Sign In Sign Up

Summarize directory contents

MF
Matt Foster • @mpf
MF
Summaruse the content of these directories

The user wants me to summarize the content of the directories in their Kindle data export. Let me look at the key directories to understand what's there.

Let me look at the key directories and files to understand this dataset.

README.md
Digital.Content.Ownership
Kindle.Devices.ReadingSession
Kindle.ReadingInsights
Kindle.ReadingBehaviorCounts
Kindle.UnifiedLibraryIndex
Kindle.KindleDocs
Kindle.Reach
Kindle.AnnotationSynchronizationDetails
outputs

Let me sample a few JSON/CSV files to understand the data structure:

ls /Users/mpf/Downloads/Kindle/Digital.Content.Ownership/ | wc -l
    6902
Digital.Content.Ownership/ Digital.Content.Ownership.1.json
Kindle.Devices.ReadingSession/ Kindle.Devices.ReadingSession.csv
 L1-5
Kindle.ReadingBehaviorCounts/ Kindle.ReadingBehaviorCounts.csv
 L1-5
main.py
Kindle.ReadingInsights/ datasets
Kindle.UnifiedLibraryIndex/ datasets
Kindle.AnnotationSynchronizationDetails/ datasets
Kindle.Reach/ datasets

This is an Amazon Kindle data export (GDPR/privacy data request). Here's a summary of the major categories:

Category Content
Digital.Content.Ownership ~6,900 JSON files — each is one digital content item you own (books, etc.) with ASIN, title, purchase date, rights, and download history. Goes back to 2010.
Kindle.Devices.ReadingSession ~13,800 rows of per-session reading data: timestamps, ASIN, device, reading duration (ms), page flips.
Kindle.ReadingBehaviorCounts Monthly aggregate reading stats: genre breakdowns, completion rates, avg reading days, KU membership, language splits.
Kindle.ReadingInsights Sessions with time adjustments, daily reading units, and titles completed.
Kindle.UnifiedLibraryIndex Library metadata: author relationships, genres, tags, and content relationship types.
Kindle.AnnotationSynchronizationDetails Highlights, bookmarks, and notes synced across devices.
Kindle.Reach Notification and push endpoint registrations.
Kindle.KindleDocs Personal documents sent to Kindle.
Kindle.Devices.* (30+ dirs) Granular device telemetry: app sessions, search events, toolbar interactions, UI impressions, notification metrics, sync metrics, etc.
Digital.* (Series, Whispersync, Sharing) Series relationships, Whispersync sync state, household sharing settings.
Audible files Listening.csv and audible_activity.json — Audible listening history.
outputs/ Previously generated analysis: a reading heatmap (HTML), timeline (JSON), and activity chart (PNG).
Scripts main.py (stub), build_timeline.py, reading_heatmap.py — Python tools for analyzing this data.
MF
What's going on with the business minded CISO, the numbers seem too high

The user is asking about a specific book - likely "The Business-Minded CISO" or similar title. They're saying the reading numbers seem too high. Let me search through the data to find this book and its reading session data.

Let me find that book and its reading session data.

grep -ri "CISO" /Users/mpf/Downloads/Kindle/Digital.Content.Ownership/ | head -5
grep -i "CISO" /Users/mpf/Downloads/Kindle/Kindle.Devices.ReadingSession/Kindle.Devices.ReadingSession.csv | head -5

The ASIN is B0D1DWVZ78. Let me search the reading sessions by ASIN:

grep "B0D1DWVZ78" /Users/mpf/Downloads/Kindle/Kindle.Devices.ReadingSession/Kindle.Devices.ReadingSession.csv
Warning: long output
grep -r "B0D1DWVZ78" /Users/mpf/Downloads/Kindle/Kindle.ReadingInsights/datasets/ 2>/dev/null | head -20
grep -r "B0D1DWVZ78" /Users/mpf/Downloads/Kindle/reading_timeline.json 2>/dev/null | head -10

Now let me compute the totals:

grep "B0D1DWVZ78" /Users/mpf/Downloads/Kindle/Kindle.Devices.ReadingSession/Kindle.Devices.ReadingSession.csv | awk -F',' '{sum += $10; count++} END {printf "Sessions: %d\nTotal ms: %d\nTotal hours: %.1f\n", count, sum, sum/3600000}'
Sessions: 187
Total ms: 226652000
Total hours: 63.0
# Find the outlier session
grep "B0D1DWVZ78" /Users/mpf/Downloads/Kindle/Kindle.Devices.ReadingSession/Kindle.Devices.ReadingSession.csv | awk -F',' '{if ($10 > 10000000) print $0}' | sort -t',' -k10 -rn
2025-07-18T19:34:17Z,2025-07-19T07:20:13Z,B0D1DWVZ78,www.amazon.co.uk,www.amazon.co.uk,Kindle for Mac,75E1D945A98F5F06A2B4BFC641BDCE09,1191416202,E-Book,42356000,1

Found the problem. Here's the breakdown for "The Business-Minded CISO":

  • 187 sessions across Jun 15 – Jul 19, 2025
  • Reported total: ~63 hours — that does seem way too high for one book

The culprit is one massive outlier session: 2025-07-18 19:34 → 2025-07-19 07:20 recorded as 42,356,000 ms (~11.8 hours). That's almost certainly a ghost session — the Kindle for Mac app was left open overnight and the session timer kept running. That single session accounts for ~19% of all reading time on this book.

Without that session: ~51 hours across 186 sessions — still high but more plausible for a ~35-day reading period if you were reading intensively. There are also several other long sessions (1+ hour) that may include idle/background time from the Mac app staying open.