Current behavior: When a carrier uploads a driver’s statement as a scanned PDF (no text layer), Glean displays the pages but cannot expose selectable text; it returns an error that text is “stored in a non-standard, encoded way” and that OCR is not available.
Impact: For arbitration cases, I must quote exact language from driver statements in my rulings. When a statement is image-only, I am forced to manually retype the needed passages, which is time-consuming and error-prone.
Desired behavior:
- When Glean detects an image-only PDF, automatically run OCR in the background (or on demand) to create a text layer.
- Allow me to:
- Select, copy, and paste text from the OCR result, and
- Search within the document using the OCR text.
Benefit: This would significantly reduce time spent on each arbitration, lower transcription errors, and make it easier to include precise quotations from driver statements in rulings.
