From 438ff5001311c47d48b1752408d1505532a7a1a4 Mon Sep 17 00:00:00 2001 From: Ilya_Shlyakhter Date: Tue, 19 Mar 2019 20:40:50 +0000 Subject: [PATCH] Added a comment --- .../comment_2_5528a5dd5f8d7746b175552a077ba313._comment | 8 ++++++++ 1 file changed, 8 insertions(+) create mode 100644 doc/todo/simpler__44___trusted_export_remotes/comment_2_5528a5dd5f8d7746b175552a077ba313._comment diff --git a/doc/todo/simpler__44___trusted_export_remotes/comment_2_5528a5dd5f8d7746b175552a077ba313._comment b/doc/todo/simpler__44___trusted_export_remotes/comment_2_5528a5dd5f8d7746b175552a077ba313._comment new file mode 100644 index 0000000000..850ea00103 --- /dev/null +++ b/doc/todo/simpler__44___trusted_export_remotes/comment_2_5528a5dd5f8d7746b175552a077ba313._comment @@ -0,0 +1,8 @@ +[[!comment format=mdwn + username="Ilya_Shlyakhter" + avatar="http://cdn.libravatar.org/avatar/1647044369aa7747829c38b9dcc84df0" + subject="comment 2" + date="2019-03-19T20:40:49Z" + content=""" +\"A remote's implementation could do the same thing\" -- right, but if git-annex implements this (through the current special remote protocol) then one implementation covers all export-supporting remotes. Also, implementing special remotes would become simpler, since they could just implement the export part of the protocol, and the key-value store then gets implemented in terms of that. But the main plus would be the uniformity: export remotes would be key-value remotes with an added ability (to store files in human-readable locations), rather than a separate untrustworthy species. +"""]]