File Formats & Media

PDF File

Definition

PDF stands for Portable Document Format. Adobe published it in 1993 and handed it to ISO in 2008; the current standard is ISO 32000-2, known as PDF 2.0. Every PDF starts with the bytes %PDF- followed by a version number, which is how software recognises one even when the extension has been stripped. Inside, the file is a set of numbered objects — a page tree, embedded font programs, images, drawing instructions — with a cross-reference table at the end that says where each object sits. Positions are absolute, so text does not reflow: an A4 page stays A4 on a 27-inch monitor and on a phone, where the reader pinches and pans instead. The MIME type is application/pdf, and sending that exact string is what makes a browser render the document in place instead of dropping a file into the downloads folder.

Why It Matters

The content type is where most PDF links go wrong. Served as application/octet-stream, the same file skips the viewer entirely and downloads, which many people read as a security prompt and abandon. Weight is the second problem: a 30-page deck exported with photographs at print resolution runs to 40 MB, while the same deck with images downsampled to 150 dpi lands nearer 3 MB — thirteen times less for something nobody will print. There is an accessibility cost too. A PDF produced by printing to file from a design tool carries no tags at all, so a screen reader meets an unstructured run of characters with no headings and no reading order.

How It Works

On request the server sends two headers that decide everything the reader sees. Content-Type: application/pdf identifies the format, and Content-Disposition chooses the behaviour: inline hands the bytes to the browser's viewer, attachment forces a save, and the filename parameter sets the name it saves under. Chrome, Firefox, Safari and Edge each ship their own viewer built on their rendering engine, so no plugin is involved. If the file is linearised — the option often labelled fast web view — the objects for page one sit at the front, and the first page paints while the remaining bytes are still arriving. URL fragments work as navigation as well: appending #page=4 opens the fourth page directly, and #search=invoice highlights matches on load.

Real-World Example

A logistics consultancy publishes a 6 MB research report at northwind-report.99helpers.site/state-of-logistics.pdf. Because 99helpers serves it as application/pdf with an inline disposition, the link in their newsletter opens the report on page one inside the browser, with no download step and no landing screen in between. Six weeks later they swap in a corrected edition under the same name, so every link already sent keeps working. The analytics then show 1,240 views against 180 downloads, which tells them most readers never wanted a copy on their own machine.

Common Mistakes

  • Serving the file as application/octet-stream or a generic binary type — the browser withdraws its viewer and forces a download, which reads as a warning rather than an invitation
  • Exporting at print resolution for a web link — 300 dpi photographs inside a 40 MB file cost the reader a long wait for pages they will only ever see on screen
  • Treating a PDF as an accessible substitute for a web page — untagged content has no headings and no defined reading order, so assistive technology gets a flat wall of text
  • Assuming page count drives size — embedded fonts and uncompressed images dominate, and a two-page flyer often outweighs a hundred-page manual

Related Terms

Put a file online in seconds

Drop in a document, an image, a page or a whole static website and share the link — free, with no build step and no server to set up.

Host a file free →