I wrote an article about the need for low-carbon and and the path towards zero-carbon computing.

In short, we need to dramatically limit the growth in emissions from computing, or by 2040 emissions from computing alone will be close to half the emissions level acceptable to keep global warming below 2°C.

And it is possible to do this.

@wim_v12e @stragu
Personally I started to adapt my computer habits to the climatecrisis by remaining offline lots of times.

Also sharing content offline with others, instead of everyone downloading the same over and over again.
I also don't listen to music in the browser, except one time when I download it initially.
And using books instead of watching the newest netflix series.

@wim_v12e what sort of things can the user do to reduce their computational footprint?

Aside from turning laptops/consoles, etc., off (not on standby), what else can the average person do?

Is this where programs that 'do one thing and do it well' philosophy come into play?

@athairbirb What you can do mainly is not replace your devices for as long as you can. That is the key factor.

@athairbirb As I see it, that is mostly a software problem: currently it's the software upgrade cycle that forces us to replace devices. So from a CS perspective we need to make sure that we can keep on upgrading the software but on the same devices.
Of course that also means devices must be repairable.

@wim_v12e We were discussing at a free software foundation meeting if re-using older computers and installing linux to give computer extended life is better for the environment. I know newer computers are more energy efficient would be good to perhaps get your input on this

@zleap It is almost always the case that the emissions caused by producing the newer, more energy efficient computer outweigh the energy gain due to its improved efficiency.
For a typical laptop you'd have to use it more than 10 years to break even.
So using the old computers for as long as possible is the best thing you can do.

@zleap See for more details on how long you have to use a device from an emissions perspective.

@wim_v12e Thanks - looks like a good read and great to see some device stats in there. Totally agree, we need to bake longer lifetimes and repair into our consciousness.

You might also like this piece by viznut and my slightly random newsletter at

@wim_v12e I'm really fascinated by how we tell new stories to change our behaviour, and how we disrupt that 'new' idea that we *need* to constantly upgrade or chase novelty.

@scribe Me too. I'd love to work with sociologists and economists on this: we need nudging but also economic incentives, both for producers and consumers.


I mean, they sent the Apollo on the moon with the calculation capacity of a Commodore 64, I think we can fix our needs in some more reasonable hardware

I use a refurbished Thinkpad myslef

@AbbieNormal Quite so. Even if we could simply stick with our current HW for the next decade or so it would already make a huge difference

@wim_v12e @cstanhope Holy fuck this is grim.

"by 2040 the energy required for computing will exceed the estimated world's energy production."

And that's just the energy after you buy the device, which is swamped by the cost of production.

I know how to make phones last longer: force Google and Apple to stop writing so much code! More software updates → greater power consumption → phones that don't last even the contracted 2 years.

@akkartik just less webshit would go a long way. for 95% of applications a classic form submission would be fine, compared to 20M JS bullshit and fullhd video backgrounds to.. fill in a form. we need to finally go back to native applications, not everything as web application. for that to happen we'd need to fix software development _and_ the debacle which NAT is though. that way you could host your stuff reliably at home.. @wim_v12e

@bonifartius @akkartik In terms of the energy cost of networking, video is currently the main component. But with augmented reality this could become a lot worse because it requires sending voxels instead of pixels, so the data would grow enormously in size.

@wim_v12e Suddenly my choice of a single 1024x768 screen resolution in my computer seems like it may be on the right side of history.

@akkartik In the abstract, more resource constrained devices would be ideal. But I am always reminded of Asimov's novel "The Gods Themselves". Basically, once people are used to a certain level of comfort, they don't want to sacrifice for anything: "they want it enough to refuse to believe they can't have it."
However, I believe maintaining the current level for longer would be more easily acceptable.

@wim_v12e @akkartik It might also be possible to change people's ideas about "levels of comfort" just like it's possible to change people's ideas about "what is trendy" etc. Make the big look ugly.

I personally feel anxiety about too big screens and too high resolutions. Moderate and low resolutions feel more homely, tangible and down-to-earth.

I don't understand why so many people want these huge TVs in their homes. It's scary. Though I assume you refer to screens for computing.

I think it is definitely worth trying to change people's attitudes towards computing. Using things for longer would actually be convenient. And "make the big look ugly" is a great idea for a campaign!

@wim_v12e Big TVs = software updates = more vulnerabilities. If they require an app, that's more loss of privacy. And after all that you won't get HD because you have the wrong cable. It's all so incredibly user hostile. @viznut

@akkartik @cstanhope
The way I see it, it's not that the write updates per se, but that the code is written such that it does only support older devices for a short time. There are some issues with supporting devices for much longer, but they can definitely be solved.

Proposal: tax software creators for the power usage of their software after installation/delivery. Perhaps by accounting for that power when calculating their carbon footprint.

Companies (especially crypto clients) will try to pass on these costs to consumers, which is fine. Increases in these costs will become visible to consumers 👍🏼👍🏼

@wim_v12e @cstanhope @amiloradovsky @vanillacherry

We definitely will need some form of economic incentive/disincentives, and taxation might be a suitable one. Regulation another one. The question is how to convince the political decision makers.

@cstanhope @vanillacherry

@wim_v12e This is all very important. The ideas were something I've been thinking and writing about for some time, but I'm grateful for the great sources you've found.

@viznut Thanks! I've read your article about permacomputing, but only after I'd done mine. I plan a series of articles to keep the momentum going.
Meanwhile I'm trying to set up research collaborations on this topic, but that is not so easy.

This is so great, I don't know how to express how wonderful this is! You're amazing, Wim.

@Eidon Thank you for saying that, Eidon! I really hope that I can make a small contribution in this way. It's a difficult project because so much needs to change, but I believe it can be done.

< The emissions from production of computing devices far exceed the emissions from operating them so even if devices are more energy efficient producing more of them will make the emissions problem worse.
The part after "so" is a repetition, adds nothing.
Attempt at re-phrasing:
"So only increasing energy efficiency of computing devices will not solve the emissions problem"

@iung I agree there is repetition but I don't quite agree that it adds nothing.
It is important to make the point that _even if devices are more energy efficient _, the emissions from production still dominate. The most common mistake people make is to assume that upgrading will save energy.

However, your rephrasing is correct so it is OK if you use that as the basis for your translation.

@wim_v12e I was writing this… when I noticed your article. But yes, we need a constant flow of this stuff!

@viznut @wim_v12e Underlying it all there is this constant push for "you need to do this thing better/smarter", whether it's making stuff or consuming stuff. We're literally and mentally burning out as a species based on that race to improve.

Coming to the conclusion that any alternative narrative maybe needs to come down to just "ok, we've got enough now". Or at least that redistribution is now the main issue, not more stuff.

@scribe @viznut I have no issue with doing things better or smarter (that's my job as a computing scientist, after all), as long as it results in a reduction of resource utilisation rather than an increase. I think we will need a lot of improvement in e.g. our software design practices and software lifetime management if we want our devices to last for much longer. We will also have to improve our repair skills. Just so we can make do with less rather than more.

@wim_v12e @viznut Same here, and I think maybe I'm starting to think about systems in the terms of 'more for less' and 'less for less'. That is, what is a system trying to produce (health? profit? etc) and what resources does it need to do it? Which situations require more output (eg growing population) and which don't (eg I have a limit on how much content I can consume...)

@scribe @viznut Yes, that is the key thing: for any system, we need to ask: what should it do? How can we do this as frugally as possible?

@viznut That's a very nice article, thanks for sharing. I agree with your analysis that the desire for "ever more" and the overuse of abstraction go contrary to what we need.

Sign in to participate in the conversation

The social network of the future: No ads, no corporate surveillance, ethical design, and decentralization! Own your data with Mastodon!