Skip to content

Instantly share code, notes, and snippets.

@aled1027
Last active August 8, 2016 18:57
Show Gist options
  • Save aled1027/8f5d7e188394d15962fc70009e5b9f44 to your computer and use it in GitHub Desktop.
Save aled1027/8f5d7e188394d15962fc70009e5b9f44 to your computer and use it in GitHub Desktop.
A Reponse to "What can a technologist do about climate change? (A personal view)"

What can a technologist do about climate change? (A personal view)

Bret Victor wrote a [nice, comprehsive post on climate change] 1. I encourage you all to read it, and I've written my thoughts on the post.

Overall

[What can a technologist do about climate change? (A personal view)] 1 is a well-written blog post with many links and cool formatting (Control-F "Blinder's Proposal" for an example). The post is written for technologists in response to the question, "how can the tech community contribute to climate change efforts?" As such, the post has a tone of being a call to action for technologists. It recognizes that technologists may struggle with finding good, useful, climate-related problems to work on, and answers the concern in full by giving examples of successful technology projects. The post also discusses Bret's views on how we should think about climate change issues: through models and numbers instead of qualitative claims.

Numbers

The post emphasizes the role of numbers in climate change discussions through the example of Cash for Clunkers. Bret presents a piece from the New York Times that advocates for Cash for Clunkers by modestly stating that with Cash for Clunkers we can "reduce pollution". The author finds this piece unpersuasive, and rewrites the piece using interactive test (with the help of [Tangle] 8, a Javascript library he wrote). The rewritten paragraph presents Clash for Clunkers with a model, showing that an investment of y dollars into Cash for Clunkers that removes cars fewer than x MPG would result in an estimated carbon dioxide reduction of z metric tons per year (Control-F "Blinder's Proposal" for the paragraph in the [article] 1).

I found Bret's paragraphs to be tremendously illustrative. I could see how changing the amount of money invested in the program (y in the paragraph above) would affect CO2 reductions. Interestingly, he didn't provide graphs, which I believe his Javascript library supports, but the model-based paragraph is sufficiently informative and intriguing.

Climate Change is a strange issue because it is a numbers issue. At the core of climate change concerns is average global temperature, which is affected by amount of CO2 in the air, which is increased by CO2 emissions. As a consequence we argue about policies that will most effectively reduce the most CO2 emissions, but when I read a sentence like "We can reduce pollution by pulling some of these wrecks off the road" (wrecks referring to inefficient, gas guzzlers), I don't have any numbers, so I don't have any sense of the impact that Cash for Clunkers may have.

On top of that, even if numbers had been included, I lack the necessary context to understand them. What is the meaning of 9.97 million tons of CO2? Bret addresses this in his rewritten paragraph by adding that that value is equivalent to 0.14% of annual U.S. greenhouse gas emissions (not sure why he would include that number, instead of just CO2).

A better method of contextualizing large values is used by David Mackay who discusses everything in kilowatt hours per day (kWh/d) ([Book page] 10) Kilowatt hours per day are nice because values are low and easy for humans to intuit. For example, a 40 watt lightbulb that is left on for 24 hours uses 1 kWh/d.

Here is the math for Clash for Clunkers, assuming that people drive an average of 20 miles per day, the program removes cars with 15 MPG, and that there are 318.89 million people in the U.S.

Cash for Clunkers kWh per day 
= (distance per day  / dist per unit fuel) * energy per unit fuel per car * number of cars
= (20 miles per day / 15 miles per gallon) * 33.7 kWh per gallon per car * 828,571 cars
= 45 kWh per day per car * 828,571 cars
= 37,285,695 kWh per dar

We would reduce kWh per day used per person in the U.S. by 8.6 kWh/day: that's a 3.4% reduction from the average 250 kWh per day an American uses ([Source on 250 kWh per day per American] 12).

However I digress, since that analysis I provided was focused on energy usage and not carbon emissions, but hopefully my math is right and that puts the Cash for Clunkers program in perspective.

Ideas for Technologists

Bret suggested a number of ideas for technologists to work on. They fell into a few categories:

  1. Improve the technology around a renewable energy
    • e.g. make solar panels move to find the Sun
  2. An easy way for technologists and entrepreneus to see successful and failed projects in climate change response.
  3. Improve the tools for communicating about climate change
    • Tools for communicating with models (as discussed above)
  4. Better tools for getting data
    • Bret asks "What if there were an “npm” for scientific models?" For those who don't know, npm is a tool for Javascript programmers to easily download and use code written by others. So Bret is suggesting that we need a better tool for downloading climate change data. I'm not sure if he has an issue with APIs and instead wants a dedicated tool. In my experience, Quandl and similar sources are nice for downloading data, but they lack data relevant to climate change.
  5. Better tools for analyzing data

I thought that these ideas were great. They moved beyond the typical advice, and offered concrete suggestions and examples, and I now have a better of idea what role technology can potentially play in climate change movement.

For one, technology can change how we communicate (as it so obviously already has). But communicating with models can dramatically improve communication, and technologists can create better tools for communicating via models. This includes allowing users to see and interact with graphs and numbers, which is becoming more and more popular by the month with the growth of interactive data journalism (e.g. [538] 14). But as far as I've seen, most of the work is not focusing on climate change, but on economics, trade and politics.

Moreover, there are many areas surrounding clean energy that could be improved. One example that Bret didn't include was the internet of things. Many open source tools are being developed for Python and other languages that allow programmers to connect devices inside of their homes. When devices can communicate, they operate more efficiently and make a home a better experience.

Some other thoughts before I sign off. Naomi Klein, author of [This Changes Everything] 13, suggests that we use communities and a closer-to-earth lifestyle to combate climate change. One instanation of this suggestion is low-carbon hobbies, such as biking and booking, as opposed to hiking and climbing, wherein a participant needs to drive to the location, or boating, which uses gas the entire time. I can imagine that technology can be used in activities and hobbies to reduce their carbon footprint. Not sure how - but a thought.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment