Issue with the 2024 sustainability chapter - Page Weight calculated differently than in 2022

[ko-jordan]

Thanks for the great job to everyone involved first of all.

I noticed that the Page Weight percentiles in /en/2024/sustainability#fig-1 were calculated differently than in 2022.

If I'm not mistaken, in 2022 the percentiles were calculated using only one specific date: 2022-06-01. This time the percentiles were calculated by using the average of percentiles of six dates (2024-01-01 - 2024-06-01).

The carbon emissions on the other hand were calculated using only data of 2024-06-01. This can lead to a bit of confusion, because it seems like page weight has decreased but carbon emissions have increased since 2022.

[burakguneli]

Please correct me if I'm wrong but as far as I can see from this query, the shared figure is data of multiple months instead of just 2022-06-01

[ko-jordan]

I believe gthub.com/.../2022/sustainability/page_bytes_per_type.sql is the query used in Figure 1 of the 2022 Sustainability chapter. The data matches at least.

[ko-jordan]

Let me try to explain my case again:
Figure 1 and the paragraph below suggest that page weight percentiles have decreased since 2022. The carbon emissions section shows that carbon emissions percentiles have increased since 2022. This is irritating to the reader, because earlier chapters explain that carbon emissions are highly correlated to page weight.

The reason for this contradiction is that the data for page weight is derived from averages of percentiles of the first six month of 2024 while the carbon emissions are calculated using only data from June 2024. In 2022 both page weight and carbon emissions were calculated using data from June 2022.

My suggestion would be to calculate the page weight with June 2024 data only.

[ldevernay]

OK, I think I get your point, we'll look into this.
Thanks for the feedback.