diff --git a/hypothesis/__init__.py b/hypothesis/__init__.py
index e5d6ab4..d172b3a 100644
--- a/hypothesis/__init__.py
+++ b/hypothesis/__init__.py
@@ -1,28 +1,54 @@
 from functools import lru_cache
-from kython import listdir_abs, json_load, JSONType
+from kython import listdir_abs
 from typing import Dict, List, NamedTuple, Optional
+from pathlib import Path
+import json
 from pytz import UTC
 from datetime import datetime
 import os
 
+from kython import group_by_key
+from kython.misc import the
+
 _PATH = '/L/backups/hypothesis/'
 
 class Entry(NamedTuple):
     dt: datetime
     summary: str
-    content: str
+    content: Optional[str] # might be none if for instance we just marked page with tags. not sure if we want to handle it somehow separately
     link: str
     eid: str
     annotation: Optional[str]
     context: str
     tags: List[str]
 
+Url = str
+
+class Page(NamedTuple):
+    highlights: List[Entry]
+
+    @property
+    # @lru_cache()
+    def url(self):
+        return the(h.url for h in self.highlights)
+
+    @property
+    # @lru_cache()
+    def title(self):
+        return the(h.title for h in self.highlights)
+
+    @property
+    # @lru_cache()
+    # TODO shit. can't be cached because of self, wtf??? how to get around it??
+    def dt(self):
+        return min(h.dt for h in self.highlights)
+
+
 # TODO guarantee order?
 def _iter():
     last = max(listdir_abs(_PATH))
-    j: JSONType
-    with open(last, 'r') as fo:
-        j = json_load(fo)
+    with Path(last).open() as fo:
+        j = json.load(fo)
     for i in j:
         dts = i['created']
         title = ' '.join(i['document']['title'])
@@ -30,7 +56,7 @@ def _iter():
         if selectors is None:
             # TODO warn?...
             selectors = []
-        content = None
+        content: Optional[str]
         for s in selectors:
             if 'exact' in s:
                 content = s['exact']
@@ -53,9 +79,21 @@ def _iter():
         )
 
 
+def get_pages() -> List[Page]:
+    grouped = group_by_key(_iter(), key=lambda e: e.link)
+    pages = []
+    for link, group in grouped.items():
+        group = list(sorted(group, key=lambda e: e.dt))
+        pages.append(Page(highlights=group))
+    pages = list(sorted(pages, key=lambda p: p.dt))
+    # TODO fixme page tag??
+    return pages
+
+
 def get_entries():
     return list(_iter())
 
+
 def get_todos():
     def is_todo(e: Entry) -> bool:
         if any(t.lower() == 'todo' for t in  e.tags):
@@ -64,3 +102,10 @@ def get_todos():
             return False
         return e.annotation.lstrip().lower().startswith('todo')
     return list(filter(is_todo, get_entries()))
+
+def _main():
+    for page in get_pages():
+        print(page)
+
+if __name__ == '__main__':
+    _main()