Save a Wikipedia article to read offline

Why a PDF and not a saved page

Saving a web page keeps a copy of the markup and a folder of assets beside it, and it depends on both surviving and on a browser willing to put them back together. A reading-list entry is often just a bookmark that goes and fetches the page again when you open it — which is exactly what you cannot do on a plane.

A PDF is self-contained by construction. The pictures are embedded. So are the fonts: only the characters actually used, but genuinely inside the file, so a device that has none of your fonts installed still shows the document the way you made it — including Greek, Cyrillic, CJK, Arabic and Indic text, which is where a saved web page most often falls back to boxes.

And it is one file. You can put it in a folder, sync it, mail it to somebody, or keep it for ten years.

Saving it

Generate the article as usual, then:

• On a computer, the viewer bar has ⬇ Save PDF.
• On a phone or tablet, ⇧ Share and ⬇ Save sit at the foot of the screen, under the pages. Share hands the file to whatever your device does with files — Files, Drive, a message, another app.

The file is named after the article, minus any characters your filesystem objects to, so a folder of these reads as a shelf rather than as document (3).pdf.

Nothing is uploaded, and there is nothing to sign up for

The article is fetched from Wikipedia by your own browser, and the parsing, the layout and the PDF itself all happen on your machine. There is no server that made your file, so there is no server that has a copy of it, and no account for it to belong to.

That is worth stating precisely rather than as a slogan: the about page lists everything your browser actually talks to and why.

The file knows what it is

A folder of files you cannot identify is not a library. Each PDF carries its own title, language, description and keywords, so your file manager and PDF reader show the article’s real name rather than a filename, and a search across your documents finds it.

The closing Sources & License page goes further: alongside the CC BY-SA credit it records a permalink to the exact revision the copy was made from — the ?oldid= form Wikipedia’s own Cite this page uses — and the date. Wikipedia moves; your copy does not. In two years that line is the difference between “an article about this” and a citable source you can put back in front of the version it came from.

What still works with no connection

Everything that points inside the document: the contents page, the PDF bookmarks your reader shows in its sidebar, and the citation links — tap a footnote marker and you land on the reference, on a device in aeroplane mode.

Links that point out of the document are still links, and still need a connection when you tap them. The QR code you can put on the first page is in the same category: it is there for when you are back online and want the live article.

Two things worth knowing before you rely on it

Both of these are the opposite of what a reader looking for an offline library tends to assume, so they are here rather than in small print.

The browser does hold the last few documents so that Back and a page reload return you to what you were reading. That is a convenience inside one browser on one device, and clearing site data clears it. It is not where a library lives — a library lives in your files.

Making the copy worth keeping

Three things are worth a minute before you save. Choose the paper size for whatever you will read it on, and turn the picture quality up if it is going on paper — both are in printing a Wikipedia article. Drop the references and anything else you did not come for, in choosing which sections to keep. And if you are packing for a trip rather than saving one article, several articles can go into a single book PDF with a contents page and a chapter each.