3 ms·
Wouldn't the browser do a HEAD first? Seems like you could also use uniquely generated ETAGS as cookies if it does. Which would be more effective with favicons
by tyingq 6y ago
Wouldn't the browser do a HEAD first? Seems like you could also use uniquely generated ETAGS as cookies if it does. Which would be more effective with favicons than the general case, given the comments about how browsers cache them.
- edoceo 6y agoBrowser does NOT do the HEAD request first. Only GET
- laurencerowe 6y agoThe browser would presumably send the ETag in an If-None-Match in the GET request though.
- evan_ 6y agoETag fingerprinting has been around for awhile, KissMetrics got sued for doing it in 2012. I don’t know if there’s a mitigation per se or if it’s just the threat of a lawsuit keeps people honest. Regardless, clearing the cache or using a different profile defeats it. https://www.google.com/amp/s/www.research-live.com/amp-page.html%3fid=4008518&name=kissmetrics-settles-etag-tracking-lawsuit https://www.google.com/amp/s/www.research-live.com/amp-page....
- tyingq 6y agoThat's the point I was making. Since favicons have their own cache that isn't cleared when the user clears the main cache, ETags would work well there. And would be less complex than the file scheme in the post.
- esprehn 6y agoMost browsers have moved, or are moving, to cache partitioning to mitigate this: https://www.chromestatus.com/feature/5730772021411840 https://www.chromestatus.com/feature/5730772021411840 Safari shipped this a long time ago, and all other browsers are following that path. It's unfortunate because it means shared CDNs become ineffective.