What does Dense Data Workbenches conclude, in numbers, about “Scan Paths and the Skeleton of a Table: Column Order, Alignment, Frozen Panes and Density”?
Answer: The skeleton is a cost paid on every row. Ordering columns by task shortens the eyes' travel per row by 27.7% under this chapter's task weights, at the price of slowing minority tasks, which need views of their own. Right-aligned numbers, fixed decimals, comma separators and tabular figures let 43.3% of neighbouring rows be compared without reading a digit. Freezing the header costs 4.8% of the viewport height and puts all 1,200 rows on screen with it; freezing the first column costs 6.7% of the width and keeps every row's identity in view. Zebra striping most likely does little, so little that a team of four could hardly detect it within a week, so it is off by default and the testing budget goes to column order and row height. Row height trades screens against wrong clicks, and tables with in-row actions keep rows at 32 pixels or more. Colour comes only once the skeleton stands, and that is the next chapter.
Scope of the numbers: The book is not accounting, audit or legal advice. Marked boxes separate published evidence, runnable calculations, sources and the author's hypotheses, each stating what would refute it; the source studied is the reporting system of the Tang finance minister Liu Yan, given in paraphrase.
Data table
The numbers below are printed verbatim by the chapter's Python program and can be reproduced by running it. standard-library Python onlyprinted values that match the answer: 1200, 4.8, 6.7
| sticky header False: rows on first screen 25, on later screens 26, screens per day 47 | |
| rows that share a screen with the header | 25 of 1200 (2.1%) |
| sticky header True : rows on first screen 25, on later screens 25, screens per day 48 | |
| rows that share a screen with the header | 1200 of 1200 (100.0%) |
| header height as share of the table viewport | 4.8% |
| nine visible columns table | 1016 px, viewport 1680 px, overflow 0 px |
| plus ten audit columns table | 2096 px, viewport 1680 px, overflow 416 px |
| columns beyond the first screen 4; fully scrolled the view moves 416 px, more than order_id's 112 px | |
| freezing order_id costs 6.7% of the viewport width and keeps every row labelled |
Show the full program
# Chapter 2. p23: Frozen header row and frozen first column: what they cost and what they buy.
# Geometry for the fictional Harbourbridge desk: a 1920x1080 screen, browser and app chrome take the top
# 240 px (assumption), so the table viewport is 840 px tall; header 40 px; row pitch 32 px: 25 rows, as p10.
# Horizontal: a 240 px side menu leaves 1,680 px; with the four hidden audit columns shown the table is wider.
import math
ROWS, VIEW_H, HEAD, PITCH = 1200, 840, 40, 32
VIEW_W, MENU = 1920, 240
VISIBLE = [112, 96, 128, 80, 96, 104, 104, 88, 208] # p20 widths, order_id first
# ten audit columns (fx rate, fee, sku, warehouse, tax and five more); only their widths matter here
AUDIT = [96, 96, 88, 120, 104, 136, 112, 96, 128, 104]
def rows_on_screen(sticky):
body = VIEW_H - HEAD
first = body // PITCH
later = first if sticky else VIEW_H // PITCH
return first, later
if __name__ == "__main__":
for sticky in (False, True):
first, later = rows_on_screen(sticky)
screens = 1 + math.ceil((ROWS - first) / later)
with_head = first if not sticky else ROWS
print(" sticky header {:<5s}: rows on first screen {}, on later screens {}, screens per day {}"
.format(str(sticky), first, later, screens))
print(" rows that share a screen with the header: {} of {} ({:.1%})".format(
with_head, ROWS, with_head / ROWS))
print(" header height as share of the table viewport: {:.1%}".format(HEAD / VIEW_H))
avail = VIEW_W - MENU
for label, widths in (("nine visible columns", VISIBLE), ("plus ten audit columns", VISIBLE + AUDIT)):
total = sum(widths)
print(" {:<23s} table {:5d} px, viewport {} px, overflow {:5d} px".format(
label, total, avail, max(0, total - avail)))
if total > avail:
right_edge = [sum(widths[:i + 1]) for i in range(len(widths))]
offscreen = sum(1 for e in right_edge if e > avail)
print(" columns beyond the first screen {}; fully scrolled the view moves {} px,"
" more than order_id's {} px".format(offscreen, total - avail, widths[0]))
print(" freezing order_id costs {:.1%} of the viewport width and keeps every row labelled"
.format(widths[0] / avail))Single source and citation
Fan, Y. (2026). Cognitive Unloading of Dense Data Workbenches, Conditional-Format Salience and Zero-Training Error-Proof Form Engineering (Version 0.1.0), Chapter 2: Scan Paths and the Skeleton of a Table: Column Order, Alignment, Frozen Panes and Density. 5D Library. DOI pending
DOI pending · monograph page goes live after publication · 5DL-B-138 · v0.1.0 · CC-BY-NC-ND-4.0
BibTeX
@book{fan2026138,
author = {Fan, Yafu},
title = {Cognitive Unloading of Dense Data Workbenches, Conditional-Format Salience and Zero-Training Error-Proof Form Engineering},
year = {2026},
publisher = {5D Library},
edition = {0.1.0},
note = {DOI pending}
}Other chapter answers in this book
- Chapter 1 · Method: Three Losses, Four Metrics, Evidence Grades and the Liu Yan Reporting System
- Chapter 3 · Preattentive Salience: Let the Anomalies Jump Out, and Only the Anomalies
- Chapter 4 · Alarm Thresholds: Signal Detection, ROC, the Cost Matrix and a Per-Screen Alarm Cap
- Chapter 6 · Zero-Training Error-Proof Forms: Constraints, Defaults, Inline Validation, Confirm-or-Undo and Batch Guards
- Chapter 7 · The Workbench Playbook: Reconciliation, Orders, Stock and Customer Records, and the Distillation Spec
- Chapter 8 · Practice: Rebuilding the Harbourbridge Reconciliation Back Office from Diagnosis to Redesign