Modern papers on search indices often contain a statement that inverted indexes (posting lists) are categorically superior to signature files (bloom filters). Here are some examples from papers published in 2016:
While this [signature file] technique provides a relatively low computation overhead, studies by Zobel et al. [1998] have shown that inverted files significantly outperform signature files.
Inverted indexes have been benchmarked as the most generalisable, and well performing structure (Zobel et al., 1998)
Every paper seems to cite Zobel et al., Inverted files versus signature files for text indexing.
However, if I'm reading Zobel et al. correctly, the argument they make isn't fundamental (e.g., an asymptotic bound or an information theoretic limit). Rather, the argument seems to be, given signature files implemented with techniques X, Y, and Z compared to inverted indices implemented with techniques A, B, and C, and the current technology of the day (disks with very high seek/access overhead), inverted indices are superior because they require fewer seeks and there are faster.
Is there a more recent comparison that compares these techniques on SSD, NVMe, or RAM, or is there a more recent comparison that looks at "new" techniques that have been invented since 1998?
