of Web Perf WG. I've thought about what are the lines connecting web performance and sustainability
… Looking at emissions, and different aspects of CO2 emission
… a lot of it is consuption on the device, e.g., less JavaScript

<tzviya> +1 to yoav

yoav: another part is people replacing device because they're too slow

<AramZS> Thanks Judy, can you let him know via Zoom chat mby? I assume he isn't on here.

yoav: one thing missing is everything server side. There's no way for servers to tell us how much energy was used. Could we make some voluntary standard, and later some enforcement mechanism

npdoty: Is that something clients may potentially want to expose?

yoav: Haven't brought it up in Web Perf WG, hope to this week. It could be interesting for clients to surface, e.g., a sustainability mode, similar to battery saving mode

<tantek> https://www.w3.org/wiki/Sustainability

<jholland> responding to yoav: the comments about developing server tracking of impact reminded me of https://www.greeningofstreaming.org/ which has started doing some work in this direction

tantek: We're going to have to liaise with other groups. We have a public wiki page, feel free to add liaison opportunities

jgraham: speaking for myself. Informal chat about CO2 emissions
… improving the web can save a fraction of a percent
… reducing carbon emissions is already incentivized

<mprorock> https://www.epa.gov/ghgemissions/global-greenhouse-gas-emissions-data

jgraham: browsers gotten faster and more energy efficient, but sites got slower
… one of the problems is that efficiency is eaten by more content
… smartphones are faster than they were, and every year total emissions of IT keep growing

<mprorock> https://escp.eu/news/reduce-your-digital-carbon-footprint-shape-greener-future

jgraham: the scale of the problem requires more than simple choices and optimizations
… if we can't get the web running on renewable/nuclear power, than any of the other things won't work
… pretty clear how to make electricity sustainable
… web has to run on renewable energy
… anything less than doesn't support the solutions
… we do not know how to make flying sustainable
… we don't know that video calls can replace flying. Don't know if this is something the W3C should have an opinion on this
… dominated by system effects and we need to look at those

<Philip> help

<Zakim> tzviya, you wanted to talk about impact

tzviya: in terms of impact, we've heard about floors and ceilings in the past RE compliance
… Need to have something obtainable - e.g. only renewables may be hard to get to
… We can talk about how this improves performance and a11y
… horizontal review doesn't work unless it's something people implement themselves

mprorock: work directly in sustainability. measure data on this
… confirm scientifically a lot
… Can do a lot of feel-good things, or tackle the real world problems
… Maybe can start help define how to describe the data for that
… customers in agriculture - how can we measure pesticide, gas, etc
… if we can do that in a standardized way and get everyone on board, can save 20%
… past the point of no return, we'll face consequences
… we can find ways to tie verifiable creds and enable real people mitigate real impact
… can we facilitate people that actually solve this problem

DanD_: thrown off by the diff between title and conversation
… relying on the UN 17 goals. You're talking about number 13
… would be useful to align to those targets

<AramZS> https://sdgs.un.org/goals/goal13

<mprorock> +1 aram

DanD_: sustainability goals are beyond energy efficiency

<Zakim> AramZS, you wanted to talk about the need to id effective mitigation strategies

AramZS: lot of desire to visibly do something about this on the brand level

<npd> I also think we could consider other sustainability goals too! energy efficiency and climate change is an urgent need, but doesn't need to be the only topic we discuss, if there is interest in doing that work

AramZS: general goal to demonstrate sustainability
… clear interest in spending money on this desire
… big concern - a lot of room for charlatans to step in
… not a sustainability person, so can't assess solutions
… need good metrics

<mprorock> +1 judy

AramZS: every measurement framework needs a solutions framework - generate a report
… A lot of the solutions are focused on carbon credits or planting trees
… doesn't seem to be an effective way to go
… In ad tech, there's a lot of data we could use and say that a solution doesn't help

<Zakim> tantek, you wanted to propose that the Sustainability CG work in public by default, as well as sustainability at W3C in general, and that we provide that request to the AB to make that happen on our behalf

tantek: proposal - wanted to get a series of next steps

<AramZS> For those interested, here's my opening research into carbon offsets https://context.center/topics/carbon-offsets/

tantek: agree to work in public by default - want to make a request to the AB to make that happen

<Judy> +1

<tzviya> +1

<mprorock> +1

… at the AC open session, ethical web principles say "the web must do no harm"
… Need to not make the problem worse

<tantek> Proposed: Sustainability CG work in public by default, as well as sustainability at W3C in general, and that we provide that request to the AB to make

<spectranaut_> +1

<jholland> +1


<gendler> +1

<phila> +1

<Cornelius> +1

<Jay> +1

<annette_g> +1

RESOLUTION: Sustainability CG work in public by default, as well as sustainability at W3C in general, and that we provide that request to the AB to make

<tantek> per Tess's presentation earlier today, first the web must not do harm

Jay: very complex problem. we consume a lot of energy
… may be out of scope for W3C activity
… very difficult to predict energy impact

Alex: sustainability CG. when people look at variables that impact, we don't have a framework that counts all the variables that cover the end-to-end journey

<AramZS> Is the group he's referring to https://www.w3.org/community/sustyweb/ ?

Alex: user pathway and developers. Got general idea about how much CO2 is used, but never had a more global idea about looking into all the different variables
… and here are the ones that we need to actually look in to
… Don't know where our blind spots are
… need to build a map of the things we need to take into account


gendler: news corp is new to the W3C. mostly consumers of deliverables

<AramZS> +1 on a principles document being really useful as a starting document

<jgraham> +1 to looking at policy/principles

gendler: principles documents really assist in pushing initiatives in corps. I propose a principles document which will really help CEOs understanding the problem

<mprorock> +1

gendler: principled writing helps make stronger internal pitches

<tzviya> +1 to a principles (and I think mprorock needs to help write it ;)

<Zakim> phila, you wanted to talk about certification and to talk about DPPs

phila: working with national bodies to put data on product conformance
… in the EU, all products have tracability to make sure where they come from
… both are about data being presented to end users
… we're not good at exposing data to people
… commercial world provides information about their stuff

<gendler> Adding for the record (forgive my formatting if I'm doing this wrong), I intended to point out that both Ethical Web Principles and ML Ethics Principles documents have been used to make internal use cases at NC as rationale for hoping for a similar sustainability document.

phila: more impact if we help people see what damage their consumer goods are doing

<mprorock> +1 phila

phila: datasphere is not recognizable
… browsers should recognize schema.org

spectranaut_: excited that this is a topic of conversation. Igalia is trying to create a positive impact on env
… struggling with questions, with no answers, but a bunch of Igalians would love to help
… fun to work on concrete stuff

<Zakim> tantek, you wanted to note that our individual or even group actions can only do so much, to have the biggest impact from a systems perspective we need collective action including political advocacy, and ask how can we consider our work here in that larger context?

tantek: talked about what we can do individually to prevent more harm. but so much we can do as individuals
… gonna need collective action beyond W3C
… wanted to check motivation to do such collective action, including political action

Judy: Think it's almost a myth that we can't do policy work. If you do good work, people will come
… several orgs try to come up with metrics. Things W3C could do to facilitate... but possibly if one jumps ahead too early on that, before having something good to point to, might not land so well

mprorock: work often on the policy side. If we can bring something concrete to the table - e.g. connect directly to the end users

<Jay> +1

mprorock: deeply rooted in what the W3C is

tantek: also connects to principles

mprorock: distributed the principles docs across and they were well received

tantek: technologies to reduce energy usage
… w3c impact on its operations
… measurement
… system effects
… horizontal review

<mprorock> Can we ensure use of invited experts, for instance from the national labs?

tantek: Principles

ivan: does measurement cover what mprorock and phila talked about?

tantek: yes

<Zakim> annette_g, you wanted to respond to mprorock's mention of national labs

annette_g: not an expert, but plenty people at the national labs are. Would see if someone is interested

tantek: wanted to capture who's interested in any of the areas

<tantek> tantek: [system effects][horizontal review][principles]

<Judy> [JudyB: I will continue working on scalability of a good hybrid meetings solution]

<jgraham> jgraham: [Principles][system effects]

<jholland> [jholland: technologies to reduce energy usage, measurement]

<mprorock> mprorock: [system effects] [measurements]

<gendler> gendler: [principles]

yoav: [energy reduction, measurement]

<ivan> ivan: [measurements]

<spectranaut_> spectranaut_: [horizontal review]

<tzviya> tzviya: [principles]

<AramZS> AramZS: [measurements][principles]

Tim Frick: co-chair of the sustyweb group. Wrote a charter. Want to know how we can collaborate

<jholland> [jholland: akamai has a sustainability commitment, if I can get access to a good standard for reporting usage I will push measurement within akamai and return feedback on problems with applying the metrics if possible]

<phila> phila: [measurement] (by which I mean certified measurements against defined methods)

<Judy> JudyB: [Hybrid] Good hybrid meeting techniques, and sharing awesome research showing the scalable carbon savings from hybrid

<phila> Sustainability CG

<gendler> https://www.w3.org/community/sustainability/

<npd> npdoty: [horizontal review][principles][w3c meetings and operations]

<jgraham> RRSAgent: make minutes

Summary of resolutions

  1. Sustainability CG work in public by default, as well as sustainability at W3C in general, and that we provide that request to the AB to make
Minutes manually created (not a transcript), formatted by scribe.perl version 192 (Tue Jun 28 16:55:30 2022 UTC).


Succeeded: s/in Q2/at the W3C AC meeting in Q2/

Succeeded: s/do to facilitate/do to facilitate... but possibly if one jumps ahead too early on that, before having something good to point to, might not land so well

Maybe present: Alex, DanD_, jholland, Nick, npdoty, spectranaut_, tzviya, yoav