How is the tech industry contributing to global warming and what can we do to reduce the impact?

deepu105 profile image Deepu K Sasidharan ・1 min read

As members of the tech industry we should be aware of how we are contributing to global warming. More then that we should talk about what we can do to reduce the impact.

For example the boom of cryptocurrency has been contributing to global warming by increased use of electricity to mine and probably wasn't the best contribution of the tech world. What are other issues that we have created or contribute to? Of course, then there is also node_modules.

Let us discuss to increase awareness


markdown guide

A few months ago, I read “How Improving Website Performance Can Help Save The Planet”, by Jack Lenox. It was fascinating since I had never thought about web dev in these terms. I speak about web development since it’s my field: we need to create more unsophisticated web stuff. Many web designer and developer don’t even know a webpage with different sizes, scripts, and resources have different impacts on the climate change.

The more resources we force the user to download to enjoy our webpage, the more servers need to pump in. This can be insignificant if you think about only one webpage, with few views. What if the same webpage is viewed millions of times in a month only? The results can be considerably damaging.

To give to the web developers’ community here on dev.to some advices, please:

  • reduce the file size of your HTML, CSS and JavaScript files;
  • reduce the size of images;
  • don’t use video if you don’t absolutely need them;
  • try to use web safe fonts;
  • don’t put things such as ads, social widgets, analytics, pop up on your webpages if you don’t absolutely need them.

The best advice, in general, is to make your webpages more lightweight, so servers can use fewer resources.

In addition to servers, your users’ devices will thank you too, because simpler pages require less CPU power. Less CPU power requires less energy, so the battery will last longer, and you’ll save cellular data, too.

Of course, then there is also node_modules


I haven't looked at any studies on crypto specifically, but this article has some interesting info.



Write efficient software, especially if it will run in large scale in the foreseeable future. In large deployments even minor speedups that only save minute amounts of power per execution accumulate to significant energy savings. Thinking of a debate I read about the relevance of efficiency for adblockers. It might not matter for a single person but if the millions of people using abp would switch to ublock origin that could add up quite a bit.
Then there are things like lot letting a CI pipeline run on every commit for all of your side projects. Just do it for master (maybe dev, but not every little change). Shutting down cloud VMs when you're not using them and many of these smaller changes.
The choice of language can also matter. A rust application will use a lot less resources and thus energy than the same thing on nodejs.


I think the single biggest thing long-term is probably going to be working harder to improve energy efficiency of hardware.

Put simply, data processing requirements aren't going to go down any time soon, so we should be focused on improving the computational power per-watt of computers.

Luckily, things are already (slowly) moving in that direction for other reasons. ARM, if it ever becomes big in the HPC or desktop/laptop markets, will likely help a lot (it's got significantly more processing power per-watt than most x86 chips do), and I suspect RISC-V will help too if it gets significant uptake. It would be nice if this movement could be faster, but the fact that it is happening at all at all is still rather reassuring.


Yes, definitely agree. Excellent point. Do you think the advancement in Quantum computing can help?


It might. Right now though, it's too soon to tell because:

  • Quantum computing only improves performance for some things, not everything. You need actual quantum algorithms, and it has to be something that can actually benefit from the parallelization involved. I/O, for example, is never going to get faster because of quantum computing, because there's just nothing that it can improve there (except possibly transparent compression or encryption, but that remains to be seen).
  • It's still rather inefficient. The absolute scale of quantum computing is too small to be useful on a practical level right now. If things keep improving, this will probably eventually change, but I doubt it will be soon.

Yes right now its not practical but I hope that in the future it helps as that is the only path forward as current technology innovation has reached the peak and Moore's law seems to be slowing down already google.com/amp/s/www.cnet.com/goog...


Certainly electricity use has gone up considerable to support the the much larger web, but also I would say the the tech boom cities (in USA) have made long commutes very common and thus increased gas consumption dramatically.

Fixing the green house gas problem is not easy, but it can be done and needs to be done fast. The only truly viable solution is nuclear energy. While I like the idea of renewable energy, none of it is reliable enough or able to generate enough power for our energy needs. Let's be honest, we are not going to reduce our electricity usage by 80% in the next 10 years, if anything it will increase by that much. So we need something that generates huge quantities of power, whether it is windy, rainy, or cloudy and we need to be able to contain the waste. Nuclear power generates tremendous amounts of energy, and while not ideal the waste is containable.

So to fix the electrical side of things we can support and vote for nuclear power initiatives. To fix the commute problem...well that is something else entirely.


Here in the Netherlands, I think we are relying fully on wind and solar energy. My household runs on 100% wind energy for example. But yes Nuclear energy is also a viable alternative and is better than fossil fuel energy.


That is great to hear! There are a few places where renewable energy is effective due to the the properties of the area and the population. Iceland for example has a lot of geothermal energy. I was mostly targeting the USA as that is what I know best.


Source of the problem

Lets find the source of the problem. Grab the world co2 emission data.
If you look at the co2 emission rate, there are some countries that stand out:

Problems have to be solved at their source. It's clear this cannot be solved in a little country like Liechtenstein, Netherlands or Denmark.

So if you want to solve this problem, you have to find a technological solution to reduce emissions in China and others.

This is a challenge, because it's not only a technological problem but also a social, economical and political problem.

Current trend

"It is difficult to get a man to understand something, 
 when his salary depends on his not understanding it."

The GDP of China is rising, and as a result will create more pollution

This is not just a problem for us, but also for the Chinese.


To summarize, create technological products for the Chinese market will have most impact. Current trend shows increase in co2 pollution.

As a country with 1,420,135,488 people, it's in their best interest to solve this problem. I'm sure they can come up with more ideas than me, I can't save the world


Not Just China, every country, especially the top 10 must act. But its unfortunate that none of them do :(


From time to time, this question pop into my head. I'm fully aware that software/hardware has a huge impact on the climate since there is a lot of resources used to craft all the products and their infra and to maintain them. This discussion is promising and I can't wait to read the answers.

I myself have a hard time finding a solution.


Yes that's the biggest hurdle, finding solutions.


Each bit saved is ~2.5fWh less electricity spent ;-)