Levels and their names

Each subparagraph of a book is set several times over, one after the other, and each setting is a level. Which levels there are depends on the language. The header of the reader has one button for each, named for what it shows — With vowels, Chunks, Plain, Nastaliq in a Persian book — under the word levels. There are no numbers on them anywhere: not on the buttons, not in the panel, not in what pointing at one says.

A level is one of a few kinds, and the kind is what a name is kept against: the sentence on its own, the chunks with their glosses, the sentence as it is ordinarily printed, the sentence in another face, and — Japanese and Chinese — the reading alone. Level 1 is your own attempt at the sentence, before any help arrives; the chunks level is the help; the others give the sentence again, as it is really printed, in its other face, or as it is read aloud.

1. What each language calls them#

These are the names a book has until you change them, and the sentence the panel prints under each (The ⚙ settings of a page). They come from Parseh’s language registry, so a language you add brings its own (Adding a language).

The buttons come in the order of the table. In Persian and Arabic the first level has the vowel marks, the third has none and is not a hovering text — there is no cloud on it, so it is the sentence as a printed page has it.

2. Hiding a level#

Press a level’s button to show or hide it — the level goes and comes back, and the rest of the page closes up. Or open the panel, ⚙ page → Levels & reading → Levels, and tick or untick it there; the two are one switch. What you choose is kept on this device, for every book: turn everything off but the sentence to read it alone, or leave only the chunks to work through the glosses. The choice is made for a kind of level, so hiding Plain hides it in every book that has one.

While hover is on, the levels it takes over rest — the chunks, the other face, the vertical one — and the panel says so under the row: Hover mode shows the glosses in the cloud instead. (The reader.)

3. Renaming a level#

The panel’s Levels row is one line per level: the tick, a field with the level’s name in it, and under it what the level shows.

Type a new name in the field and the button on the page has it a moment later, and so does its tooltip, and so does the sentence hover mode’s own tooltip says (the text alone (the levels With vowels and Newspaper)…). The rules:

Twelve characters at most

, on one line — what a button holds. A name that is too long is cut as it is typed.

Empty is the default.

Clear the field, or type the default name itself, and the level goes back to the name in the table above.

A name is for a language.

Plain in Persian and Plain in Arabic are two names, so you can call one Newspaper and leave the other. A book in that language wears it, whichever book it is.

It follows you.

A name is a word you chose, not a way a screen is set up, so it is kept by the computer and every device you read on wears it: rename a level on the phone and the computer’s reader has it the next time it loads.

It works in every reader

, however long ago the book was built: the names are put on the page when it opens, with no rebuilding.

4. Diacritics#

A Persian or Arabic sentence can be written with the small marks above and below the letters that say the vowels, a doubled consonant and a stop. A reader who knows the word does not need them, and a page full of them is hard to look at. Diacritics puts them away — in the reader, where it is ⚙ page → Levels & reading → Diacritics, and in a video’s own panel, Watching & reading → Diacritics — and puts them back.

Which marks.

Exactly the ones the third level has never had, and that the lookups fold away: in Persian the eight of U+064B to U+0652 — fathatan, dammatan, kasratan, fatha, damma, kasra, shadda (the doubling) and sukun (the silent stop); in Arabic the same eight and the superscript alef (U+0670). One rule, whichever level or video. No letter, hamza or madda is touched.

Which levels.

The first and the second — With vowels and Chunks — in a book; the lines of the transcript in a video. The third level, Plain, is as it ever was: no marks at all, as the language is ordinarily printed, with no cloud. Diacritics does not change that level, and it is not the same thing as it: with Diacritics off you still have the first level, its words still open their cloud when you point at them, the chunk’s gloss in it — the third level is the same text without the hover.

Display only.

The marks are put away from the screen and from nowhere else. The book’s files, the search, the lookups, the cards, the printed PDF and a bundle are exactly as they were. Selecting text with the mouse and copying it gives what is drawn, without the marks; but Shift-click, the cloud’s copy button, a card and the chunk editor give the book’s own text, with its marks, and so does the search — a card made from a chunk with the marks put away carries the vowelled text.

Chapters that come later.

A book of several chapters brings them as they are wanted; a chapter that arrives after you turned the marks off arrives without them.

A video

has the switch only if the video’s lines hold marks — a Persian video whose captions are written without any has nothing to put away, and no row. The lines, the cloud and the colours work as before; copying, cards, the editor and every lookup keep the phrase as it was written.

On this device.

The switch is kept on this device, and the marks are shown to begin with. It is in the panel only — a language with no such marks (Italian, Japanese…) has no row.