Talk:Sustainability Initiative

From Meta, a Wikimedia project coordination wiki
(Redirected from Talk:Environmental impact)

Talk page archive[edit]

See Talk:Sustainability Initiative/Archives

Carbon offsetting: whether to use it and if so how?[edit]

Some of the discussion above mentions carbon offsets as a possible mitigation. It doesn't make sense to offset where emissions can be avoided by changing the energy source (i.e. with servers), but it might make sense to offset where emissions are unavoidable (i.e. with intercontinental travel).

I thought it might be helpful to create a separate section to try to create consensus around:

  • whether carbon offsets are worth using at all, and
    • if not, what to do instead; or
    • if so, whether the WMF should be encouraged to partner with a specific provider.

Zazpot (talk) 05:18, 9 November 2017 (UTC)[reply]

Hi Zazpot, I agree with you on this. --Gnom (talk) Let's make Wikipedia green! 11:37, 2 April 2018 (UTC)[reply]
Google just announced that its total purchase of renewable energy in 2017 exceeded the amount of electricity used by Google's worldwide operations. --Gnom (talk) Let's make Wikipedia green! 16:07, 7 April 2018 (UTC)[reply]

Should WMF offset air travel emissions? If not, then what; if so, then with whom?[edit]

  • Cautious no. Although some carbon offsetting schemes are likely to be genuinely effective, many have ended in failure, and numerous efforts to sort the genuine ones from the charlatans have not been kept up-to-date.[1][2] Because of the difficulty identifying effective offset schemes, I would prefer the WMF to instead:
  1. Adopt a decentralised Wikimania:
    • keep having a rotating host city in a different country each year; but
    • spend some of the event's funding on providing meetup spaces in cities around the world with existing concentrations of Wikimedians, for the duration of the event, and encourage participants to teleconference in from those meetups instead.
  2. Adopt a travel policy stating that in order to be reimbursed for travel expenses, WMF staff and volunteers travelling on WMF business must travel by bus, rail or carpool instead of car, taxi or air, as long as the cost of the former would be no more than twice the cost of the latter. Zazpot (talk) 05:18, 9 November 2017 (UTC)[reply]
Hi Zazpot, what do you think about this project? Microsoft Leads Movement to Offset Emissions With Internal Carbon Tax --Gnom (talk) Let's make Wikipedia green! 11:37, 2 April 2018 (UTC)[reply]
Hi Gnom, first thoughts: that article is an obvious PR coup for Micro$oft. I would encourage everybody to be sceptical of anything that Micro$oft says - especially if it is something that makes them look good - unless it has been externally audited by a trustworthy, independent team that has access to internal documents and that lacks conflicts of interest. It isn't clear that this has happened here. Even if the underlying facts are true, the New York Times should know better than to publish a piece that is so heavy on hagiography and so light on verifiability.
As for internal carbon pricing, I suppose the key quote is:

“When we started talking about carbon emissions not in metric tons, but in terms of dollar amount, the business people could understand it,” Ms. DiCaprio said. “We’re all speaking the same language now: What is the cost to my group?”

If an organisation's staff needs, as Microsoft's staff apparently does, environmental impact to be translated into dollar cost to business unit in order for the organisation to best reduce environmental impact, then yes, it makes sense to perform that translation. Otherwise, performing the translation is just unnecessary overhead. So, in the context of the WMF, the question would seem to be: is the WMF staffed with people who would most effectively reduce environmental impact only with the aid of such a translation? Answer that, and the answer to the question of whether or not to adopt an internal carbon-equivalent tax should be clear.
This is easier said than done, of course. If the WMF has not already piloted such a scheme, or anything like it, then they probably do not know, objectively, the answer to that question. So, perhaps they should try a small scale pilot - perhaps covering just two or three "business units", with another two or three as controls - to start forming an answer to that question. Zazpot (talk) 16:30, 2 April 2018 (UTC)[reply]

Sustainable investment policy adopted by Wikimedia Italia[edit]

See wmit:Politica di investimento finanziario. Apart from some local ethical financial instruments, in short the board has directed the president to (only) buy ETFs with an ESG score in Morningstar of 4/5 or more, or similar. Of course there are more sophisticated ways to do this, but WMIT only needs to park relatively small amounts of money for not so long periods in a sensible way. More information will follow. Thanks, Nemo 16:22, 16 March 2018 (UTC)[reply]

Wow, that's wonderful, Nemo! I'm currently looking into this topic as well, so I'm excited to hear more. --Gnom (talk) Let's make Wikipedia green! 18:29, 16 March 2018 (UTC)[reply]
For now I've translated the two most interesting criteria: wmit:Politica di investimento finanziario/en. Note that the ETF options available for our case (rather short-term investment) are very limited compared to a case like the Wikimedia Endowment's, and we probably won't actually buy any for several more months as it will take some time to implement the policy. --Nemo 18:47, 16 March 2018 (UTC)[reply]

Geocoordinates /Photos of protected plants and animals[edit]

In my opinion, an initative on Commons should be startet to sensitize users to environmental problems of geo-coordinates. Protected plants like Orchidaceae often are lifted by people. Also many protected plants are sensitive to soil compaction. It is no need to know where a protected bird has its nid. Y 've spoken about this problem with several people working for nature conservation authorities or cooperating with them. Most of them evaluate coordinates as an additional risk for protected plants and animals. --Belladonna2 (talk) 09:02, 25 June 2018 (UTC)[reply]

Feel free to do it but that's a different topic and has got nothing to do with this sustainability initiative. NNW (talk) 10:16, 25 June 2018 (UTC)[reply]
This behavior can seriously dammage species and also influence in a negative way wildlife conservation programs. Why should this not be a topic for a sustainability Initiative? --Belladonna2 (talk) 10:51, 25 June 2018 (UTC)[reply]
Is this one of the three main goals? No. There are many things that damage nature, coordinates in exif data may be one (but not necessarily like using coal as a source of energy), but it's also driving by car when you may take your bike or eating less meat. That all would help to save the planet but not everything can be goal of this initiative. NNW (talk) 11:12, 25 June 2018 (UTC)[reply]
Btw: addressee of this initiative are the foundation and the national chapters. They all do not upload photos of animals or plants with coordinates. You need to talk to the community. NNW (talk) 11:26, 25 June 2018 (UTC)[reply]
These coordinates will stay in web forever. The dammaging potential can not be reduced by an other behavior. This is the significant difference between your examples and my concern. I don`t understand why it`s not possible to discuss the topic. Sustainability means also wildlive conservation. Have you barred this goal from the initiative? And who is responsible for this? --Belladonna2 (talk) 11:30, 25 June 2018 (UTC)[reply]
I didn't write that it is impossible to discuss it. I just wrote that this is the wrong place. I am sure you are able to find the history of this page and the name of the user who created it. NNW (talk) 11:48, 25 June 2018 (UTC)[reply]
Hi, we touched on this topic at Wikimania 2015: "Non-copyright restrictions to free knowledge". --Gnom (talk) Let's make Wikipedia green! 20:43, 25 June 2018 (UTC)[reply]
I cannot see, that this topic has been discussed. I`ll show you now some articels dealing with this subjet.
Berichte der bayrischen botanischen Gesellschaft, Artikel Süddeutsche Zeitung, Landschaftsinformationssystem der Naturschutzverwaltung Rheinland-Pfalz
In my opinion the freedom of knowledge cannot ignore the situation of endangered species, which are threatended by the freedom of knowledge. In this regard, a discussion about approaches would be a really contribution to sustainability. --Belladonna2 (talk) 18:45, 1 July 2018 (UTC)[reply]
Yes, can you give a specific example concerning Wikipedia? --Gnom (talk) Let's make Wikipedia green! 22:01, 1 July 2018 (UTC)[reply]
On commons you'll find many photos taken of highly protected species with exact positioning. These dates are directly linked to Geohack. In which way these informations will be continued to use- for example by Google Earth- is incalculable. There is also a strong trend to postulate geo-coordinates in the context of competitions like Wiki loves Earth. Raising relevant user groups awareness of this topic would be a step in the right direction.--Belladonna2 (talk) 08:20, 2 July 2018 (UTC)[reply]
Thanks, can you point me to a few examples? --Gnom (talk) Let's make Wikipedia green! 09:24, 2 July 2018 (UTC)[reply]

Any updates?[edit]

Has anything major been done since Wikimedia 2015? Any chance for an update at the next metrics & activities meeting? Cheers, Daylen (talk) 20:45, 28 June 2018 (UTC)[reply]

Hi Daylen, thanks for asking! The information on the page is nearly up to date, with two exceptions:
  • On 19 June 2018, the Wikimedia Foundation published a new investment policy that names sustainability as a core objective. (Note: This policy is for the WMF's own investments, and not for the Wikimedia Endowment.)
  • The draft Wikimedia Foundation annual plan for 2018/2019 contains an Environmental Sustainability Program. The final version of the plan will probably be published tomorrow, 30 June 2018.
I will try to update the page soon. I'd be happy to give an update at the next Metrics and Activities Meeting, if requested! Kind regards, --Gnom (talk) Let's make Wikipedia green! 06:52, 29 June 2018 (UTC)[reply]
Done. --Gnom (talk) Let's make Wikipedia green! 05:04, 30 June 2018 (UTC)[reply]

Wiki Loves Vegan[edit]

Wiki Loves Vegan is a wiki community group advocating for vegan participation options at Wikimedia events.

Most events serving food at Wikimania for example do not have vegan options. When vegan options exist they often are not labeled. When social events serve food for some people but not others then that makes the event less accessible for broad participation. Vegan food is the most accessible food choice as every diet includes some vegan food.

There are many reasons why people choose veganism. Among them, en:Environmental vegetarianism to promote sustainable living is one reason.

How would the Sustainability Initiative feel about advocating that all major Wikimedia events have a clearly labeled vegan option at meals? Wiki Loves Vegan could coordinate discussion around this if the Sustainability Initiative can support the idea. Right now having vegan food is either haphazard or a luxury option. I would like for it to become a norm. Blue Rasberry (talk) 13:33, 25 September 2018 (UTC)[reply]

Hi User:Bluerasberry, making vegetarian or vegan food at events would obviously help. To my knowledge, this has already been tried and met with criticism from Latin American participants. --Gnom (talk) Let's make Wikipedia green! 17:09, 27 September 2018 (UTC)[reply]
@Gnom: Are you aware of any documentation on past experiments or requests? Do you have any recollection of where I could look to possibly find old discussions?
I think there is no need to make any fast decision but I wanted to put the idea out. Blue Rasberry (talk) 00:36, 28 September 2018 (UTC)[reply]
I think Wenke Storn (WMDE) and her team might know from the past Wikimedia Conferences. Kind regards, --Gnom (talk) Let's make Wikipedia green! 08:12, 28 September 2018 (UTC)[reply]
Hi Bluerasberry, hi Gnom, Unfortunately we don’t have any documented discussion or feedback regarding vegan food at the Wikimedia Conference or other bigger events we organize. In our event registration forms, we always ask for food preferences and provide vegetarian and vegan as options to choose from. Across all events, the percentage varies a lot – approx. 15-40% choose vegetarian, 3-8% choose vegan. It depends on the type of event and the respective participant/target groups. Wenke Storn (WMDE)

After http://www.ipcc.ch/report/sr15/ it's really hard to say that it's responsible to continue business as usual with meat at events. Cf. https://www.bbc.co.uk/news/science-environment-45814659 Without having any hasty decision, Wikimedia Foundation could set a target for the events it funds, such as "reduce meat consumption by 90 %", and then leave the specific implementation to the organisers. --Nemo 10:55, 23 October 2018 (UTC)[reply]

Sustainable ETFs[edit]

Today on FT: BlackRock stakes claim on sustainable investing: Fink forecasts such ETF assets will rise from $25bn to over $400bn in a decade. Which reminds me I still have to share an earlier paper article on FT. --Nemo 10:46, 23 October 2018 (UTC)[reply]

See also https://www.unpri.org/annual-report-2018/how-we-work/the-pri-in-numbers Nemo 12:50, 11 January 2019 (UTC)[reply]

Tetra Pak[edit]

Worth keeping a PR eye on given donations to WMF: https://www.theguardian.com/environment/2018/dec/09/billions-discarded-tetra-pak-cover-vietnams-beaches-towns --Nemo 15:41, 9 December 2018 (UTC)[reply]

Rankabrand[edit]

Hi @Aubrey, Gnom, and Nemo bis:,

I just discovered the website rankabrand.org who has a Wikipedia sustainability score report which is very bad. We could use this indicator as an help as some points could probably be easily improved (PUE, footprint, etc.). Anyway, I think it worth at least looking more into it (and maybe benchmarking with other similar websites).

Cdlt, VIGNERON * discut. 11:22, 13 December 2018 (UTC)[reply]

Thanks, VIGNERON! How "credible" is this website? --Gnom (talk) Let's make Wikipedia green! 14:06, 13 December 2018 (UTC)[reply]
@Gnom: that's a good question. I just checked the basics : Alexa rank (low but not super low, the site has quite a visibility) and the Who we are page (which seems fine), and they don't confuse Wikipedia and Wikimedia, but there is definitely room for improvement ! Cheers, VIGNERON * discut. 14:25, 13 December 2018 (UTC)[reply]
Considering that they mix "Wikipedia" and "Wikimedia Foundation" liberally, and their only source appears to be this page (but they didn't even bother to use a permanent link), I'd say it's a totally unreliable website. (They even managed to pick the pre-2010 logo.)
Their score is also questionable to the point of being useless: it's just a checkbox-ticking exercise where it only matters whether one has said or done something about a certain topic, but not what they actually achieve, so for instance they give the same score whether one has reduced emissions by 55 % or 15 % ([3] [4] [5]). Judging from some Google News searches, their reach is nearly non-existent and people continue to prefer more reliable studies by entities like Greenpeace. Nemo 20:21, 17 December 2018 (UTC)[reply]
@Gnom and Nemo bis: it's maybe not the most important website but we should care nonetheless. I will try to contact them, at least to put the correct logo. Do we have something that can help check some box (I don't think so, but I don't know everyting ;) ). Nemo: where do they mix "Wikipedia" and "Wikimedia Foundation" (so I can ask them to correct it). Cheers, VIGNERON * discut. 16:44, 7 February 2019 (UTC)[reply]
Hi, I've just been contacted by someone at Rankabrand and I am trying to call them next week. --Gnom (talk) Let's make Wikipedia green! 14:19, 10 February 2019 (UTC)[reply]
@Gnom: A few of us were chatting about this website the other day - were you able to speak to anyone there? I'd love to hear how the conversation went, if so. Samwalton9 (WMF) (talk) 15:00, 19 April 2019 (UTC)[reply]
Hi Samwalton9 (WMF), I exchanged a couple of messages with them and we agreed to speak over the phone, but I didn't manage to reach them. I will try again. --Gnom (talk) Let's make Wikipedia green! 14:38, 20 April 2019 (UTC)[reply]
Hi Samwalton9 (WMF), I spoke with Mario Dziamski of Rankabrand today and we agreed to stay in touch to keep Wikipedia's profile on their website up to date. --Gnom (talk) Let's make Wikipedia green! 11:40, 15 May 2019 (UTC)[reply]

Foot prints of servers and clients[edit]

To me as a layman, it seems evident that the client computers, which are used world wide in order to read and edit Wikimedia projects, which most of the time are some language versions of Wikipedia, that these computers are using far more electric power and non-sustainable ressources than the Wikimedia servers. In fact, 222 kilo Watt, which is like 2220 household light bulbs burning all the time, seems relatively little, compared to all the computers and infracstructure needed for transmission of the information internet signal, and for running all of the client computers. So my guess is, in order to have any positive environmental impact, it would be far more effective to focus on the clients side rather than the server side.

E.g. if you connect to the internet with any computer running on sustainable energy, there should be some encrypted and reliable certificate, which can be read, confirmed and rewarded by other computers. E.g. if you connect to the Wikimedia server with a computer running on certified sustainable energy, then Wikimedia would reward you, like with a small bonus paid out to your bank account in US$ or any other viable currency, perhaps like 1 cent per hour. And it won't bancrupt Wikimedia, if Wikimedia decides to run the servers on sustainable energy too, since then the client computers will acknowledge the certificate provided by Wikimedia, and pay the same amount of money back as a reward.

This way, Wikimedia could actually encourage the use of more sustainable power, or not? Of course the fundamental problem remains, that the production of sustainable electrical power is far more expensive than non-sustainable power, and probably Wikimedia isn't able to solve this problem on it's own, Rosenkohl (talk) 13:04, 4 February 2019 (UTC)[reply]

Another idea: make pure text and pictures with no layout the standard for Wikipedia readers and new accounts.

At the moment, "vector" skin is the standard layout, and if you register at Wikipedia, you can change to several other skins. I guess with js.-skripts you can even turn out most the layout. However, most of the average readers will use the standard layout. But transfering the layout from server to user computer consumes time and energy, since it takes place very often every day, each time a reader or editor is opening a new page or article. So I wonder: Can we estimate, how much electric power could be saved on the server, transmission network and client side by returning to simple pure text articles as the standard, without a skin, and pictures only as far as they belong to the encyclopaedic article?

After all, the text and pictures is all you actually need as reader of an article. And, if people remember and find out how smooth, fast and energy efficient pure text works on Wikipedia, they may start to prefer to use it on other sites of the internet too, Rosenkohl (talk) 23:10, 21 February 2019 (UTC)[reply]

It's trivial to disable images on your browser. What really consumes electricity is all the CPU required by JavaScript, especially as the client's performance per watt is likely to be dismal compared to the servers' and routers'. Nemo 21:40, 24 February 2019 (UTC)[reply]

Wikipedia events -> Green events?[edit]

I would like to follow the previous discussion on Wikipedia events - at least in different European countries there are now standards how to make conferences and similar events environmental friendly. I know, that there is for example accreditation for Green meetings in Austria, there are different guides also on web, for example this one (Btw. Green events page has to be improved much). The Green meeting is not only about vegetarian/vegan food, but also about reduction of waste (plastic phase-out), about information (very important because of influence of WP), transport etc. etc. I have never been in any Wikipedia event, but I work in the field of environment and also meetings which I participate are changing quite lot. Btw. vegetarian starts to be standard. :-) Jirka Dl (talk) 05:46, 5 February 2019 (UTC)[reply]

Great, thanks for sharing Jirka Dl!
I'm pinging Axel Pettersson (WMSE) and Eric Luth (WMSE) since in this tweet Axel said that minimizing waste is « for sure something we're looking into. ». I hope this can help.
Cdlt, VIGNERON * discut. 16:50, 7 February 2019 (UTC)[reply]
The usual suspects are active on reducing the carbon impact of events too: https://climatecare.org/carbon-offsetting/ . The biggest component is usually the cumulative distance flown and the kind of food eaten. Making all food vegan and offsetting all the flights is a decent mix of concrete action and self-incentive for transport-efficient venues. Nemo 13:45, 9 February 2019 (UTC)[reply]
Yes, vegan food could be perfect, but I think, that vegetarian local food could be better acceptable for most of participants. Not wasting food is also important. There are also other "small" items which can make change - not using single use plastics, recycling waste, give paricipants proper information on public transport etc etc. As an example of the "green meeting" you can find green conference information about our conference in Vienna in 2016. Jirka Dl (talk) 20:41, 10 February 2019 (UTC)[reply]
Sure. The trade-off also depends on the local availability of tasty options. Here in Helsinki all the events I attended recently were vegan-only and there was no waste. Nemo 21:42, 24 February 2019 (UTC)[reply]

Contacts[edit]

I really need to start a list of experts from outside the Wikimedia movement I have spoken with about this project... Here's a start:

  • David Pomerantz, Greenpeace USA
  • I participated in the ClimateAction.tech sustainability accelerator (cohort 1/2018), where I spoke with (among others)
    • Chris Adams,
    • Sunya Ojure, Director of Sustainability at Salesforce
    • Chelsea Mozen, Sustainability Lead at Etsy
    • Andrew McWilliams, Sara Diaz, and others, consultants at Thoughtworks (technology consultancy)
  • Andreas Feiner, Head of ESG Research and Advisory, Arabesque Partners and Thomas Schafbauer, Director, Arabesque Asset Management (sustainable investment analysts)
  • Matthias Schmuderer, Enerbasics (sustainability consultancy)
  • Alexander Rossner, Zukunftswerk (renewable energy consultancy)
  • René Post and Chris Adams, The Green Web Foundation (NGO promoting green hosting)
  • Christian Kroll and Wolfgang Oels, Ecosia (the search engine that plants trees)
  • Mario Dziamski, Rankabrand (sustainability ranking website)

--Gnom (talk) Let's make Wikipedia green! 11:37, 15 May 2019 (UTC)[reply]

Wikimania 2018 in Cap Town[edit]

Was it a good idea of Wikimedia-Foundation to keep Cap Town for the Wikimania 2018? The Cap Area had several years without rainfalls. And the people there had in 2017/2018 a lot of Problems to get enough Water. And i wrote a letter to WMF to look for another Town for that Wikimania 2018. They wrote back "we keep it" the South African Tourist Office wrote to us "all Tourist, so the Wikimania-Guest too, will get enough water. The SA Department for Tourism give all Hotels enough water, it doesnt matter the all peaple in the Cap Areal get nearly no water. The tourist will get enough water". Is that polically correct? Is that how Wikimania should act? And yes the the conference had no problem with water. What a signal to the world! The Natur in South Africa has a big Problem with the climate change. The Capenis with its big amount of endemic species and the problem of Invasiv Species, and WMF was totally blind for this acute risk of Climate Change :-( --BotBln (talk) 08:25, 26 April 2019 (UTC)[reply]

Updated stats?[edit]

Can we get some updated information about WMF’s sustainability efforts?

  • Greenpeace USA’s “Click Clean Scorecard”: 2015 grading
  • Renewable energy for servers: August 2017 stats
  • Power source graphic: May 2016 data
  • Green energy roadmap: submitted February 2017
  • Remote participation: 2018 details described, 2019 details not yet published
  • Wikimedia Endowment: “investigating” where the Endowment should be moved to “in 2019”
  • Slide deck image: Wikimania 2017

Perhaps @EBjune (WMF), Rfarrand (WMF), and Lgruwell-WMF: could add updating their sections to their to-do lists and {{ping}} here? — OwenBlacker (Talk) 00:18, 30 June 2019 (UTC)[reply]

Some suggested edits as of June 2019[edit]

A few things should be amended given the change in staffing and resources:

  1. "Ellie Young, Events Manager, Wikimedia Foundation" - She is no longer with the Foundation. You should likely contact Joël Letang, Events Team Manager to see if he is the right person to place in here. [6]
Done --Gnom (talk) Let's make Wikipedia green! 12:01, 30 June 2019 (UTC)[reply]
  1. "The Wikimedia Foundation should offer video streaming for all sessions at Wikimania and other Wikimedia events" - As the main person who has spearheaded video recording in the last few Wikimania events, I would suggest this be more generic to say, "video recording of all sessions at Wikimania and other Wikimedia events" since there should be no needed requirement it be "live" or "streamed." This is a fine distinction but it does have implications in that streaming is often a higher technical hurdle that is not necessarily a requirement. As long as the session is memorialized and made available later, that is what is most important.
Done Couldn't agree more. --Gnom (talk) Let's make Wikipedia green! 12:01, 30 June 2019 (UTC)[reply]
  1. "Remote participation at Wikimania" - Rather than just making remote participation the only focus of this section, we could also talk about mitigating the impact of Wikimania as an in-person gathering from the very start of the bidding/selection process, in order to select better hub locations that minimize long-distance travel. A previous comment here mentioned the impact of flying a large percentage of attendees to locations outside the usual travel patterns of Wikimanias past, such as Cape Town, and this is a legitimate question. This should also be balanced against the outreach role of Wikimania being closer to communities that further the mission and the goal of knowledge equity for underserved languages and cultures. It's not an easy balance to strike, but it seems to be worth mentioning in the Wikimania section.
That's true. In the spirit of keeping the number of individual demands of the Sustainability Initiative low, I'd propose leaving this out for now, however. --Gnom (talk) Let's make Wikipedia green! 12:01, 30 June 2019 (UTC)[reply]
  1. Sustainability_Initiative/team - "Juliet Barbara" is no longer with the Foundation, so perhaps this should be changed?
Juliet specifically requested to be part of the team after she left the WMF. However, I have to say that the team never had any activity even though I tried to start a few conversations via email. --Gnom (talk) Let's make Wikipedia green! 12:01, 30 June 2019 (UTC)[reply]

Thanks! -- Fuzheado (talk) 02:13, 30 June 2019 (UTC)[reply]

Hi Fuzheado, Thank you for your comments! See my replies above. --Gnom (talk) Let's make Wikipedia green! 12:01, 30 June 2019 (UTC)[reply]

shutdownforclimate[edit]

Should we join on global climate strike on 20. September 2019? Template: http://shutdownforclimate.net Different levels seems possible for me for Wikipedia: complete blackout strike of Wikipedia, info banner on Wikipedia or individual infos on user-pages. --Kolossos (talk) 15:06, 22 August 2019 (UTC)[reply]

Renewable energy vs sustainable/low carbon energy[edit]

Hi everyone. Regarding "renewable energy for the Wikimedia servers", I imagine you're thinking of wind, solar, hydroelectricity, and maybe geothermal or tidal energy. That stuff is great. We should absolutely try to power the servers on as much of it as possible. However, the category of "renewable energy" includes biofuels, which are sometimes good and sometimes very, very bad. The English Wikipedia has an article summarizing the issues. I suggest that the proposal clarify that the intention is to power the servers with low-carbon electricity sources, or perhaps "green energy", to avoid unintended consequences. Cheers, Clayoquot (talk) 18:43, 16 September 2019 (UTC)[reply]

Sure, 'green energy' is good for me. Would you like to make the change? --Gnom (talk) Let's make Wikipedia green! 08:14, 17 September 2019 (UTC)[reply]
Yes please! Clayoquot (talk) 05:56, 18 September 2019 (UTC)[reply]

WMF sustainability report published[edit]

Wikimedia Foundation Sustainability Assessment and Carbon Footprint

The Wikimedia Foundation just published its first-ever sustainability report. This is a big milestone. I will update the page over the weekend. --Gnom (talk) Let's make Wikipedia green! 14:25, 19 September 2019 (UTC)[reply]

Blog post[edit]

Wikimedia Sverige just published a blog post that covers the initiative (in both English and Swedish). Ainali (talk) 19:20, 3 December 2019 (UTC)[reply]

Green Web Check by Green Web Foundation[edit]

Update 2020: According to the Green Web Foundation, Wikipedia.org and Wikimedia.org both are NOT hosted green. --Martina Nolte (talk) 22:58, 20 January 2020 (UTC)[reply]

Optimize Code[edit]

Hi, how much potential in saving energy could be unleashed by optimizing code? Both code run at the servers but mustly code send to the clients and run there. Probably not much in HTML, but maybe in JS? JS can cost lots of CPU time and energy and I believe our code allows for optimization there. Is that an issue for the Sustainability Initiative? And who should be asked to look at this? --h-stt !? 15:00, 21 January 2020 (UTC)[reply]

Hi h-stt, as far as I understand, there are already considerable efforts under way in this regard. See here (from 2019, in German) and here (from 2014, in English). --Gnom (talk) Let's make Wikipedia green! 15:46, 21 January 2020 (UTC)[reply]
THX. Looks good to me. I guess there still is potential for further optimization but obviously the relevant people are already at work. Let's keep up the spirit. --h-stt !? 15:51, 21 January 2020 (UTC)[reply]
Does anyone know if the WP-Logo has been reduced? In the German newspaper FAZ was suggested end of 2019 that the logo could be reduced without loosing information (in German: "Würde die Wikipedia nur ihr Logo um die Hälfte komprimieren, was verlustfrei möglich wäre, könnte man dafür ein paarmal zwischen Berlin und London hin- und herfliegen." FAZ, 31.12.2019). Thank you --Hadi (talk) 15:52, 10 February 2020 (UTC)[reply]

Tracked in Phabricator:
Task T245362

Hi Hadi, thank you for pointing this out – I just filed a bug report. --Gnom (talk) Let's make Wikipedia green! 16:01, 16 February 2020 (UTC)[reply]
All these claims about energy consumption (or saving) caused by increasing (or decreasing) the bandwidth usage, or even the CPU time, are generally bogus. I've not yet seen any serious study on the topic. For instance, it's pointless to reduce CPU usage if the processor doesn't actually reduce consumption; reducing consumption in one place may increase consumption elsewhere; variable energy consumption may be the minority of energy consumption. Even a rather ok study did not measure the actual consumption but just what the processor states; many just consider CPU cycles; Rong 2016 shows that servers are directly responsible for a minority of the consumption even if you limit your view to the datacentre.
If you look at the Global power usage graph, you'll notice that it's almost completely flat, although traffic varies quite a bit. On the individual client device at home things are of course different, but it's very hard to predict, let alone directly measure, how energy consumption will be affected by optimisation of some HTML or JavaScript or other.
Finally, it's really misleading to focus on a specific application at a time. For instance, if we reduced our servers' energy consumption by 50 % and made the sites slower by 10 %, resulting in 10 % less page views and 5 % more Wikipedia users who go watch a movie on some DRM-heavy video streaming website projected on a 50" screen instead of spending hours in some text-only rabbit hole, do we want to bet how much the global energy consumption would go up? We should really focus on making our websites and software more responsive and fast to load even on low-power devices, so that people never feel the need to "upgrade" to hungrier devices. That's the only reasonable mode of operation.
In conclusion, a personal suggestion: if you care about the energy consumption for computer usage, tell everyone to install uMatrix/NoScript/others to make bloated websites lighter, as well as GNU/Linux to extend the life of PCs. There's a relatively short guide at privacytools.io which is mostly ok. Nemo 18:00, 17 February 2020 (UTC)[reply]
Interesting points Nemo and I fully agree that Wikimedia should not be only a cluster of websites that optimize content and software, but also as one of the veryVeryVERY few non-commercial websites at the top - it should influence the culture of computing from consumerism and abundance into a frugality with both others in present and shared future. Zblace (talk) 08:38, 13 December 2020 (UTC)[reply]

Equinix data center[edit]

@Gnom: Equinix now claims on that their US data centers are powered by 100% renewable energy, which presumably would include the one that the WMF uses in Ashburn (source - https://www.equinix.com/data-centers/design/green-data-centers/). It is worth noting that Equinix's website is ambiguous to whether the data centers in Virginia are powered directly by renewable electricity or if the consumption is "offset" through RECs. Also, are you aware of any updates from the Foundation regarding this initiative in 2020, aside from the cancellation of Wikimania this year? Thank you. Daylen (talk) 09:06, 21 June 2020 (UTC)[reply]

Hi Daylen, thank you for this pointer! I just had a quick look and it says that they have in fact reached "100%" renewable energy in the United States, which "comes from Green-e RECs from wind in the U.S.". So yes, not the progress we would like to see here. Please let me know if you have a different understanding of the information available on the Equinix website.
To answer your question, there are unfortunately no updates, although I have heard that the Wikimedia Foundation is still planning to follow through on their commitment from last year to allow community participation in their existing 'Sustainability Consortium'. --Gnom (talk) 21:30, 22 June 2020 (UTC)[reply]

Carbon footprint of 1.2 kilotonnes[edit]

"carbon footprint of 1.2 kilotonnes" I guess this should be megatonnes.--2003:C0:3F34:98B5:F8B7:EBEF:DFF4:30C5 10:48, 9 August 2020 (UTC)[reply]

Hi, thank you for your comment. Did you check the footnote provided for that figure? —-Gnom (talk) 12:25, 9 August 2020 (UTC)[reply]
Yes, and there it is (1,142.00+14.92+7.25) T CO2-eq/activity. And as it is an american publication I assume the "," to seperate powers of 10³ and the "." to seperate decimals lower than 1.--2003:C0:3F34:98B5:F8B7:EBEF:DFF4:30C5 12:54, 9 August 2020 (UTC)[reply]
Ok, I see that the value is correct. But if I want to adapt the german translation it does not work, I got the message:
Publishing the translation failed: This action has been automatically identified as harmful, and therefore disallowed. If you believe your action was constructive, please inform an administrator of what you were trying to do. A brief description of the abuse rule which your action matched is: Antispam
Thank you for the pointer, I just corrected the mistranslation. To avoid the error message that you received, I recommend creating an account. --Gnom (talk) 21:10, 9 August 2020 (UTC)[reply]

1.2 kilotonnes equal annual carbon footprint of 80 persons[edit]

Hello User:Gnom, "powering the Wikimedia servers [...] resulting in a carbon footprint of 1.2 kilotonnes, which equals the annual carbon footprint of 80 persons. Regarding your revert you see this differently? --Alex42 (talk) 22:23, 24 April 2021 (UTC)[reply]

Hi Alex42, one question is whether we would want to base this one figures for U.S. residents. Another is whether readers would (wrongly) think that a "person's" carbon footprint in this sense only contains their direct emissions rather than their personal share in the emissions of the entire country. --Gnom (talk) 23:24, 24 April 2021 (UTC)[reply]
The conversion from kWh to CO2 was done using US figures because the servers are located there. For further conversions it makes sense to stay consistent. In every case these CO2 emissions are absolutely negligible. --Alex42 (talk) 07:35, 25 April 2021 (UTC)[reply]
You are correct that these figures are relatively low, but they do contribute to the current climate crisis. And just like everyone else, we are responsible to do our part – and, in my opinion, also to set an example for other websites that may be consuming much more energy than us.
On your other point, as a global movement, I don't think we should make a comparison to U.S. figures. --Gnom (talk) 09:31, 1 May 2021 (UTC)[reply]

2019 WMF carbon footprint report and sustainability impact statement published[edit]

The Wikimedia Foundation has now published its "carbon footprint report" and "sustainability impact statement" for 2019 alongside with a blog post titled How the Wikimedia Foundation is continuing efforts to reduce its carbon impact:

I have commented on Talk:Sustainability. Please join the discussion! --Gnom (talk) 09:45, 27 August 2020 (UTC)[reply]

I just updated the page a bit. A discussion is going on here: Talk:Sustainability#Sustainability_Consortium_Sharing_Lessons_Learned. --Gnom (talk) 20:42, 22 November 2020 (UTC)[reply]

18 September 2020: community discussion and editathon[edit]

Hi, check out the news for the sustainability community discussion and climate justice editathon on 18 September. --Gnom (talk) 22:13, 14 September 2020 (UTC)[reply]

Carbon Emissions of the Web’s Most Popular Sites by Uswitch[edit]

Hi, you might be interested by this article that says that Wikipedia.org is one of the "cleanest" website evaluated during the study. Congratulations :) Pamputt (talk) 12:24, 17 March 2021 (UTC)[reply]

Podcast about the Sustainability Initiative[edit]

People following this page might be interested in this interview I made with User:Gnom about the initiative.

(It's also available in a more common podcast format.) Ainali talkcontributions 12:46, 27 March 2021 (UTC)[reply]

KDE positions[edit]

2 open positions at KDE e.V. for environmental sustainability project Nemo 22:13, 5 April 2021 (UTC)[reply]

New data centre in Marseille, France[edit]

I just heard that the WMF is opening a (small) new data centre location in Marseille, France. The website of the colocation provider states that "100% of power consumption [is] produced from renewable sources". Yay! (On the other hand, there seems to be no movement concerning our big data centre sites in Virginia and in Texas...) --Gnom (talk) 12:23, 17 May 2021 (UTC)[reply]

Fun, feels like a throwback to the old times with the lone caching server in France. Nemo 19:42, 17 May 2021 (UTC) :)[reply]

Proposal for reworking the lead[edit]

Currently, the lead reads as follows (as at Aug 2021):

The sustainability initiative aims at reducing the environmental impact of the Wikimedia projects. The reason for this is simple and can be found right in Wikipedia: Anthropogenic climate change is a reality and it is our shared responsibility to prevent it, while the Wikimedia projects have a significant carbon footprint, especially through power consumption and air travel.

Per this discussion, I would recommend two enhancements to this page: Firstly that there is clarification that this is a community-led initiative intended to support and encourage efforts by WMF to act with greater environmental responsibility. And, secondly, that greater focus is made to generate greater interest, support and sign-up from individual Wikimedia editors. I would offer this expanded alternative for consideration:

The sustainability initiative aims at reducing the environmental impact of all Wikimedia projects. It is a community-led stratagem, intended to influence and encourage the Wikimedia Foundation into greater environmental responsibility in all its activities and by its users.
The rationale for the initiative is simple and can be found on Wikipedia itself. Put simply: Anthropogenic climate change is a reality and it is our shared responsibility to prevent it. The Wikimedia projects have a significant carbon footprint, especially through power consumption and kerosene-fuelled travel, whilst their users and editors have the ability to take small but collectively significant individual actions to reduce their carbon emissions and overall environmental impact, both on- and off-line.

My own view in making this suggestion is that, not only should the community be supporting, challenging and encouraging the WMF to be sustainable, but we should also, as a group of individual Wikimedians, be prepared to show that we will act as sustainably as possible in our own lives, both online and offline. The sustainability initiative seems the best way to do this, and I really don't feel this would be going too far, as it makes sense not only to ask our parent organisation to take steps to reduce its environmental impact, but also to do the same ourselves. I hope others would agree? Nick Moyes (talk) 10:10, 31 August 2021 (UTC)[reply]

Thank you, Nick Moyes! To start, I made a small change in the first sentence; let's see what others say. Next, we are in the middle of a great debate, see Individual action on climate change. --Gnom (talk) 16:56, 31 August 2021 (UTC)[reply]

Curious if carbon-capture has been evaluated?[edit]

I found this https://climeworks.com/net-zero-solutions-for-businesses and was wondering if carbon-capture has been evaluated at all? Like, if WMF can measure the total carbon output of servers and computers and offices and flights etc. then WMF could in theory buy 2x that from an outfit like this just to make sure it compensates for all that pollution? Victorgrigas (talk) 03:17, 4 December 2021 (UTC)[reply]

The problem is that it does not work. --Gnom (talk) 23:35, 4 December 2021 (UTC)[reply]
My mistake - I meant to say Direct Air Capture, which (I believe to be) different than carbon-capture. Victorgrigas (talk) 15:39, 7 December 2021 (UTC)[reply]
It requires a vast amount of energy and money to keep 'pie in the sky' aloft, and I would have thought tree planting would be a far more effective route for capturing and offsetting CO2 production. (I believe it worked quite successfully during the Carboniferous Period, just as it should today.) It would be wonderful, however, to see WMF move to using servers that rely on much higher proportion of renewable sources of fuel to deliver our content. Nick Moyes (talk) 16:04, 7 December 2021 (UTC)[reply]
I don't disagree, I have suggested before (sorry can't find the link now) that WMF buy a forest in some strategic way for this purpose. What I'm getting at is if this direct air capture method has been evaluated or not? Victorgrigas (talk) 14:34, 8 December 2021 (UTC)[reply]
@Victorgrigas: I very much doubt it, nor do I think for one moment that it's the role of WMF to investigate unproven technologies when far simpler solutions already exist that are known to work. The push that WMF should make would be better directed at their server suppliers who consume the most energy, or WMF could make greater effort to switch to alternative providers, and reduce its global C footprint. As the lead in the page you refer to says: "DAC can act as a carbon dioxide removal tool, although as of 2021 it is not profitable because the cost per tonne of carbon dioxide is several times the carbon price." I think that's all we really need to say on that. Except that I don't see that much critique of the process on the page at Direct air capture. Nick Moyes (talk) 20:12, 8 December 2021 (UTC)[reply]
Respectfully, I don't think that just because something is expensive it should be outright ignored. By my napkin math, WMF has roughly a carbon footprint of 2500 tons, and DAC costs roughly $250-$600 per ton currently (and per your point reforestation costs $50 per ton)[1], so that's between $625,000 and $1,500,000 for DAC and $125,000 for reforestation. Now, how fast can reforestation work vs. DAC? Will funding DAC now create cheaper DAC later? And why would it have to be one (reforestation) or the other (DAC)? In my thinking, if WMF were to actively implement DAC or reforestation or any other method, it could be used in fundraising efforts that could in theory pay for itself entirely. Suppose there was a wealthy potential donor to WMF who would be interested in DAC more than reforestation? I would argue that all of these regardless of cost could be a good. But per your point above about servers and their fuel source, I fully agree that pollution is occurring in the first place, and that polluting fuel source should be changed. Victorgrigas (talk) 01:08, 9 December 2021 (UTC)[reply]
I'm sorry, but it's a job for scientists and economists to investigate theoretical benefits of, as yet, unproven technology. Meanwhile, WMF should simply focus on actions which immediately reduce its huge carbon footprint. Diversionary tactics help no-one. Nick Moyes (talk) 01:40, 9 December 2021 (UTC)[reply]
Hear hear. Even if carbon capture can work as a solution, wouldn't it be much better to focus finances on efforts to lower emissions at the source? EpicPupper (talk) 04:39, 23 November 2022 (UTC)[reply]

DAC for the sake of DAC has approximately no role in a path to carbon neutrality. However, some models predict that the DAC necessary for production of e-fuels and e-chemicals might end up offsetting the emissions from things which are more difficult to electrify. See some charts.

It's not up to Wikimedia to fund the chemicals industry so there isn't much we can do here (apart from supporting open science so that innovation happens faster and more equitably). Nemo 15:37, 12 February 2023 (UTC)[reply]

Combatting Software-Driven Environmental Harm With Free Software[edit]

Some useful references from this KDE presentation at FOSDEM:

--Nemo 15:34, 12 February 2023 (UTC)[reply]

Anything new on the data centers?[edit]

How is it going with regard to the big centers? Has anyone seen any kind of communication since 2020-11? --h-stt !? 19:27, 3 July 2023 (UTC)[reply]

Following a discussion on a Movement Strategy sub-page, ASarabadani (WMF), has kindly picked up the job of providing an update. --Gnom (talk) 10:55, 2 November 2023 (UTC)[reply]
I have prepared a table to collect the relevant information which we can fill out (ideally also indicating a source for each figure) to prepare for the update. --Gnom (talk) 11:35, 2 November 2023 (UTC)[reply]
server name
and location
data centre
provider
energy consumption
2022[1]
energy sources[2]
2022
comments
eqiad
Virginia, USA
Equinix 2,295,476 kWh 9.2% renewable
(+ 38.8% nuclear)
(2021)[3]
provider states that it procures RECs for non-renewable energy share
(no use of renewable energy beyond general grid)[4];
data for the general Virginia electricity grid
codfw
Texas, USA
CyrusOne 1,123,770 kWh 85% renewable
(2022/2023)[5]
provider states that it does not rely on RECs[6];
"renewable power sources will supply approximately [...] 85% of CyrusOne’s [electricity] in North Texas"[5]
eqdfw
Texas, USA
Equinix 2,141 kWh 25.7% renewable
(+ 9.5% nuclear)
(2021)[3]
provider states that it procures RECs for non-renewable energy share
(no use of renewable energy beyond general grid)[4];
data for the general Texas electricity grid
eqord
Illinois, USA
Equinix 1,926 kWh 7.1% renewable
(+ 28.5% nuclear)
(2021)[3]
provider states that it procures RECs for non-renewable energy share
(no use of renewable energy beyond general grid)[4];
data for the general Illinois (RFCW) electricity grid
ulsfo
California, USA
United Layer 69,006 kWh 39.6% renewable
(+ 8.1% nuclear)
(2021)[3]
no statement from provider available;
data for the general California electricity grid
esams
Netherlands
EvoSwitch 91,140 kWh 100% renewable[7] provider states that it procures 100% renewable energy

for this data centre site

knams
Netherlands
Interxion 2,373 kWh 100% renewable[8] provider states that it procures 100% renewable energy

for this data centre site

eqsin
Singapore
Equinix 62,472 kWh 2.9% renewable[9] data for the general Singapore electricity grid
drmrs
France
Digital Realty 48,490 kWh 100% renewable[10] provider states that it procures 100% renewable energy

for this data centre site

total 3.696.794 kWh 36.20% renewable
(+ 24,26% nuclear)

Sources

  1. Wikimedia Foundation Environmental Sustainability Report 2022 (PDF). p. 22. 
  2. In this table, the term 'renewable' excludes nuclear energy.
  3. a b c d eGRID Summary Tables 2021 (PDF). United States Environmental Protection Agency. 
  4. a b c "Map of Initiatives". Equinix Corporate Sustainability Program. 
  5. a b "CyrusOne Adds Solar and Wind Energy in North Texas". CyrusOne. 2022-05-03. 
  6. CyrusOne: 2023 Sustainability Report (PDF). p. 47. 
  7. "Sustainability". Evoswitch. 
  8. Interxion Amsterdam Data Centre Campus Fact Sheet (PDF). p. 5. 
  9. Andres, Gabrielle (2023-02-07). "CNA Explains: Singapore's energy sources and the future of its electricity supply". Channel News Asia. 
  10. "Marseille MRS1". Digital Realty.