A Permacomputing Primer

2023-07-31

From analyzing environmental data to powering cleaner technologies, computers undeniably have a significant role to play in tackling the climate crises. Yet, while increasing computational power has long been cited as a driving force for improving efficiency and reducing emissions, in reality, things are not quite that simple. Instead, the demand for computing technology has resulted in a paradox: the technologies supposed to empower us are leading to significant waste, insatiable consumer demand, and increased pollution.

2023-07-31

A study by Lancaster University suggests that ICT currently accounts for around 2.1–3.9% of global emissions. And crucially, these emissions (despite the increased efficiency of computers) are growing. If we assume continued relative growth, ICT’s relative contribution will exceed 14% of the 2016-level worldwide GHGE by 2040. In other words, around 10 GtCO₂e, or about 80% of the acceptable CO₂ emissions budget.

In response, tech optimists often argue that innovation will save us from disaster. However, even a near-future utopia of super-efficient computation does not solve the problem. That is because, as computer science professor Wim Vanderbauwhede writes:

“We can’t rely on next-generation hardware technologies to save energy: the production of this next generation of devices will create more emissions than any operational gains can offset.”

Instead, in the face of ecological crises and the potential collapse of the current global order, we need to drastically change our approaches to technology to prevent disaster or prepare for it. And in this vain, attempts have been made to theorize, investigate and implement an alternative technological culture that centers environmental and sociological concerns through ‘radically sustainable computing.’ Increasingly central to this premise is the concept and community of practice known as “permacomputing.”

What is permacomputing?

Permacomputing is an approach to computation inspired by permaculture that encourages sustainable technology and networking. At its core, permacomputing seeks to find the answer as to whether humanity can utilize computing in a manner that positively contributes to the environment, the biosphere, and society. Or, as author, Jonny Hankins writes:

“From the starting point that technology has harmed nature, the concept aims to re-center technology and practice and enter into better relations with the Earth.”

Just as permaculture seeks to empower resilient and regenerative practices in agriculture, permacomputing hopes to do the same in the technological sector. But what does this all mean in practical terms? How can permaculture concepts ever be applied to computation? To answer that, we first need to identify the issues permacomputing seeks to tackle.

The problem with modern computation

The last century has seen an explosion in available computational power thanks to improving data storage and algorithmic logic. But this ever-increasing abundance, taken for granted as it is, has led to wastefulness, underpinned by presumptions of perpetual growth and infinite resources.

These reckless industrial approaches to computation manifest in a multitude of ways, including:

Ridiculous hardware requirements for trivial tasks
The mass production of inordinate consumer “smart” devices with short-lived manufacturer support (as opposed to products that offer real social value
A lack of regulations regarding the production and disposal of valuable resources such as microchips
Planned obsolescence
A bloated internet, worsened by corporate-backed ad-tech
Energy-consuming centralized network processing and data storage

Arguably, these issues are exacerbated further by a cultural detachment from modern technology, especially mediums such as streaming, where the consumption of resources appears intangible, as “virtual” or “in the cloud.” But perhaps the most significant issue preventing a more sustainable tech sector lies with current computational lifespans.

As a report by the European Environmental Bureau details, to curtail emissions, current devices such as phones and laptops should aim for lifespans of around 25 years. But thanks to planned obsolescence and the insatiable demands of consumer culture, we are not even close to reaching such targets. Indeed, the average lifespan of a smartphone is around 2.58 years (worryingly, this figure is on the downtrend), laptops are not lasting much longer, and even servers in data centers only last around 3–5 years.

But not only is the tech industry’s endless production of short-lived consumer products causing environmental damage, but it also leaves the industry vulnerable to the ramifications of this wasteful mindset.

Post-collapse computing

As noted in the paper Collapse Informatics and Practice: Theory, Method, and Design by Bill Tomlinson and other authors, “continuous growth is, by virtue of limited global resources, unable to continue indefinitely.” And research into multiple fields, including environmental studies and anthropology, suggests the collapse of the current global civilization due to these limitations is a real near-future possibility.

Some scholars, such as Jared Diamond, even go so far as to suggest that this transformative process has already begun or will begin within the next several decades:

“[A]t current rates most or all of the dozen major sets of environmental problems . . . will become acute within the life-time of young adults now alive.”

To be clear, collapse in this context should not be equated with apocalypse but the decline of the worldwide corporation and supply chains that underpin global civilization. Such a scenario would likely be tied to shortfalls in energy and resources and environmental damage. The result would lead to scarcities of many kinds but would be particularly destructive to the ICT sector, especially the already volatile production and distribution of microchips.

Such an imminent collapse cannot be easily predicted, but as Tomlinson and co note, it is vital to consider “how to develop sociotechnical systems for use in these scenarios,” however likely they are to occur.

In other words, we either have to change rapidly to prevent a collapse, or change rapidly to prepare for it. What we cannot do, is wait and see.

Permacomputing aims to inspire this transformative process.

What might a permacomputing culture look like?

In permaculture, all actions are assessed by their impact on the local ecosystem. For example, an individual practicing a permaculture approach to agriculture will consider how removing vegetation affects local wildlife or soil quality. Or they may ponder the implications of changing an overland water flow based on the potential changes it may cause downstream. In each case, they will put the long-term resilience and ‘renewability’ of the ecosystem first to ensure the future sustainability of their practice, even at the expense of short-term profit. Permacomputing aims to apply this mindset to technology, beginning with the all-too-crucial manufacturing stage.

Manufacturing new computer hardware requires lots of energy and resources, not to mention the creation of undesirable byproducts. The production of one of the most vital components of computers, microchips, is especially resource-intensive. As a result, according to permacomputing principles, they should be treated as precious resources — because they are — and their lifespans maximized. They would not be reduced back to raw materials until absolutely necessary. Instead, permacomputing advocates argue that chips should have redundancy and bypass mechanisms to keep them functioning after certain internals wear out. In other words, manufacturers should produce for planned longevity and not planned obsolescence.

Closely tied to this concept of planned longevity is the idea of manufacturing with disassembly in mind. That is, making sure all elements of a device can be disassembled for repair once it reaches the end of life. The result would prolong the initial life cycles of hardware and enable the reclaiming of components. In practical terms, this means taking actions such as using fasteners instead of adhesives, modulation, open documentation, and making components compatible with everyday tools.

“Unlike the nebulous goal of designing a sustainable product, designing a product for disassembly is a more concrete, quantifiable approach.” — from the wiki.xxiivv entry on permacomputing

Another vital consideration in permacomputing is creating tech with collapse resistance in mind. To be clear, this should not be mistaken for “developing for collapse” (though such projects have their place). Instead, collapse resistance means creating with an openness to utopia-scenarios while preparing for collapse or dystopia. Doing so may mean removing unreliable dependencies (including hefty hardware requirements), creating scalable systems, and ensuring documentation is made freely available.

Again, the key focus here is ensuring the long-term usability of computational ecosystems instead of increasing short-term capabilities at any cost. To that end, a networking system in a permacomputing sphere would preferably be small, simple, adaptable, and vernacular, utilizing local resources, while scaling to requirements (as opposed to 24/7 full availability).

Of course, such concepts as these and others are easier to propose in theory than to execute in practice, especially with current consumer market expectations. But while real-world examples of permacomputing are hard to come by, they do exist.

Permacomputing in practice: examples

There exist various software utilities, protocols, art movements, and other projects that, intentionally or not, incorporate elements of permacomputing thought. Some examples include:

The Small File Media Festival: An annual festival founded in 2020 designed to raise awareness regarding the high carbon footprint of streaming media and to propose and celebrate alternatives.
Low Tech Magazine: a website run entirely on solar power that offers low-waste tech tutorials
Collapse OS: An operating system and series of tools designed to enable people to program microcontrollers following a civilizational collapse.
Project Gemini & Gopher: alternative internet protocols that are adware-free and minimalist by design.
The Small Web: a movement based on minimalist static websites and old web aesthetics.
Solarpunk: an optimistic art movement that envisions a sustainable and egalitarian world.
The Fairphone: A smartphone that features some elements of permacomputing principles, including replaceable parts, long-term support, and the use of recycled materials.

The above examples each serve to provide some proof behind the concept of permacomputing. However, real substantial change in the technological sector would require a transformation of the current computing culture. And such a change faces several significant hurdles.

The challenges facing permacomputing

For the moment, permacomputing and other associated concepts, such as ‘frugal computing,’ remain fragmented and underground, despite academic interest. And there are many obstacles to achieving a real lasting transformative effect on mainstream tech culture. Perhaps the most prominent of which, is the difficulty of imagining an alternative tech culture in the first place.

Due to the current state of capitalist realism, it is nearly impossible to imagine a tech industry that is not focused on unlimited growth. And as the programmer and blogger Ville-Matias “Viznut” Heikkilä notes in a 2021 essay:

“Even the most interesting real-world examples (such as the Soviet computing culture) exist somewhat in the shadow of Western developments and ideologies. So, there’s no real “other” to contrast the growth-obsessed mainstream computing with.”

Another issue is that if computers are to have longer lifespans, this will ultimately result in more opportunities to exploit their vulnerabilities. A move towards permacomputing standards would then, require substantial improvements in cybersecurity.

Finally, to stop corporations from utilizing practices such as planned obsolesce to increase profits at the cost of sociological and ecological destruction, firm regulations would be required. The current state of global politics, and the role of microchips and semiconductors in that realm, means that any voluntary ‘slowing down’ of technological progress could have ramifications on a global stage.

Nonetheless, while imagining significant changes in the tech industry is difficult from our current standpoint, such change is both possible and necessary.

Redefining progress in tech

In both the permacomputing and wider degrowth spheres, it is held that to move forward sustainably, progress must be reframed as an abundance of ideas rather than an abandoning of the old in favor of the new. To this end, in the permacomputing space, limitations are considered liberating, and a source of creativity, the abandoning of maximalism is viewed as positive, and coarseness is associated with quality, not sacrifice.

This reframing of progress is not to be confused with forced deprivation of “going back in time”. Instead, permacomputing seeks to: remind us that progress does not move in a linear path and that even primitive technologies can produce innovation; to prepare for the next century’s technological challenges, accounting for both pessimistic and optimistic projections; to step away from the current state of affairs where old ideas are endlessly rehashed with newer specs and to re-center ecological and social integrity as the digital climate moves forward.

Creating a sustainable form of ICT is undeniably a lofty challenge. But it is by no means impossible. And while a permacomputing utopia is unlikely, the ideas and frameworks it puts forward may yet serve to inspire the next generation of computer ecosystems. Following decades of excess, the biggest hurdle may simply be a matter of relearning that less really can be more.