Notes beside books and videos

A reading edition is a text and its gloss, and a video is a transcript and its gloss; neither has room to say the other thing — that this idiom is the one from chapter two, that the speaker has just changed register, that the next four lines are a quotation. A note is that room. You write it in the seam between two lines of a book or a video, and it is a studio document: not one like it, the same — written in the same editor, drawn by the same renderer, with the same dialect, the same hover tools, the same glosses and exercises. What differs is only where it lives.

1. Writing one#

Between any two lines of a book’s reader, or of a video’s transcript, there is a small +. Press it and a new note is made there, and opens straight in the editor, over the page. Afterwards the seam carries a mark with the note’s title: point at it to see the note’s first lines, click it to open the note over the page. The reader’s and the player’s own pages say more about the seams and the marks: see the books and videos sections.

The window a note opens in has edit, which turns it to the editor, or read, which turns it back to the note’s page, and ✕; Escape closes it wherever the pointer is, even inside the note. Closing it draws the marks again, since the note’s title may have changed.

A new note starts short:

---title: A notelang: entarget: faanchor: after sub 1.2-6f8f21d7175b---Written by hand, about the line above. Everything the studio can do goeshere; nothing here is ever built to a PDF.

target:

is the book’s or the video’s language, so the note opens in the face and direction of the text beside it.

anchor:

says where the note sits: after or before a line, sub for a book’s subparagraph or cap for a video’s caption, and which one. The book’s line keeps its name through a rebuild; a caption is named by the second it starts at. Moving a note is editing this one line. A note whose anchor names a line that is no longer there is not lost: the reader shows it at the end, marked adrift, so you can open it and set its anchor again.

title:

is A note — or A note 2, A note 3…, since two notes of one book cannot share a name. Give it a name of its own; that is the name links to it spell.

2. A library of their own#

The notes of one book, or of one video, are a library of their own, kept in that book’s or video’s folder (its markdown/ subfolder) rather than in the studio’s library. So:

They travel with the book or the video

— in its download, in its bundle, in the shelf’s backup — and never mix with the studio’s own documents.

Names are unique among the notes of that book

and not across the toolbox: two books can each have a note called Vocabulary. The name-conflict dialog works there as it does in the studio.

Links reach the notes of the same book.

A link from one note to another — [the first note](doc:A note), or through Doc link…, which lists that book’s notes — works, is followed by renames, and appears in the other note’s ↩ Linked from. A link from a note to a studio document, or to another book’s note, finds nothing.

Every page and button is there

— scoped to the notes. A note’s page has its tags, its ⚙ panel, its contents, its glosses, its exercises with Check exercises and ⤢ Enlarge; its ← Library opens the library of that book’s notes, with the same search, tags, Upload .md / .zip, Backup, Load from backup and Download N shown as the studio’s. (A document made there with + New has no anchor, and the reader shows it at the end of the book, adrift.)

Exercise decks

— a note has + Deck and + Add all exercises, and its editor Load from a deck…, whenever Parseh knows which book or video the note belongs to; a deck’s copy of an exercise links back to the note.

3. What a note does not do#

A note is never printed. It is read on its page and nowhere else: Build PDF answers a note beside a book or a video is read on its page and nowhere else: it is never built to a PDF and never taken away as LaTeX. Copy it into the studio if you want either. — and Download ▾ → LaTeX (.tex) and → PDF stay greyed out. To print one, copy its text into a new document of the studio.

No prompt asks for one. A note is your own reading of your own book, and nothing in the toolbox asks a model to write one: the LLM prompt knows nothing of notes beside a book or a video, nor of anchor:. The buttons are still there, as every button is — the notes library has LLM prompt and Paste LLM answer, a note’s editor has Exercises ▾ → Generate with LLM… — and they work as they do in the studio; a document pasted there is a note with no anchor:, shown at the end, adrift, like one made with + New.

The preview on the mark is plain text. Pointing at a note’s mark in the reader shows its title and its first lines — at most 280 characters, the front matter, the footnotes and the pictures left out, and each exercise shown only as <Exercise> — as text and nothing more: a note can come in a bundle from somebody else’s toolbox, and merely passing the pointer over it must not do more than show its words. A touch screen has no pointer to rest, and a tap opens the note.

Notes written before links named documents by name are brought up to names the first time the notes of that book or video are opened after Parseh starts, as the studio’s library is (see Names and links): notes that were all called A note, as every new note once was, become A note, A note 2, A note 3…, the oldest keeping the name.