diff -r a0e322641810 -r 2d8e39919742 tests/test-evolve-phase-divergence.t
--- a/tests/test-evolve-phase-divergence.t	Fri Mar 30 19:37:06 2018 +0200
+++ b/tests/test-evolve-phase-divergence.t	Fri Mar 30 19:38:55 2018 +0200
@@ -106,6 +106,8 @@
 
   $ cd ../alice
   $ hg amend -m 'tweak a'
+
+XXX: pull should tell us how to see what is the new phase-divergent changeset
   $ hg pull ../public
   pulling from ../public
   searching for changes
@@ -128,8 +130,14 @@
     phase-divergent: 4d1169d82e47 (immutable precursor)
   
 
-Trying to see instability on public changeset
+
+XXX-Pulkit: Trying to see instability on public changeset
 
+XXX-Pulkit: this is not helpful
+
+XXX-Marmoute: public changeset "instable themself"
+XXX-Marmoute: I'm not sure if we store this information and it is useful to show it.
+XXX-Marmoute: We should maybe point the user toward `hg obslog` instead`
   $ hg evolve -r 4d1169d8 --list
   4d1169d82e47: modify a
   
@@ -137,6 +145,7 @@
 Understanding phasedivergence using obslog
 ------------------------------------------
 
+XXX: There must be mention of phase-divergence here
   $ hg obslog -r . --all
   @  98bb3a6cfe1a (2) tweak a
   |
@@ -146,6 +155,10 @@
 Solving the phase divergence using evolve command
 --------------------------------------------------
 
+(We do not solve evolution other than orphan by default because it turned out
+it was too confusing for users. We used to behave this way, but having multiple
+possible outcome to evolve end up scaring people)
+
   $ hg evolve
   nothing to evolve on current working copy parent
   (do you want to use --phase-divergent)
@@ -171,6 +184,9 @@
   hg revert --all --rev 98bb3a6cfe1a;
   hg commit --msg "phase-divergent update to 98bb3a6cfe1a"
 
+XXX: evolve should have mentioned that draft commit is just obsoleted in favour
+of public one. From the message it looks like a new commit is created.
+
   $ hg evolve --phase-divergent
   recreate:[2] tweak a
   atop:[1] modify a