Aim for this test is to start discussion regarding website impact on climate and environment. It is not perfect but hopefully a start.
We are giving websites a relative rating depending on the impact they would have IF they had the same visitor count and technical solution. We are doing this by taking the weight of the url in KiB and calculating a value from this. We then compare that value with a reference values and gives you a rating. The reference values represents the percentile for all urls checked by Webperf.se. This is updated manually and you can see when it was done latet by looking at the date in top of /tests/energy_efficiency_carbon_percentiles.py).
If you know any other way we could automatically compare impact on climate and environment a certain url has, PLEASE let us know :)
This section has not been written yet.
This test is using Google LightHouse in the background so please follow instructions on page about Google Lighthouse Based Test
Below are the steps that you need to do to calculate a new carbon percentile reference file. As you can read above, this are required if you want to have a up to date reference regarding carbon footprint. It is also needed if you want to have your own reference to rate against, for example your own websites last year or your closes competition.
You do this by running Energy efficiency against a list of all sites you want to compare against. For example:
python default.py -i webperf.csv -t 22 -o data\carbon-references-2022.json
You now have a baseline to create your carbon percentiles from.
You do this by running carbon-rating.py, you can see a full list of argument by writing python carbon-rating.py -h
We recommend running it as follows:
python carbon-rating.py -i data\carbon-references-2022.json -o tests\energy_efficiency_carbon_percentiles.py
For webperf-core to use your new percentiles it has to be placed and named as follows
tests\energy_efficiency_carbon_percentiles.py
No frequently asked questions yet :)