To Hell With Fair Use: IBM Patents a Printer Which Doesn’t Belong to Its Buyer
DRM has long enabled a file’s creator to limit how it is used. We’ve seen this most often in PDFs, but also office docs and even ebooks, where for example Adobe has a print/readaloud tag in Epub ebooks which can be disabled the publisher.
Now IBM has taken this old idea and applied it to printers.
For those of us living in the 21st-century, a printer was an old way of sharing a document. Way back in the 20th-century, before the internet and email, before smartphones, and even before computers, a printer was a machine which used ink to "print" a copy of a document on paper.
You can think of it as analog, offline cloud storage with limited capacity and durability, and effectively no distribution. (I’m pretty sure those who want to see a printer can find one in the Smithsonian, or some other museum.)
And now IBM wants to apply DRM to this technology from yesteryear, all with the goal of preventing copyright infringement.
Torrentfreak writes that:
unauthorized printing can be problematic for copyright holders, such as book authors, IBM says, and this week the company filed a patent application for a technology that aims to counter the problem.
Simply titled “Copyright Infringement Prevention,” the patent’s main goal is to ‘restrict’ the functionality of printers, so they only process jobs when the person who’s printing them has permission to do so.
It works as follows. When a printer receives a print job, it parses the content for potential copyrighted material. If there is a match, it won’t copy or print anything unless the person in question has authorization.
“The computer, in response to identifying any text, images, or formatting indicative of potential copyrighted material, identifies potential copyrighted material within the file.”
“The computer determines whether the file may be printed based, at least in part, on the identified potential copyrighted material,” the patent description adds.
The patent describes various variations on this approach, and IBM notes that ISBN numbers, United States Copyright Office records, and other public resources could be used to define the copyright status of a work.
I don’t know what’s crazier: that everyone is calling this a printer patent, when in fact software does all the work; or that this patent is obviously intended to deprive a printer’s buyer of control over their property as well as their rights under copyright law.
The patent application says that it wants to make sure that the owner has permission to print a file, and that would be all well and good except for the fact that there are a hundred and one different parts of copyright law under which I don’t have ask anyone’s permission.
Depending on your legal venue, this includes fair use, fair dealings, public domain, personal copying levy (aka the "you must be a pirate" tax), and creative commons.
This patent proposes to be able to navigate the thicket of laws, rights, permissions, and licensing, some of which dates back to the 17th century, and adjudicate whether a printer’s owner can print a file?
Good luck with that.
This is an unworkable idea on any level, but sadly it will still come to pass. Sure, you might not want one, but colleges, libraries, and institutions might get a printer encumbered by this patent if someone pitched it as a feature. It wouldn’t be hard to see why it’s a bad idea, but not everyone is going to think it through and realize that this isn’t going to solve any problems while at the same time it will deprive many people of their rights.
If you see one in the wild, do let me know. You can bring the axe, I will bring my bat’leth, and we can slay the monster together.
image by Col Frankland
Darryl May 20, 2016 um 1:46 am
Slaying the monster simply won’t be necessary. It will sit gathering dust, alone, unloved and unwanted until it is given away for a new life as a paperweight or a doorstop.
Popup May 20, 2016 um 5:33 am
Have you heard of the EURion constellation?
It’s an image that cannot be printed on modern printers! Since two decades it has been added to paper money, and colour printers can detect the pattern, and refuse to print images which contain it.
Nate Hoffelder May 20, 2016 um 10:29 am
Interesting. So that would be prior art, wouldn’t it? (Tow, counting Bill’s comment below.)
This invalidates the patent, but that’s no guarantee that it will kill the idea.
Bill Rosenblatt May 20, 2016 um 4:13 pm
I’m not suggesting that the Stefik patents for Xerox PARC are prior art to what IBM is attempting to patent here (FYI, it’s a patent application, not an issued patent). The approaches are different. Stefik’s approach was DRM. This approach is forensic – examine the work and use the same sorts of techniques that copyright monitoring services might use to determine whether it’s potentially infringing – identifiers, metadata, fingerprinting-style pattern recognition, etc.
I’m scratching my head over what the incentives or uses cases could possibly be here. Just as Xerox might have done in the mid-late 1990s, but also particularly since infringement by hardcopy is sort of like robbing Fort Knox by tweezers…
Nate Hoffelder May 20, 2016 um 4:51 pm
My mistake, Bill.
I can see the market for this idea (as terrible as it is), although i don’t see how it would be practical on a technical level. Any analytical methods would lead to too many false positives. That’s why I started this post with DRM; I can see how building the permissions into the doc files (DRM) is the only practical way to do it.
Bill Rosenblatt May 21, 2016 um 7:43 am
Ok, I’ll bite: why do you suppose anyone would want to buy a printer like this? Or why would a company either charge more for it than a standard printer or eat the difference in product cost (including the patent license royalty)?
Put another way: who would a) be that afraid of being sued, b) be convinced that simple non-technical measures (e.g. sign here or check this box to attest that you are only printing material to which you have the rights) would not allay that risk, c) be convinced that this device would allay the risk, and d) not be worried about pissing off customers? Is there anyone?
Nate Hoffelder May 21, 2016 um 9:46 am
I’m not saying it’s a good idea, but I could see someone pitching the idea to US colleges. Some schools are buying network-level content filters in order to fight piracy. I could see how this might be sold to them as an anti-piracy measure.
Bill Rosenblatt May 21, 2016 um 6:53 pm
Right, those resulted from the Higher Education Opportunity Act of 2008. The requirements in that act are wishy-washy and don’t get to the level of sophistication that’s specified in that IBM patent application. College IT administrators now know that they can do far less than that in order to "tick the box" and be in compliance with the Act. And besides, it was the music industry who caused that to happen; publishers weren’t really involved.
puzzled May 20, 2016 um 5:39 am
Just wait until a company that has bought this system changes its company name…
Bill Rosenblatt May 20, 2016 um 9:08 am
Actually I don’t think that will be necessary.
Technology of a similar ilk was invented at Xerox PARC roughly 20 years ago. Xerox probably (though I have no confirmation of this from the source) had intended it to enable them to build printers that followed machine-readable instructions embedded in encrypted digital documents which controlled rights to print those documents and under what conditions. Xerox did attempt to commercialize the technology, though not for that purpose. Xerox never did build DRM-enabled printers.
puzzled May 21, 2016 um 7:33 pm
As interesting as this is, and as few sales as it will lead to, anyone who installs this printer system will remove it as soon as it refuses to print a RFP or other document that leads to losing money from not being able to make a sale.
The last thing anyone needs in the half hour before printing the final version of a proposal is the damn printer refusing to print the document.
SHANLEE May 22, 2016 um 12:58 pm
Would IBM be storing all this data somewhere? Could they use it to sell data to advertising companies? Could the Government/Police subpoena data if they suspect domestic or international terror plots using printed "bomb/terror making" books or pamphlets? Looks like a NSA dream. Just another way to figure out what we are doing.
SHANLEE May 22, 2016 um 1:03 pm
NOT that I would be doing anything illegal! I’m a writer. if I am writing a book about terrorists and I get on a site that shows how to make a bomb, would the NSA be knocking at my door after I printed out something?