Column headings say NEW and WHEN, not UNREAD and TIME

The chat picker's unread count is headed NEW and its last-message age
WHEN; the channel list's unread count is NEW; the live feed's timestamp
lane is WHEN. Short words over a badge and an age, which UNREAD outgrew
by a cell and had to borrow the gap for. The command line's columns keep
their names: they are a documented contract of their own.

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>

meshterm/tools/chat.py

@@ -695,7 +695,7 @@ _BADGE_WIDTH = 5
 _AGE_WIDTH = 3
 #: Floor for the conversation lane. The header word is ``CONVERSATION`` (12 cells) and the
 #: lane carries a two-cell gap after it, so anything narrower than this runs the heading
-#: straight into ``UNREAD`` on a mesh whose every name is ``Bob``.
+#: straight into ``NEW`` on a mesh whose every name is ``Bob``.
 _LABEL_MIN = 12
 #: Ceiling for it. The lane is sized to the longest name it actually holds (see
 #: :func:`_lanes`), and this is where that stops paying: one very long name would otherwise
@@ -767,8 +767,10 @@ def _picker_header(lanes: _Lanes, width: int) -> str:
 
     The indent covers the select screen's pointer column (2 cells, drawn on choice rows but
     not separators) plus the marker lane, so each header lands exactly over its column.
-    UNREAD borrows its lane's trailing gap — the badge lane itself is one cell too narrow
-    for the word — which still leaves a space before the age column.
+    ``NEW`` counts the unread messages and ``WHEN`` is the last message's age — short words,
+    because the lanes under them are a badge and an age, and a heading longer than its lane
+    runs into the next one's. ``WHEN`` still borrows its lane's trailing gap (the age lane is
+    three cells), which leaves a space before the preview.
 
     Resolved against the render width (the header row is pinned, so it must stay one row):
     on a terminal too narrow for the whole line, ``LAST MESSAGE`` gives its cells back a
@@ -778,8 +780,8 @@ def _picker_header(lanes: _Lanes, width: int) -> str:
     return column_header(
         [
             Lane("CONVERSATION", lanes.label + 2),
-            Lane("UNREAD", _BADGE_WIDTH + 2),
-            Lane("TIME", _AGE_WIDTH + 2),
+            Lane("NEW", _BADGE_WIDTH + 2),
+            Lane("WHEN", _AGE_WIDTH + 2),
             Lane(("LAST MESSAGE", "LAST MSG", "LAST")),
         ],
         width,

meshterm/ui/channels.py

@@ -540,7 +540,7 @@ def _lanes_header(slot_w: int, name_w: int, scope_w: int, width: int) -> str:
             Lane("SLOT", slot_w + 2 + cell_len(channel_glyph("Public", None)) + 1),
             Lane("CHANNEL", name_w + 2),
             *([Lane("SCOPE", scope_w + 2)] if scope_w else []),
-            Lane("UNREAD", _BADGE_WIDTH + 2),
+            Lane("NEW", _BADGE_WIDTH + 2),
             # Right-aligned, because the values under them are: an age and a count are
             # padded to the right edge of their lane, so a left-aligned label would sit
             # off the digits it names.

meshterm/ui/livefeed_screen.py

@@ -588,7 +588,7 @@ class LiveFeedScreen(Screen):
 
         Laid out lane for lane against :meth:`_feed_row`, the pointer column included, so
         every label sits over the values it names. Nothing here sorts — the feed is a
-        stream, and its one order (newest first) is what the ``TIME`` lane reads down — so
+        stream, and its one order (newest first) is what the ``WHEN`` lane reads down — so
         no column carries the contact list's sort triangle or its lit active-column style;
         these are signposts, not controls. On a terminal too narrow for the class label
         the header drops its ``CLASS`` too, leaving the icon lane unlabelled rather than
@@ -600,7 +600,7 @@ class LiveFeedScreen(Screen):
         down the whole column.
         """
         header = Text("  ", style="muted")  # the pointer lane
-        header.append(fit_cells("TIME", _TIME_LANE))
+        header.append(fit_cells("WHEN", _TIME_LANE))
         header.append(fit_cells("CLASS", _ICON_LANE + _FEED_CLASS_WIDTH + _LANE_GAP))
         header.append(fit_cells("SUBJECT", lanes.subject + _LANE_GAP))
         header.append(fit_cells("SCOPE", _SCOPE_LANE))

meshterm/ui/menus.py

@@ -364,7 +364,7 @@ class Lane:
             terminal. Only a lane the line actually overruns is ever shortened.
         width: Cells the label is padded to — the lane's own width *plus* the gap before
             the next lane, so a label wider than its column absorbs that gap instead of
-            shifting every lane after it right (``UNREAD`` over a narrower badge lane).
+            shifting every lane after it right (``WHEN`` over the chat picker's 3-cell ages).
             Zero (the default) for a trailing lane, which just runs to the edge.
     """
 

tests/test_chat.py

@@ -1807,6 +1807,7 @@ async def test_picker_pins_its_column_header_over_the_group_heading(
     visible, above, _below = frame._visible_slice(screen, screen.render_body(72), 8)
     top = [ansi.sub("", row).strip() for row in visible[:2]]
     assert top[0].startswith("CONVERSATION") and top[0].endswith("LAST MESSAGE")
+    assert top[0].split()[1:3] == ["NEW", "WHEN"]  # short words over a badge and an age
     assert top[1] == "── 👤 Direct ──"
     assert above is True
     # Too narrow for the whole line, the trailing label shortens — never wraps. The

tests/test_livefeed.py

@@ -263,14 +263,14 @@ def test_livefeed_column_header_sits_over_the_lanes_it_names() -> None:
     """The header names each lane, and every label lands on the values beneath it."""
     screen = _screen(seed=[_obs(node="3d63", kind="telemetry", snr=12.8, rssi=-105.0)])
     header, row = _stripped(screen.render_body(90))[:2]
-    assert header.split() == ["TIME", "CLASS", "SUBJECT", "SCOPE", "SNR", "RSSI"]
+    assert header.split() == ["WHEN", "CLASS", "SUBJECT", "SCOPE", "SNR", "RSSI"]
     # Measured in display cells, not characters — the class icon is one character wide
     # but two cells, so a character index would report every later lane one column early.
     # The clock is read off the row itself: a literal needle would only match during the
     # minute it was written in.
     stamp = re.search(r"\d\d:\d\d:\d\d", row)
     assert stamp is not None
-    assert _col(header, "TIME") == _col(row, stamp.group(0))
+    assert _col(header, "WHEN") == _col(row, stamp.group(0))
     assert _col(header, "CLASS") == _col(row, "telemetry") - _ICON_LANE
     assert _col(header, "SUBJECT") == _col(row, "Hub")
     # Each reading's label starts where its lane does, like every other label.
@@ -285,7 +285,7 @@ def test_livefeed_column_header_is_pinned_and_carries_no_sort_cue() -> None:
     screen = _screen(seed=[_obs(node=f"n{i}", age_s=i) for i in range(40)])
     screen.note_viewport(12)
     lines = _stripped(screen.render_body(100))
-    assert "TIME" in lines[0]
+    assert "WHEN" in lines[0]
     assert not any(mark in lines[0] for mark in ("▲", "▼"))  # nothing here sorts
     screen.handle("end")  # the list's last row — Back
     screen.handle("up")  # …and one up from it, the oldest packet: the window scrolls
@@ -475,7 +475,7 @@ def test_livefeed_windows_inside_the_fixed_screen() -> None:
     lines = screen.render_body(100)
     assert len(lines) <= 24  # column header + feed window == the viewport, never more
     body = _plain(lines)
-    assert "TIME" in body  # the pinned column header is still there
+    assert "WHEN" in body  # the pinned column header is still there
     assert "↓" in body and "more" in body  # hidden feed rows are counted below
 
 

tests/test_scope_display.py

@@ -218,7 +218,7 @@ def test_the_picocalc_feed_names_the_class_without_an_icon() -> None:
     try:
         screen = _feed(_scoped("harbour"))
         header, row = _stripped(screen.render_body(53))[:2]
-        assert header.split()[:2] == ["TIME", "CLASS"]
+        assert header.split()[:2] == ["WHEN", "CLASS"]
         assert _col(header, "CLASS") == _col(row, "chan text")
     finally:
         set_platform(REGULAR)
@@ -411,12 +411,12 @@ def _channel_header(scopes: dict[int, str]) -> str:
 
 def test_the_channel_list_draws_scope_only_when_a_channel_has_one() -> None:
     """A column of blanks says nothing; one scoped channel brings the lane back."""
-    assert _channel_header({}).split() == ["SLOT", "CHANNEL", "UNREAD", "LAST", "MSGS", "ACTIVITY"]
+    assert _channel_header({}).split() == ["SLOT", "CHANNEL", "NEW", "LAST", "MSGS", "ACTIVITY"]
     assert _channel_header({1: "harbour"}).split() == [
         "SLOT",
         "CHANNEL",
         "SCOPE",
-        "UNREAD",
+        "NEW",
         "LAST",
         "MSGS",
         "ACTIVITY",