Ship's log.
Every entry, newest first. Some unfold into a longer note: what was built, what broke and what it taught us. Facts and numbers, no announcements.
Image to Text OCR and Linklet are live in Microsoft Edge Add-ons, a day after submission. Their sites now send Edge visitors to the Edge store.
Image to Text OCR is live in the Chrome Web Store, three days after submission. Its site keeps the free online tool and now points to the store after three reads a day.
Image to Text OCR is submitted to the Chrome Web Store. Its site already reads images and PDFs in the browser, no install needed.
Tesseract.js reads table borders as letters. What fixed our OCR
Tesseract.js turns grid lines into brackets, collapses tables into one column and never admits it cannot read. Three fixes that worked, with the reasons.
Nooman is docked. A new publisher gets two slots in the Chrome Web Store, and its slot goes to the next ship, Image to Text OCR.
A new Chrome Web Store publisher gets two extension slots
Since August 2026 a new Chrome Web Store publisher can have two published extensions. What the limit looks like, when it bites, and how we freed a slot.
New charts: this site redrawn.
Nooman 1.0.1: search through Chrome's own API, per store policy.
Linklet built, site launched.
Gallera Studio sets sail. Nooman built, site launched.