Judge Rules Internet Archive Guilty of Copyright Infringement

[Kiwi]

Today Internet Archive were found guilty of copyright infringement in a US District Court with the judge ruling their arguments around being a "Library" having no legal protection for the scanning of 127 titles listed in the complaint brought by publishers Hachette, HarperCollins, Penguin Random House, and Wiley.

A nice article on the ruling can be found here

The chief issue relates to their acquiring a physical copy and then scanning it rather than sourcing a legitimate digital copy from the publishers then making it available to anyone and whereby the publishers didn't receive compensation as normally would be derived from libraries as per negotiated agreements.

This all blew up during COVID when IA dropped their one copy per person rule and made files freely available without any limits on the number of people reading a particular title at any one time.

[kitsunebi]

Relevant to the magazine-scanning community is that the judge ruled that scanning something is not considered transformative, since that's an argument I've seen certain scanners make in the past to legitimize their efforts.  Scanning something and turning it into a searchable database - yes, but just scanning something into a readable archive of images as we do, no.  So if it wasn't obvious before (as it should have been), yes Virginia, magazine scans aren't immune to copyright law.

[Kiwi]

Didn't marktrade use that excuse at one point when plying the Patreon waters?

[slider1983]

Scanning something and turning it into a searchable database - yes, but just scanning something into a readable archive of images as we do, no.

Could you clarify this for me? Isn't scanning something to make it readable in a digital form the same as hosting a website with that content? Or is it more that scans by people who did them that's allowed? 🤔

[Kiwi]

kitsunebi can probably clarify better but I think it means scanning the pages and then using the resulting text output in a database is okay because the information is NOT in a form identical or representative of the physical format it was taken from as it is simply text. It doesn't have images or anything else from the original it was obtained from.

[kitsunebi]

It's important to note that an OCRed PDF would also considered a derivative work and is a violation of copyright.  "Searchable database" does not mean "searchable PDF."  It really shouldn't need too much explanation, but if a copy is made of something that has the intent of being A COPY OF SOMETHING, it's a derivative work and is illegal unless permission is given by the copyright holder.  That applies to all magazine scans, which strive to accurately depict the contents of a magazine from cover to cover.

Nothing new here, and nothing all of us didn't already know (the scanner who tried arguing against this has long since left the scene.)

[slider1983]

Nothing new here, and nothing all of us didn't already know (the scanner who tried arguing against this has long since left the scene.)

Well obviously I didn't.    So to clarify a complete scan would be a copy but random scanned pages would be okay as those don't make up a complete scan. Also as long as said scans are being used searchable in a database and not for download that's fine as well.

[kitsunebi]

So to clarify a complete scan would be a copy but random scanned pages would be okay as those don't make up a complete scan. Also as long as said scans are being used searchable in a database and not for download that's fine as well.

No, that's incorrect. Any scan would be a violation of copyright, whether it be a full scan or a single page.  A scan is literally a copy, and serves the same purpose as the original (to be read), so it is considered a derivative work.

Also, to clarify what is meant by a "full-text searchable database" - this DOES NOT mean that a full text copy of the magazine's contents is available for users to read.  Once again, that would amount to being a copy that fulfills the same purpose as the original.  A searchable database is just that - someone can input a search term and get a result directing them to where that term can be found. 

For example, let's suppose Kiwi made a full-text database of every mag here at OGM.  A user could search for "Sid Meier" and get a list of every issue and page number where that name appears.  However, the user would not have access to either scans or text copies to those pages - they would still have to locate a copy of those magazines on their own.  This is why "full text searchable databases" are mostly in use by libraries, since after doing the search, someone can then legally check out the mag/book in question. 

Prior to the judge's decision that a full-text searchable database was legal, making a full-text copy for any reason was illegal, and it was indeed libraries that were initially sued by the copyright holders.  Now it falls under fair use so long as it is considered transformational - meaning that it serves a different purpose from what it was originally created for.  In this case, it is transformative because even though a full-text copy exists on a server somewhere, it is not accessible to be read, and can only be searched for the purpose of data location/research, thus giving it a different purpose than the original.

Clear?

[slider1983]

Right I get it now. Thanks. 👍