Not sure about stock tickers, but for domains it seems to pull an image from the <head> of a page. Looks like it's trying to find a favicon or the largest application image listed. Seems to skip some things like twitter's meta image tag, so maybe there's a preset list of tags it'll search for?
Comments
What is the source for the logos? They are sometimes jpg with bad compression artifacts.
Not sure about stock tickers, but for domains it seems to pull an image from the <head> of a page. Looks like it's trying to find a favicon or the largest application image listed. Seems to skip some things like twitter's meta image tag, so maybe there's a preset list of tags it'll search for?