Download this PDF slide deck | Link to a commentable, linkable, navigable Miro board of this deck with references | Link to talk recording on youtube
Below is a transcript of the talk given for the British Computer Society Insights conference in 2021.
Below is a fairly unedited script – you can use it to search for a specific term, but all the links and references are in the publicly visible miro board linked above on this page.
hello there, I’m chris.
I’m a director of the green web foundation, a non profit set up to track the transition of the internet away from fossil fuels.
I also am one the organising team of climate Action.tech a community of technologists working on climate at various levels.
in the time I have with you I’m going you introduce some terms, that I have found helpful when thinking about tech and climate.
I’ll then talk about some of the levers we have for change as technologists.
then finally I’ll wrap up with summary to take away with you.
But first, let’s start with Climate Emergency, and Climate Response
Normally when I do talks about climate and tech, it’s common for me to start with a bunch of scary statistics, and photos to get across the that we need to act now.
If you’re watching this, I figure you know we need to act, but you might not know how many other people feel the same way now.
There is a large and growing collection of governing bodies who have declared emergencies now, from the local and city level, to the national like the UK, and even at supra-national level, like the European Union.
What we tend not to talk about so much is why we are here.
We’re in this state because we haven’t found a way off fossil fuels.
Engineers we tend to like to jump straight to solutions when faced with problems, it’s really, really important when talking about climate to remember that effective action is about the need to take deliberate decisions, as well as thinking we can innovate our way out of problems.
To give an idea how much the cliamte emergency is related to our inability to get off of fossil fuels, it’s useful to look at the data to help understand it.
What you’re looking at is information from the openclimate data website, showing the same data that goes into the big IPCC reports.
See the big brown bit on these these datasets?
That’s us burning fossil fuels. Each year, we emit around 35 gigatonnes of carbon dioxide into the sky. Because carbon dioxide in the atmosphere traps heat, it heats up our world further.
Even when we went into lockdown, it decreased, but not by much. We might have decreased emissions by 2-3 gigatonnes.
But you don’t need to be a scientist realise that taking 35 minus 3 is not zero. Even during the COVID lockdown we were still putting extra CO2 in the atmosphere.
If we want to avoid more of the damage we’re seeing on the news daily, we need to reduce emissions. Every year we put off meaningful action, we make things harder for ourselves.
Instead of paying in small amounts over longer time and starting early, when we delay we build up more and more.
This chart was made in 2017, but the general direction of travel is about the same – just like a code base where we’ve put off paying down the technical debt, we’ve done the same thing on a planetary scale.
Now, the scale, and rapidity of changes we need to avoid the worst impacts of climate crisis is leading of new kinds of language being used by scientists.
We might need a bit more context for these words, so I’ll try to summarise it quickly.
Since the 1990’s representatives from almost every single country in the world have been meeting at a big conference called COP, each year to figure out how to act meaningfully on climate together.
In 2015, after 21 goes at it, we finally had a breakthrough, which you might have heard referred to as the Paris agreement. One of the decisions was that yes, we really do need to act on, and we definitely need to cut emissions to stay within 2 more degrees of global warming.
While agreeing on 2 is better than no agreement, number of smaller island states, basically said “this is fine, but we still drown at 2 degrees, and the human costs is immense. What about 1.5?”
Globally we commissioned scientists to see the difference was between 1.5 and 2 degrees, and what kind of changes we would need to make globally.
They released another massive report in 2018. The short version is “yes. 1.5 degrees is much, much less bad than 2” and three words were used in a message to policy makers to describe the changes we now need to make, in the face of all the delay we’ve seen.
This languge is important here.
If we’re in a climate emergency, then talking about personal carbon footprints won’t cut it.
When we talk about carbon footprints, we are usually talking about how to make them as small as possible, before we can go make back to our normal lives.
We don’t need that, because it’s not about one person, or even one company any more. It’s about the entire system, and as the scientists say, it’s about rapid, far reaching, unprecedented changes now.
In response to this state, I think we should talk less about personal carbon footprints, and instead talk more about climate response.
It’s not about getting back into eco-heaven and being personally free of eco-sin.
It’s now about safety, and avoiding harm – climate emergencies require honking great, ambitious climate responses.
Levers as technologists
Ok, so if you come along with me and accept that a) we’re in a climate emergency and b) we need to be thinking about ambitious climate responses, it’s fair to ask what we can do as technologists.
I think it’s helpful to group the kinds of things we can do into three kinds of levers we have as technologists where our skills are applicable
Having these lets us at least see when we might be missing something obvious, or think about stuff we might do anyway in a different light.
Ok, the levers then.
consumption – can i change how much we need?
intensity – can i change how much we pollute?
direction – can i change where we are heading?
I’ll explain these in more detail
One lever we have as technologists that we might be familiar with is consumption.
We’re taught as engineers to value efficiency and avoid waste, and to get a good example of this lever at play, it’s useful to look at the internet itself and what is has enabled.
If you were to ask this question, you’ll get a wide range of replies depending on who you speak to.
Because most of us have a hard time picturing what a gigatonne of carbon emissions is, when I answer this question I tend to talk in relative terms – comparing it to stuff you might have heard of.
For the purposes of this talk, you can think of the tech sector as having a carbon footprint of around 1-3% of global carbon emissions. This is in the same ballpark as:
- the entire shipping industry, or the entire aviation industry
- the carbon footprint of Germany or Canada
- all combined emissions of commercial and heavy industry in Europe
Once you you have an idea of the footprint, you might what different parts have.
I’ve found this chart from the Ericsson’s recent report on digital carbon footprints useful.
You can see in this chart that making computers causes emissions, because remember – we need to melt sand before we can start turning it into microchips.
Unsurprisingly, running computers uses loads of energy, which causes further emissions. This chart also shows us that because networks and datacentres are in used 24/7, energy makes up a larger part of the footprint than gadgets. This means that energy efficiency is a big lever here.
In fact, energy efficiency has of computing has been something of a success story.
Over the last ten years we’re been able to benefit from loads of valuable digital services, and use has exploded, but when you look at the total energy use by datacentres and internet infrastructure, it hasn’t changed that much.
Imagine what this pandemic would have been like without all the digital services we really on for example
If we look in more detail at these numbers we see this this has been made possible largely by us switching from moving away from running code in relatively inefficient, smaller scale datacentres, with relatively low levels of utilisation, to running code in hyperscale, cloud datacentres which can be have ten times as many applications using the same hardware.
Also, because they’re so much larger, they can afford to replace older inefficient kit more quickly, making better use of Moores law.
This has also come at the cost of a much less diverse tech ecosystem, but from an energy use point of view, things could be worse
It’s worth noting though that Moore’s law has been slowing down, and increasingly, as gains will need to come from closer to the ‘top’ of the stack, where we live, rather than being able to relying further down the stack at the hardware level.
This means that as a developer, understanding performance and efficiency is likely to become more important over time.
So that’s consumption.
But as Saul Griffith says of Rewiring America, you can’t efficiency your way to zero.
I use the term intensity here as a more general term here, to introduce another lever, and this is likely one we’re not so familiar with.
Let’s use the internet again as an example, as it’s something we know.
The best example I can think of for this would be the simple fact that while the internet is probably the biggest machine in the world, it’s also the biggest machine in the world mostly powered by fossil fuels.
We might not see the impact directly, and when we talk about cloud computing, we tend not to think about burning coal or fossil gas but we should.
When we burn fossil fuels, there is avoidable harm being caused, not just through the carbon emissions as we mentioned before, but also from the air quality.
New, peer reviewed research on the impact of air pollution from fossil fuels is pretty sobering. This pollution was responsible for one in five early deaths in 2018 globally.
To be fair, this isn’t an internet problem so much as an entire energy system problem.
This here is a chart from the UK think tank Ember climate, showing how the global electricity mix has changed.
This chart from shows shows how it’s changed over the last ten years.
We are seeing changes, and you can see that green wedge of renewable energy is growing, but given how late in the day it is winning too slowly is the same as losing now.
Earlier in May, the International Energy Agency released a report, describing what a climate response to stay within 1.5 degrees of warming would look like.
This is the first time such an influential body has shared a roadmap, and it shows that it’s possible, but as the scientists say, the changes would have to be rapid, far reaching and unprecedented.
We can decouple the harm caused by the underlying energy systems from our use of energy.
But to the only way we can do this is leave fossil fuels in the ground, and in the past. They explicitly say no more exploration for fossil fuels
Taking the place of fossil fuels would be truly breathtaking amount of new renewable energy generation instead – more than four times as much as we have today.
It’s not the only game in town – we still would have nukes, and geothermal, and maybe some hydrogen.
But make no mistake, most of the work is being done by wind and solar, and I’ll explain why – as it’s related to a law we know, called Moore’s law.
Moore’s law doesn’t happen because computer magically get cheaper – they get better because as we make more of something, we get better at making it. This huge increase in scale is what changed the price in chips, and what we are relying on transition from fossil fuels.
And this scaling effect, is something that tech firms have played an admirable role in, and is another success story.
What you see here is a chart from the IEA, showing power purchase agreements – where companies agree in advance to pay for all the power generated by a wind or solar farm, so it can be built.
These agreements are one of the most effective ways to get new wind and solar built, and scale it up. We can’t change the physics of climate change, but we can change the economics.
The blue bits represent the efforts of tech companies like Google, Microsoft, and so on to do this so far – they’ve been the biggest corporate players by a long way.
It’s not just US tech companies though.
The UK is a world leader in offshore wind now.
About ten years ago the UK government made a deliberate decision to change the economics of offshore wind, by scaling up its usage.
It put out a massive tender to buy power at a fixed, fair price for the next 15 years, but they used a special instrument called a contract for difference to encourage energy companies to compete to deliver the best bid, but also deal with the risk around the new technology.
If the provider’s costs were higher than the amount fixed price for power, the government agreed to pay the difference, reducing the risk to the company.
However, if the company was able to to deliver the power cheaper, over the 15 year period. the provider it would have to pay back the money into the public purse.
This was really successful! Rather than paying to subsidise private companies, the UK government is on track to have 600 million pounds paid back to the tax payer by 2027, as the cost of offshore wind has come down faster than anyone thought possible.
It’s not just huge tech firms, and the UK government that is doing work to green the underlying grid we rely on.
One company I’m a fan of is Ripple Energy, a UK based tech startup that’s now using the ability of the internet to coordinate finance and to allow crowdfunding and shared ownership of renewable energy a much smaller scale.
They make it possible now buy a share of a wind turbine, that has a name and that you can point to on a map. In return you get lower rates on electricity you pay for, because you’re now a part owner of an energy asset, not just a regular rate payer.
This model also works for businesses too and I find it much more compelling than offsets – if you’re a tech firm and your emissions are caused by fossil fuels on the grid, this way you are addressing the source of the problem, and buying power in a way that is both scalable, and greener than a lot of green energy tariffs, and gives you a wind turbine you can point to on a map.
Ok, we’ve covered consumption and intensity now.
Time for direction.
This is the one we’re probably the least likely to have encountered, because a lot of the time, if you’re a technologist you might be more familiar with being rewarded for solving a problem rather than defining a problem.
I use direction here because many of the problems we associate with technology, I would argue we see because we are not really fully thinking through the consequences of going down a particular road. In other cases they happen because we are leaving people out of the conversatio, or in some cases not even questioning how we got to where we are in the first place.
If you’ve ever used the Signal messenger app, you’ve likely used some of Cade Diehm’s work, and I find this quote really useful for thinking about how we use technology.
An accellerant is a substance that takes whatever is already there, and speeds it up.
Certain enzymes are accelerants for reactions, but you might also use it to describe how petrol when thrown on bonfire is an accelerant.
Sure it can be useful, but it’s easy to cause harm when you mishandle it – and there is an expectation that you know this when you use it.
The other reason I like it is that it emphasises deliberate choices.
We spoke about climate emergencies before, and we saw how Ripple Energy are working to accelerate community ownership of renewables.
Much of mysociety’s recent work on climate is all about accelerating the process of civic engagement on a local level, because that’s where so much needs to happen.
The ideas is wonderfully simple. They collect all the climate Action plans for all the local councils that can find, and make them easily findable, and searchable.
This makes it trivially easy to search, and see if your local council has a plan yet. But it also makes it easy to see which other councils have one too and compare them, without having to learn how to use every single council site or know where to look.
It provides a useful first step towards engaging as a member of society, and to use democratic structures that are already there, rather than trying to reinvent new ones from scratch.
Direction is also important because if you don’t think about where you are headed, things you do that you think are helping, can end up doing more harm than good.
You might have heard of the rebound effect, or Jevon’s Paradox, named after William Stanley Jevons who, first wrote about this in 1845.
He noticed that every time steam engines were made more efficient, the total amount of coal used jumped, because people found new uses for them, and used them more.
You can see the same effect moving from 3G to 4G, to the number of meetings we had on zoom as they became easier to arrange.
More concretely, this effect is one you see with ride sharing services.
When Lyft and Uber first came into the market it was easy to see them as a sustainability win, as they could replace car ownership.
This sound good, but the convenience was a double edged sword.
Ride sharing in many cities can end up displacing greener forms of transport, especially when they are subsidised to be even more attractive, pricewise compared to other options.
Direction in this case refers to understanding the implications of going down a route, and if necessary adjusting when you have a better idea of the consequences.
Direction even applies to startup accelerators in the tech field.
Climate Subak was set up earlier this year to take the model of an accellerator that typically cranks out investment ready startups, designed to to raise subsequent rounds of VC funding. However, it instead cranks out non-profits with ideas that can be scaled by subsequent rounds of funding from impact investors for greater impact.
Instead of measuring impact in returning an initial fund, they measure impact in terms of policy changes delivered, or carbon reductions through behaviour change.
You see many of the same techniques as you would see in a typical tech start up accelerator, but the direction is different, and especially the outcomes.
So I think that’s all I can fit in the time I have with you for now, but here’s quick summary:
We’re in a climate emergency – we really, really, can’t afford to delay.
Climate response – lets move on from talking about climate neutrality, and making things small, and instead focus on climate response – we need rapid, far reaching and unprecedented changes now.
And finally, your levers as a technologist – consumption, intensity, and direction.