summaryrefslogtreecommitdiff
path: root/README
diff options
context:
space:
mode:
authorPiotr Szarmanski2023-08-03 00:31:49 +0200
committerPiotr Szarmanski2023-08-03 00:31:49 +0200
commit552dfc187707185940cb11c31e66e47ca3efacca (patch)
tree27715de1728a2d3cdd4c04af48563e620625219a /README
Init.
Diffstat (limited to 'README')
-rw-r--r--README28
1 files changed, 28 insertions, 0 deletions
diff --git a/README b/README
new file mode 100644
index 0000000..4e2a626
--- /dev/null
+++ b/README
@@ -0,0 +1,28 @@
+ybackup is an ERIS (http://eris.codeberg.page/)-based backup/archive system. It
+takes a directory as an input, and packs all files in that hierarchy into ERIS
+data. Each file receives an individual URN, which is recorded along the file
+permissions, mtime and the filename in an index. The index reproduces the
+file-hierarchy, by representing directories, files and symlinks as objects.
+
+Features:
+
++ Thanks to ERIS, the data is automatically deduplicated.
+
++ Convergent encryption by default (i.e. the files are encrypted based on their content)
+
++ An additional secret can be used for encryption to protect against attacks on convergent encryption.
+
++ Backups are specified by URNs and so are specific files within them, meaning
+that you could e.g. share a file with anyone without much effort.
+
++ It's transport and storage agnostic and could work on any medium permitting read/write access to 1/32kb blocks of binary data.
+
++ Incremental backups.
+
+Anti-features:
+
++ Backups cannot be trivially removed, which means that storage space
+consumption will increase as more backups are made;
+
+
+