diff --git a/rtv/content_generators.py b/rtv/content_generators.py
new file mode 100644
index 0000000..7fe2056
--- /dev/null
+++ b/rtv/content_generators.py
@@ -0,0 +1,135 @@
+from datetime import datetime, timedelta
+
+
+def clean(unicode_string):
+    """
+    Convert unicode string into ascii-safe characters.
+    """
+
+    return unicode_string.encode('ascii', 'replace').replace('\\', '')
+
+
+def humanize_timestamp(utc_timestamp, verbose=False):
+    """
+    Convert a utc timestamp into a human readable relative-time.
+    """
+
+    timedelta = datetime.utcnow() - datetime.utcfromtimestamp(utc_timestamp)
+
+    seconds = int(timedelta.total_seconds())
+    if seconds < 60:
+        return 'moments ago' if verbose else '0min'
+    minutes = seconds / 60
+    if minutes < 60:
+        return ('%d minutes ago' % minutes) if verbose else ('%dmin' % minutes)
+    hours = minutes / 60
+    if hours < 24:
+        return ('%d hours ago' % hours) if verbose else ('%dhr' % hours)
+    days = hours / 24
+    if days < 30:
+        return ('%d days ago' % days) if verbose else ('%dday' % days)
+    months = days / 30.4
+    if months < 12:
+        return ('%d months ago' % months) if verbose else ('%dmonth' % months)
+    years = months / 12
+    return ('%d years ago' % years) if verbose else ('%dyr' % years)
+
+
+class SubmissionGenerator(object):
+    """
+    Facilitates navigating through the comments in a PRAW submission.
+    """
+
+    def __init__(self, submission):
+
+        self.submission = submission
+
+    @staticmethod
+    def flatten_comments(submission):
+        """
+        Flatten a PRAW comment tree while preserving the nested level of each
+        comment via the `nested_level` attribute.
+        """
+
+        stack = submission[:]
+        for item in stack:
+            item.nested_level = 0
+
+        retval = []
+        while stack:
+            item = stack.pop(0)
+            nested = getattr(item, 'replies', None)
+            if nested:
+                for n in nested:
+                    n.nested_level = item.nested_level + 1
+                stack[0:0] = nested
+            retval.append(item)
+        return retval
+
+
+class SubredditGenerator(object):
+    """
+    Grabs a subreddit from PRAW lazily and store in an internal list for repeat
+    access.
+
+    params:
+        session (praw.Reddit): Active reddit connection
+        subreddit (str): Subreddit to connect to, None defaults to front page.
+    """
+
+    def __init__(self, reddit_session, subreddit=None):
+
+        self.r = reddit_session
+        self.r.config.decode_html_entities = True
+
+        if subreddit is None:
+            self._submissions = self.r.get_front_page(limit=None)
+        else:
+            self._submissions = self.r.get_subreddit(subreddit, limit=None)
+
+        self._submission_data = []
+
+    @staticmethod
+    def strip_praw_submission(sub):
+        """
+        Parse through a submission and return a dict with data ready to be
+        displayed through the terminal.
+        """
+
+        is_selfpost = lambda s: s.startswith('http://www.reddit.com/r/')
+
+        data = {}
+        data['title'] = clean(sub.title)
+        data['created'] = humanize_timestamp(sub.created_utc)
+        data['comments'] = '{} comments'.format(sub.num_comments)
+        data['score'] = '{} pts'.format(sub.score)
+        data['author'] = clean(sub.author.name)
+        data['subreddit'] = clean(sub.subreddit.url)
+        data['url'] = ('(selfpost)' if is_selfpost(sub.url) else clean(sub.url))
+
+        return data
+
+    def get(self, index, n_cols):
+        """
+        Grab the `i`th submission, with the title field formatted to fit inside
+        of a window of width `n`
+        """
+
+        assert(index >= 0)
+
+        while index >= len(self._submission_data):
+            data = self._strip_praw_submission(self._submissions.next())
+            self._submission_data.append(data)
+
+        # Modifies the original original dict, faster than copying
+        data = self._submission_data[index]
+        data['split_title'] = textwrap.wrap(data['title'], width=n_cols)
+        data['n_rows'] = len(data['split_title']) + 3
+
+        return data
+
+    def iterate(self, index, n_cols):
+
+        while True:
+            yield self.get(index, n_cols)
+            index += 1
\ No newline at end of file
diff --git a/rtv/subreddit.py b/rtv/subreddit.py
index 226e138..2f8375e 100644
--- a/rtv/subreddit.py
+++ b/rtv/subreddit.py
@@ -2,22 +2,44 @@ import praw
 import textwrap
 import curses
 
-from utils import humanize_timestamp, flatten_tree, clean
+from content_generators import SubredditGenerator
 
-def strip_submission(sub):
-    "Grab info from a PRAW submission and prep for display."
+class SubredditViewer(object):
 
-    out = {}
-    out['title'] = clean(sub.title)
-    out['created'] = humanize_timestamp(sub.created_utc, long=False)
-    out['comments'] = '{} comments'.format(sub.num_comments)
-    out['score'] = '{} pts'.format(sub.score)
-    out['author'] = clean(sub.author.name)
-    out['subreddit'] = clean(sub.subreddit.url)
-    out['url'] = ('(selfpost)' if sub.url.startswith('http://www.reddit.com/r/')
-                  else clean(sub.url))
+    def __init__(self, stdscr):
+        self.stdscr = stdscr
 
-    return out
+    def loop(self):
+
+        while True:
+            cmd = self.stdscr.getch()
+
+            # Move cursor up one submission
+            if cmd == curses.KEY_UP:
+                self.move_cursor(-1)
+
+            # Move cursor down one submission
+            elif cmd == curses.KEY_DOWN:
+                self.move_cursor(1)
+
+            # View submission
+            elif cmd in (curses.KEY_RIGHT, ord(' ')):
+                pass
+
+            # Enter edit mode to change subreddit
+            elif cmd == ord('/'):
+                pass
+
+            # Refresh page
+            elif cmd in (curses.KEY_F5, ord('r')):
+                pass
+
+            # Quit
+            elif cmd == ord('q'):
+                pass
+
+            else:
+                curses.beep()
 
 def draw_submission(win, data, top_down=True):
     "Draw a submission in the given window."
@@ -49,46 +71,26 @@ def draw_submission(win, data, top_down=True):
         win.addnstr(row, 1, '{author} {subreddit}'.format(**data), n_cols)
 
 
-class SubmissionGenerator(object):
-    """
-    Grab submissions from PRAW lazily and store in an internal list for repeat
-    access.
-    """
+def focus_submission(win):
+    "Add a vertical column of reversed background on left side of the window"
 
-    def __init__(self):
+    n_rows, n_cols = win.getmaxyx()
+    for row in xrange(n_rows):
+        win.chgat(row, 0, 1, curses.A_REVERSE)
 
-        self.r = praw.Reddit(user_agent='reddit terminal viewer (rtv) v0.0')
-        self.r.config.decode_html_entities = True
 
-        self._submissions = self.r.get_front_page(limit=None)
-        self._submission_data = []
-
-    def get(self, index, n_cols):
-
-        assert(index >= 0)
-
-        while index >= len(self._submission_data):
-            data = strip_submission(self._submissions.next())
-            self._submission_data.append(data)
-
-        # Modifies the original original dict, faster than copying
-        out = self._submission_data[index]
-        out['split_title'] = textwrap.wrap(out['title'], width=n_cols)
-        out['n_rows'] = len(out['split_title']) + 3
-
-        return out
-
-    def iterate(self, index, n_cols):
-
-        while True:
-            yield self.get(index, n_cols)
-            index += 1
+def unfocus_submission(win):
+    "Clear the vertical column"
 
+    n_rows, n_cols = win.getmaxyx()
+    for row in xrange(n_rows):
+        win.chgat(row, 0, 1, curses.A_NORMAL)
 
 
 def draw_subreddit(stdscr):
 
-    generator = SubmissionGenerator()
+    r = praw.Reddit(user_agent='reddit terminal viewer (rtv) v0.0')
+    generator = SubredditGenerator(r)
 
     main_window = stdscr.derwin(1, 0)
     main_window.erase()
@@ -99,6 +101,7 @@ def draw_subreddit(stdscr):
         n_rows = min(max_rows-current_row, data['n_rows'])
         sub_window = main_window.derwin(n_rows, max_cols, current_row, 0)
         draw_submission(sub_window, data)
+        focus_submission(sub_window)
         sub_window.refresh()  # Debugging
         current_row += n_rows + 1
         if current_row >= max_rows:
diff --git a/rtv/utils.py b/rtv/utils.py
index fad4a79..de085ae 100644
--- a/rtv/utils.py
+++ b/rtv/utils.py
@@ -1,4 +1,3 @@
-from datetime import datetime, timedelta
 import curses
 from contextlib import contextmanager
 
@@ -52,52 +51,3 @@ def curses_session():
             curses.nocbreak()
             curses.endwin()
 
-def humanize_timestamp(utc_timestamp, long=True):
-    """
-    Convert a utc timestamp into a human readable time relative to now.
-    """
-    timedelta = datetime.utcnow() - datetime.utcfromtimestamp(utc_timestamp)
-    seconds = int(timedelta.total_seconds())
-    if seconds < 60:
-        return 'moments ago' if long else '0min'
-    minutes = seconds / 60
-    if minutes < 60:
-        return '%d' % minutes + (' minutes ago' if long else 'min')
-    hours = minutes / 60
-    if hours < 24:
-        return '%d' % hours + (' hours ago' if long else 'hr')
-    days = hours / 24
-    if days < 30:
-        return '%d' % days + (' days ago' if long else 'day')
-    months = days / 30.4
-    if months < 12:
-        return '%d' % months + (' months ago' if long else 'month')
-    years = months / 12
-    return '%d' % years + (' years ago' if long else 'yr')
-
-
-def flatten_tree(tree):
-    """
-    Flatten a PRAW comment tree while preserving the nested level of each
-    comment via the `nested_level` attribute.
-    """
-
-    stack = tree[:]
-    for item in stack:
-        item.nested_level = 0
-
-    retval = []
-    while stack:
-        item = stack.pop(0)
-        nested = getattr(item, 'replies', None)
-        if nested:
-            for n in nested:
-                n.nested_level = item.nested_level + 1
-            stack[0:0] = nested
-        retval.append(item)
-    return retval
-
-
-def clean(unicode_string):
-    "Convert unicode string into ascii-safe characters."
-    return unicode_string.encode('ascii', 'replace').replace('\\', '')
\ No newline at end of file