Hope: High-Speed Order-Preserving Encoder

Flash take: huh cute, these students wrote a paper formalizing something so simple I can have opinions about it

After browsing paper: holy cow that's awesome, where do these rocket scientists get their crazy ideas and when can we have it in Postgres

tl;dr: its basically a compression algorithm that produces a compressed stream whose sort order matches the sort order of the original uncompressed data

Minor complaint about the paper itself: all the datasets they picked can be expected to have very high redundancy. Would be nice to have included one approaching-random collection

Getting a 500 from Github.

Here is their paper:

https://arxiv.org/pdf/2003.02391.pdf

Is this like a succinct data structure? Can the compressed data be used in place without uncompression?

aren't there prior art? This would be super useful for a string sorting algorithm I am trying to implement in Julia