4 ms·
We can also think of using less CIs and automated jobs as beneficial for environment (less energy used, ..)
by 11235813213455 6y ago
We can also think of using less CIs and automated jobs as beneficial for environment (less energy used, ..)
- Trasmatta 6y agoThis would likely be an insignificant blip in terms of the world's total energy and CO2 output. I think focusing on those types of micro-optimizations is unhelpful as they don't make any meaningful difference, and take up way too much attention. It's like making people feel guilty for taking a shower in a drought, when residential water use is less than 1%. Every person could go the rest of their lives never again taking a shower, and it wouldn't make a difference.
- 11235813213455 6y agoYour way of thinking is very common, not wrong, but terrible, why? The world's total energy and CO2 output is simply the addition of each human footprint, all industries, transports are more or less directly related to the human end-consumer. So if everyone decide like you that their own behavior is insignificant, which is true, nothing is changing, and as you know the current trend on environment is not sustainable. I don't use a car, I don't use air-conditioning, I find natural methods and alternatives for most things, and product to reduce my footprint, and I hope everyone at some point will do that, because it'd change everything if everyone cut by a half their footprint, it's like twice less humans. My comment on CI is half-serious, but I still avoid to put many OS/enviroment targets (for example just Linux/nodejs-14 is enough) and just running in on the main git branch
- Trasmatta 6y agoMy point is that we should focus on actually effective changes. Human nature is such that focusing on the individual and the small changes that each person can make will not be effective at the pushing the needle as far as we need it to go. We need effective legislative, technological, and infrastructural changes. It's good if people reduce their personal carbon footprint, I just think we focus too much on the micro optimizations. Game theory and human nature are such that we won't ever get where we need to be unless the we have effective top down solutions.