diff options
author | Joey Hess <joeyh@joeyh.name> | 2018-03-06 15:26:57 -0400 |
---|---|---|
committer | Joey Hess <joeyh@joeyh.name> | 2018-03-06 15:26:57 -0400 |
commit | 75f5160c29608a102084965456716d94f4c1abcf (patch) | |
tree | 1206be39a7911a94ea7ff82a0be1df9603e19b34 | |
parent | 18395df2e1007d872cf63d895fe6343e7e4dd473 (diff) |
devblog
-rw-r--r-- | doc/devblog/day_486__time_to_ditch_rsync.mdwn | 14 |
1 files changed, 14 insertions, 0 deletions
diff --git a/doc/devblog/day_486__time_to_ditch_rsync.mdwn b/doc/devblog/day_486__time_to_ditch_rsync.mdwn new file mode 100644 index 000000000..bd7cf0d4b --- /dev/null +++ b/doc/devblog/day_486__time_to_ditch_rsync.mdwn @@ -0,0 +1,14 @@ +I'm excited by this new design +[[todo/accellerate_ssh_remotes_with_git-annex-shell_mass_protocol]]. + +git-annex's use of rsync got transfers over ssh working quickly early on, +but other than resuming interrupted transfers, using rsync doesn't really +gain git-annex much, since annexed objects don't change over time. And +rsync has always involved a certian amount of overhead that a custom +protocol would avoid. + +It's especially handy that such a protocol was already developed for +`git-annex p2p` when using tor. I've not heard of a lot of people using that +feature (but maybe people who do have reason not to talk about it), but +it's a good solid thing, implemented very generically with a free monad, +and reusing it for git-annex-shell would be great. |