Capture From Safari
Save selected Japanese text directly from the browser without manually copying material between unrelated applications.
NipponCapture began with a simple problem: useful Japanese appears constantly while reading online, but saving it for later often interrupts the reading itself.
A word may be unfamiliar. A sentence may demonstrate a useful grammar pattern. An article may contain vocabulary tied directly to the user's interests. The material is valuable precisely because it appeared in something the person had already chosen to read.
NipponCapture is being built as a bridge between that moment of discovery and later study. A selected word or sentence can be captured from Safari together with its surrounding context, page title, source URL, and time of capture, then reviewed and exported when the user is ready.
What NipponCapture Does
Preserve Context
Captures can retain the selected text, sentence, surrounding paragraph, page title, source URL, and capture time so vocabulary is not separated from the material that made it meaningful.
Review Before Export
Saved material enters a local inbox where it can be reviewed, organized, archived, or prepared for later study instead of being immediately pushed into a growing pile of flashcards.
Build From Real Reading
The study material comes from the user's actual browsing and interests: articles, reports, literature, manuals, fitness writing, research, or anything else they were already motivated to understand.
The Design Philosophy
“The material worth studying is often already passing through your attention.”
Language-learning systems often begin by deciding what the learner should study, then presenting that material in isolation. NipponCapture begins from the opposite direction: the learner is already reading something, has already encountered a gap in understanding, and has already demonstrated that the surrounding subject matters enough to hold their attention.
The goal is not to turn every unknown word into homework. It is to make capture cheap enough that useful material can be preserved, then leave the decision about what deserves further study until later.
Why It Exists
Reading authentic Japanese creates a constant tension between immersion and study. Looking up every unknown word destroys momentum. Ignoring everything preserves momentum but allows useful discoveries to disappear. Copying text into notes, spreadsheets, dictionaries, and flashcard tools creates friction and fragmented records.
NipponCapture is intended to reduce that friction. The capture itself should take only a moment. Organization, explanation, tagging, export, and deeper study can happen later, outside the reading session.
Over time, the captured material can also become a record of what the user actually encounters: recurring vocabulary, fields of interest, sources read, unfamiliar kanji, and patterns that may be more personally relevant than a generic curriculum.
Project Status
| Area | Current State |
|---|---|
| Safari capture | Working context-menu capture from selected text into the local application. |
| Captured data | Stores selected text, sentence, paragraph, page title, URL, capture date, and archive information. |
| Local inbox | Working capture inbox with reload, duplicate protection, export, and archive behavior. |
| CSV export | Working export workflow with cumulative data preservation. |
| Study workflow | Tagging, richer review states, Anki-oriented export, explanations, and kanji support are planned or under exploration. |
| Public availability | Not currently released. The project remains an experimental local workflow under active development. |
Development Notes
-
JUL 2026Initial Safari capture workflow established, including local App Group storage, capture review, duplicate protection, CSV export, and archiving.
-
CURRENTRefining capture states, multi-selection, deletion, tagging, review behavior, and the path from raw capture to useful study material.
-
LATERAnki-oriented export, Japanese-first explanations, kanji and radical information, richer dictionaries, and larger reference datasets may become part of the broader system.