It would be great to gather insights on the adoption of the loading attribute on img tags over time, including how many sites use it with a value of “lazy” vs “eager”.
Accumulating data for this adoption over time, e.g. from currently until end of the year would be super valuable, especially in order to identify potential spikes and their cause (e.g. the attribute being adopted by certain browsers or CMS platforms).
Some prior research into this: Chrome Image Lazy Loading - Sites Already Using it on Week 1!
@felixarntz are you looking to get a live look at the current adoption rates or is it more of a year-end retrospective? If the latter we can revisit the queries in the linked thread, otherwise it requires something more elaborate.
@rviscomi Thanks for the context, it would be great to have more of a “live” view of current adoption over time. For example, could we do a daily/weekly/monthly (whatever is feasible here, weekly is probably a good middle ground) rundown of how many websites use the loading attribute on img tags? A preferable outcome would be e.g. creating a figure that has individual graphs for the three possible values “lazy”, “eager”, and “auto”, and show the week-by-week adoption in number of sites.
This sounds like a good case for adding a new metric to the State of Images report.
If we can only provide one stat (as a percent of adoption over time) which of these adoption rates is the most meaningful to measure: sites that have img[loading] (any value), sites that specifically have img[loading=lazy], sites that have any of the valid attribute values, or something else?
As for the frequency, we’re limited to monthly granularity.
SGTM! I think adoption of img[loading="lazy"] specifically would be most meaningful.
Ok great, I’ve filed this FR: https://github.com/HTTPArchive/httparchive.org/issues/205
Closing the loop, I’ve added this metric to the Images report: https://httparchive.org/reports/state-of-images?start=2020_07_01&end=latest&view=grid#imgLazy. Adoption is currently at 5%.