Skip to content

Paleography View ​

The paleography view collects every instance of each syllable in the edition and lays them out as cut-outs from the manuscript image — one row per syllable, every attested writing of it side by side. Open it with the table icon on the image pane toolbar.

The paleography view: rows for a, i, e, o and ka, each showing every instance as an image cut-out

This is the sign chart for this hand, built from the edition rather than drawn by hand — and because each cut-out is the actual writing, variation within a row is variation in the scribe's practice, not in someone's transcription of it.

Finding a syllable ​

The search box filters by syllable. It matches whole syllables, not substrings, and takes several at once separated by commas:

ta, ṭa, da, ḍa

That puts the four rows adjacent, which is the arrangement you want when the question is whether the scribe distinguishes them.

Rectangle or polygon ​

The Polygon switch changes how each cut-out is clipped:

  • Off (default) — a padded rectangle around the akṣara. Shows the sign in its immediate context: neighbouring strokes, the state of the surface, the line above and below.
  • On — clipped to the recorded outline of the akṣara itself. Isolates the sign from everything around it, which is what you want when comparing forms.

Switch between them freely; it is the same data drawn two ways.

Selecting ​

Click a row to highlight every instance of that syllable in the transliteration. Click again to clear. The Details pane collapses while a row is selected — the row is the subject now, not any one sign.

Click a single cut-out to select that instance: the transliteration scrolls to it, and you can follow it back to its place in the text. Hovering a cut-out previews the same highlight.

Glycerine Aksara — Systemik Solutions