A scanned PDF is a photograph of paper. Nothing in it can be searched or copied until the words in the picture have been recognised — which is what OCR does.
Not with PalPDF. OCR runs inside your browser, which is why the first run pauses briefly to load the language data.