The short version
A Scribd link is not really a link to a page. It is a way of writing down a
number. 227246311 is the document, and everything around that
number is decoration: www, the language subdomain, the words in
the path, the tracking parameters. Get the number and you have the document.
The trouble is that the same document is written five different ways, and a tool that recognises four of them will refuse the fifth in exactly the same words it refuses a random web page. Here is the whole list, so you can see which one you have been handed.
Every shape a Scribd link comes in
| It looks like | What it is |
|---|---|
/document/123456789/title |
The shape people remember. Scribd redirects it to /doc/ now, so you will rarely see it in an address bar any more — but plenty of old links, forum posts and messages still use it, and it still works. |
/doc/123456789/title |
The current one. This is what every scribd.com page puts in its address bar today, and what you get when you copy from any of the language sites. |
/documents/123456789/title |
The plural, which also redirects to /doc/. You mostly meet this in older shared links. |
/presentation/123456789/title |
A slide deck. Different word, same document — the number is the document and the word before it only says how it was uploaded. If yours starts /presentation/ and a tool said it was not a Scribd document, that tool was wrong. |
/embeds/123456789/content |
What the page you were reading embeds, rather than the page itself. It works, and it carries no title, so a tool has nothing to call the file. |
What is not a document link
Three of these are Scribd pages and are not documents, which is worth knowing because two of them look almost identical to the real thing:
-
/docsis a section of the site./docs/123456789/titleis not a document, and a tool that accepted it would be guessing. -
/publication/123456789is one character from/presentation/and is something else entirely: it redirects to a person's profile. It is a user, not a document. -
/user/somebodyis a profile, and there is nothing to convert.
If you have the text, not the link
Sometimes the document link is inside a paragraph somebody pasted, or inside a message, and all you have is a wall of text with a URL in the middle of it. Pulling the links out by hand is tedious and easy to get wrong, which is usually how the wrong link ends up in a converter.
There is a free link extractor for that. It runs entirely in your browser — the text you paste is read on the page and never sent anywhere — and it lists the documents it found by their number, so two links to the same document come back as one rather than as two.
It also throws out the trailing punctuation that survives a copy and paste.
…this one), and… leaves a bracket on the end of a
link, and a link with a bracket on the end is not a link. The number is what
identifies the document, so extracting the number and rebuilding the link
from it fixes that automatically.
Getting from the link to the file
Once you have a link in one of the shapes above, paste it in and you have the document as a PDF. Two things worth knowing before you do:
- The file is a high-resolution picture of every page, so it reads and prints but you cannot search inside it. That is true of every tool that converts Scribd documents, and the reason is worth knowing before you spend an hour on a document you need to search.
- Anything behind a login, a paywall, or a download restriction on Scribd is not available to any converter. If the public page shows it to you without signing in, it will convert.