I never really understood with Google keeps pushing Brotli. LZ4 and ZStd both predate it by a couple of years, and seem to offer superior performance overall. Is it a Google-NIH thing?
They already had a very similar feature. _15 years ago_ with SDCH that got removed in 2018-- <a href="https://chromestatus.com/feature/5763176272494592" rel="nofollow noreferrer">https://chromestatus.com/feature/5763176272494592</a> <a href="https://daniel.haxx.se/blog/2008/09/09/shared-dictionary-compression-over-http/" rel="nofollow noreferrer">https://daniel.haxx.se/blog/2008/09/09/shared-dictionary-com...</a><p>Presumably they have some reason to think that this will be useful now when it wasn't before?
Makes me wonder if rsync-style differential download would be more generally useful than "shared dictionary for shared brotli". The implementation is letting you reference a previously downloaded artifact using a hash value, which is the same starting point.
Interesting two comment snippet from HN in 2018 about this: <a href="https://news.ycombinator.com/item?id=18720554">https://news.ycombinator.com/item?id=18720554</a>
>The client will store a hash of the uncompressed response<p>>client will add ...request header as well as a sec-available-dictionary: <SHA-256><p>loving it, another thing to track clients with