diff --git a/Data-structure-survey.md b/Data-structure-survey.md
index 1b20895..7393f92 100644
--- a/Data-structure-survey.md
+++ b/Data-structure-survey.md
@@ -57,6 +57,10 @@ Fixed-size array stack
 
 ## notes
 
+Used as object caches
+
+Ondrej: will be gone after the netmgr refactoring and replaced with per-thread object pools/lists
+
 Locked.
 
 
@@ -81,6 +85,7 @@ There's a min-heap in libuv, implemented as a binary tree. The
 standard algorithms (as in `isc/heap`) use an array which should have
 better performance.
 
+Ondrej: The usage in isc_timer has been already removed in the loopmgr branch
 
 # `isc/ht`
 
@@ -120,6 +125,11 @@ Case-sensitive and case-insensitive variants, chosen at run-time.
 Uses chaining, so it is heavy on indirection and heavy on allocation.
 Modern hash tables use open addressing.
 
+Ondrej: there's work to do on the isc_ht:
+
+1. make header-only variant of the hashtable, so we can unify the usage across the codebase (e.g. no extra allocations when storing item)
+2. there's a branch, where I implemented robin-hood hashing, but I never finished because of 1)
+3. the resizing was added only recently, we still miss shrinking (should be easy enough to do)
 
 # `isc/pool`
 
@@ -138,6 +148,7 @@ This is a bit different from the lock pools in `rbtdb`, because
 `isc/pool` returns a random item from teh pool, whereas `rbtdb`
 chooses `locknum` deterministically based on the node's hash value.
 
+Ondrej: this is already gone on main
 
 # `isc/radix`
 
@@ -181,6 +192,7 @@ Ought to be unnecessary. Might be worth cleaning up as part of some
 general parser refactoring (`isc/lex` could probably be made faster
 using `re2c`).
 
+Ondrej: Using [rage](http://www.colm.net/open-source/ragel/) is on my TODO list for too long...
 
 # `isc/taskpool`
 
@@ -200,6 +212,7 @@ like that was the plan when `isc/pool` was added).
 There appears to be a bug in `isc_taskpool_expand()` when it fails to
 create a new task: the existing taskpool will be left empty.
 
+Ondrej: this is already gone on main
 
 # `isccc`