Shots: get an animated gif from a Wayback Machine archive
github.com
github.com
Of course, they probably store their collection in huge WARC files not clearly deliniated by domain. I wonder how heavy it would be to generate WARC subsets on-the-fly, given that they already have a good index for the data?
[0] http://digitalpreservation.gov/formats/fdd/fdd000236.shtml [1] http://www.iso.org/iso/catalogue_detail.htm?csnumber=44717