PDF Glossary: OCR, Flatten, Linearize, PDF/A, and More Explained
Updated July 13, 2026
PDF terminology gets thrown around a lot without much explanation. Here’s what the common terms actually mean, in plain language, with a link to the HivePDF tool that does each one.
OCR (Optical Character Recognition)
Turns the picture of text in a scanned page into actual, selectable text, layered invisibly underneath the image. Without it, a scanned PDF is just a photo — you can’t search it, copy from it, or have a screen reader read it aloud. → OCR PDF
Flatten
Bakes interactive content — form field values, annotations, layers — directly into the static page content, and removes the interactive elements themselves. A flattened form can’t have its answers changed; a flattened annotation becomes part of the page rather than a separate editable object. → Flatten PDF
Linearize (a.k.a. “Fast Web View”)
Reorganizes a PDF’s internal structure so the first page can start rendering before the entire file has downloaded — useful for large PDFs viewed in a browser, where a linearized file shows page 1 almost immediately instead of waiting for the whole download.
PDF/A
An archival subset of the PDF format designed to render identically decades from now — it requires embedding all fonts, forbids external references, and avoids features (like encryption or JavaScript) that could stop working over time. Used for legal records, government filings, and anything with long-term retention requirements. Most everyday PDFs don’t need it.
Redact
Permanently removes sensitive content rather than just covering it with a black box — the difference matters because a simple black rectangle often leaves the original text intact and extractable underneath. Real redaction removes the content itself. → Redact PDF · see also how to redact a PDF safely
Metadata
The hidden information attached to a PDF beyond its visible content — author, title, creation date, the software that made it, sometimes even edit history. Worth checking (and sometimes stripping) before sharing a document publicly. → Edit PDF Metadata
Encrypt / Protect
Adds a password requirement to open a PDF (and optionally restricts printing, copying, or editing even for someone who has the password). Different from redaction — encryption controls who can open the file at all, not what’s visible once it’s open. → Protect PDF
Compress (rasterize vs re-encode)
Shrinking file size, usually by reducing image quality or resolution. Rasterizing converts a page into a single image (common for scanned pages); re-encoding just recompresses existing images on a page while leaving real text alone. The distinction matters because rasterizing a text page destroys its selectable text — a good compressor (like HivePDF’s) only rasterizes pages that were already images to begin with. → Compress PDF
Fillable form field
An interactive element in a PDF — a text box, checkbox, or dropdown — that a reader can click and type into directly, as opposed to a “flat” PDF that just looks like a form but has nothing to click. → Fill PDF Form
Client-side / in-browser processing
Software that runs entirely inside your web browser using WebAssembly, rather than uploading your file to a server. The defining trait: it keeps working with no internet connection once the page has loaded, because there’s no server in the loop. → how client-side PDF processing works