Introduce concept for cluster search
diff --git a/lib/Krawfish/Posting/Bundle.pm b/lib/Krawfish/Posting/Bundle.pm
index 272da8b..cb8616b 100644
--- a/lib/Krawfish/Posting/Bundle.pm
+++ b/lib/Krawfish/Posting/Bundle.pm
@@ -7,8 +7,8 @@
 # TODO:
 #   This is quite similar to K::P::Group
 
-# This is a container class for multiple
-# Krawfish::Posting objects
+# This is a container class for multiple Krawfish::Posting objects,
+# used for (among others) sorting.
 
 # Constructor
 sub new {
diff --git a/lib/Krawfish/Posting/Match.pm b/lib/Krawfish/Posting/Match.pm
index 299a13b..6062e43 100644
--- a/lib/Krawfish/Posting/Match.pm
+++ b/lib/Krawfish/Posting/Match.pm
@@ -1,10 +1,13 @@
 package Krawfish::Posting::Match;
 use parent 'Krawfish::Posting';
 use Krawfish::Util::String qw/squote/;
-use JSON::XS;
 use warnings;
 use strict;
 
+
+# Matches are returned from searches and can be enriched
+# with various information
+
 # Get or set field to match
 sub fields {
   my $self = shift;
@@ -36,7 +39,22 @@
   };
 };
 
+sub sorting_criteria;
 
+sub snippet;
+
+sub segment_id;
+
+sub match_id;
+
+
+# serialize to koralquery
+sub to_koral_query {
+  ...
+};
+
+
+# Stringification
 sub to_string {
   my $self = shift;
   my $str = '[';