diff --git a/doc/bugs/adb_replaces_symlinks_by_plain_text_files/comment_2_f5db61aa7efb7a916d0ffebd563cc2c7._comment b/doc/bugs/adb_replaces_symlinks_by_plain_text_files/comment_2_f5db61aa7efb7a916d0ffebd563cc2c7._comment
new file mode 100644
index 0000000000..d042b41f10
--- /dev/null
+++ b/doc/bugs/adb_replaces_symlinks_by_plain_text_files/comment_2_f5db61aa7efb7a916d0ffebd563cc2c7._comment
@@ -0,0 +1,18 @@
+[[!comment format=mdwn
+ username="amerlyq+annex@12121d70a602f094228689a0a24d348d478a8af1"
+ nickname="amerlyq+annex"
+ avatar="http://cdn.libravatar.org/avatar/3d63c9f436b45570d45bd003e468cbd3"
+ subject="comment 2"
+ date="2022-01-04T13:10:13Z"
+ content="""
+Probably skipping symlinks for both import/export is for the best.
+
+My personal expectations (pipe dream?) are to export symlinks as real files -- but only compare during import and warn if different.
+
+Because if I see file/symlink in some repo -- it's placed there for a reason, and after sync I expect to find all the expected files in their folders (even if through multiple copies). 
+Silently ignoring of symlinks forces me to remember to check that whatever I want to sync does not contain them.
+Otherwise I may be in a very delicate situation, if files needed for work were not synced to phone and I have no access to PC at that moment ;)
+
+On the other hand, I remember literally months I spent debugging and configuring samba, and reading their history of symlinks implementations.
+It contains so much frustrating corner cases, that banning symlinks completely isn't a bad idea, actually.
+"""]]