benchmarking
benchmarking copied to clipboard
s/attic/borgbackup/
I found your benchmarks and noticed you refer to borgbackup 1.1.0b6 as "attic".
While attic is the project borg was forked from, it has developped way beyond attic meanwhile (and we're not finished yet). So maybe rather refer to it as borgbackup. While borg can import attic repos, the repo format is intentionally different (changed magic value).
Attic hasn't developped at all since the fork (still at 0.16) and has known/unfixed issues, so guess you do not need to include it in your benchmark any more.
Think of 'Attic' as a category including both the original Attic and the borgbackup fork that share the same basic design. I know the original Attic hasn't been maintained for 2 years, but borgbackup is still a variation of it, not a completely new one.
Considering borgbackup repo meanwhile has almost 8x the changesets of the attic repo I think my wish for it being called by its own name and not the name of the meanwhile rather stale (dead?) parent project isn't far fetched.
You don't benchmark categories, you benchmark software. If the software you tested is Borg you should call it "Borg" or "BorgBackup". It's not Attic, especially after all this time and especially after Attic has gone unmaintained. In addition to being rude to the Borg author, this title just makes this comparison look less relevant and interesting than it actually is.