| Wednesday April 16th 2014

Feedburner

Subscribe by email:

We promise not to spam/sell you.


Search Amazon deals:

HOWTO: Defeat Sealed PDFs, A DRM Nightmare


sealed-icon.pngHave you ever heard of a “sealed” PDF? I don’t mean password-protected or branded with your name (as some eBook sellers do). This is a different beast. Let me introduce a company called SealedMedia. Their business is protecting electronic documents (PDFs), a form of Digital Rights Management (DRM) which has always been a topic of ongoing debate. One of their most notable clients: Harvard Business School Press.

The Problem:

I had to write a paper on case #698-004, entitled “We’ve Got Rhythm! Medtronic Corp.’s Cardiac Pacemaker Business,” which is a great case by the way. Unfortunately, my experience purchasing and reading the case was not.

The HBSP Online Store allows visitors to purchase cases individually, in PDF format, well actually SPDF format. The .spdf extension indicates the document has been “sealed” by SealedMedia, and consequently Adobe Reader cannot properly render it without the SealedMedia plug-in and a valid license.

When you purchase the case, you receive a license with a login name and password. When launching Adobe Reader, you must authenticate to the license server to view the document:

sealed-login

The license and the entire concept itself bestows the following restrictions:

  • Your license is only valid for six months. When the license expires, you can no longer view the document without paying for another.
  • The SealedMedia plug-in must be installed to the view the document. This will exclude any public computer (work, library, university) which does not already have the plug-in installed by the system administrator and restricts the installation of new software. Additionally, the plug-in is only compatible with specific versions of Adobe Reader. At the time of writing, SealedMedia only supports up to Adobe Reader 7.0.9, whereas 8.0 was released two months ago.
  • You must be connected to the Internet to view the document. The SealedMedia plug-in for Adobe Reader must connect to the license server and validate your login name and password to fetch your license, every time you double-click the file to open it.
  • While viewing the document, you cannot “print to image” or “print to PDF”. These are features of most operating systems, letting you output documents to other common file formats, instead of an actual printer, using the Print command. If you attempt to do this, you will receive a message stating that you need to purchase a print license.
  • Screen capturing or “Print Screen” does not work. As soon as you enter screen capturing mode, the entire document becomes hidden (I haven’t tested on Mac OS X):
  • You cannot copy and paste anything from the document to another application. If you want to incorporate any text into a report or use it in any other manner, you will have to copy the text by hand.
  • Your license can only be “checked out” to one computer. Once you log in and view your document (on your home PC for example), you cannot then view it at work without relinquishing the license from your home PC first. If you fail to”check in” your license, you will receive a”license already checked out” error. Your only alternative is to wait for the license to time out, which varies between companies using the technology (12 hours for HBSP).

That’s quite a list of restrictions for one document, averaging $6.50 a pop. It is certainly unpleasant knowing that once you agree to legitimately purchase a document from a publisher, these restrictions come bundled with it.

Think you can just take a screenshot? Think again.
sealed-screencap

Think About It

  • The policy is far too restrictive. I’m confident HBSP generates enough revenue selling hard copy case studies and legitimate electronic copies to care about unauthorized duplication. Where do you draw the line between security and convenience?
  • Harvard Business School Press may not be a candidate for solutions like SealedMedia. It seems that many individuals, universities, and businesses count on the instant availability of high quality case studies and other materials in a compact, easy to manage electronic format – without the accompanying restrictions which only distract and inhibit users from working naturally.
  • It is imperative for a company offering this type of solution (especially in a plug-in form) to stay current with the industry by keeping their software updated and properly stating the compatibilities and incompatibilities on their web site (adopting a support forum, posting fixes and workarounds, etc.).

The Solution:

The purpose of this post is to help other users get FAIR USE of .spdf materials that they have paid for but that are practically unusable due to a combination of bugs in the Sealed Media software and missing or obfuscated disclosures from vendors (in my case Harvard Business School that publishes HBR articles).

Specifically, in my case, I purchased an HBR case online since I was about to travel internationally where I would not have reliable Internet or printer access nor did I have time to print the article before leaving.

Due to the following combination of mostly non-disclosed “features” (aka bugs) in the DRM solution, I was not able to access the document that I had purchased again most of these issues were either not disclosed at all or obfuscated. I am sure that many fewer people would choose to download a soft copy over buying a hard copy if the following limitations were clearly disclosed prior to purchase:

1. Opening file requires Internet access and validation EACH AND EVERY TIME YOU OPEN THE FILE (including user name, password, and 1-2 minute wait to unlock); the file needs to be re-validated at least every 12 hours
2. Unlocking the DRM requires downloading of a plug-in which brings potential spyware, bugs, and other stability/security risks
3. This plug-in only works on Windoze computers and not on Linux (and at the time I got it not on Macs)
4. This plug-in only works with certain versions of Adobe Acrobat – most glaringly, it did not work with the current Adobe version at the time
5. The plug-in only works with Internet Explorer and doesn’t work with Firefox or presumably other browsers
6. The spdf file “expires” after 6 months” analogous to writing a document in disappearing ink
7. The spdf file cannot be saved in other unburdened formats for easier personal use. Note you cannot even print to file
8. Even after validation, individual phrases or tables cannot be copied and pasted for valid “fair use” of copyrighted material
9. The spdf file cannot even be read on another computer without a complicated process of uninstalling and reinstalling the key

So, the solution is as follows:
1. Set up a cups “print to pdf” virtual printer on a separate Linux machine (a similar approach probably should work for a Mac). I used the freely available cups-pdf program. Make sure printer is samba shareable over the local network
2. Browse to the pdf printer from your Windoze machine and double-click to install a printer driver (I used the already present HP 1200C/PS driver which enables high resolution color pdfs)
3. Print to the resulting driver. The .pdf file will then be (automatically) saved to your Linux Desktop

Note more detailed info on how to set up a samba (Windows) networked print-to-pdf printer is available at this link (it’s actually easier than it looks but this link includes lots of extra config details, many of which may not be necessary if you are using a Linux distro with a good default samba and cups configuration).

Interestingly, I was initially concerned that passing the text from spdf to postscript and then back to pdf would both lose resolution and increase the file size substantially. However, on my limited testing, I have noted no resolution loss (either of text or graphics, even at 1600% blow-up) and the file size only increased by 30%. Also, since you couldn’t cut and paste text in the original, you don’t lose anything here in the copy. My guess is that the spdf format is already missing all/most of the meta-information and is basically stored similar to a postscript page description language.

As a side note, such workarounds once again illustrate the futility of using DRM — even more so badly implemented DRM that prevents the user from reasonably using his/her legitimately purchased materials. In fact, had spdf been reasonably implemented, I would never have been motivated to spend the time to figure out a way to save it to standard pdf just so that it could be used as intended. Furthermore, were it not for my frustration with this crippled pdf format and the devious way it was sold, I would never have dreamed of posting my personal solution to the net. The solution though is laughably simple and does’t even involve hacking or reverse engineering the DRM itself – just printing it to a pdf printer on a local network.

Net result is I have had a bad customer experience with Harvard Business School and Sealed Media, making me less likely to purchase from them again and overall making me less sympathetic to the DRM interests of copyright holders in general. Ugh, DRM. Yuck!

Related Posts: On this day...

Reader Feedback

50 Responses to “HOWTO: Defeat Sealed PDFs, A DRM Nightmare”

  1. Jimmy says:

    I love you guys. The XPS method as stated by the first post, works like a charm. Awesome find dude.

  2. Jake says:

    None of these methods work any longer (I havn’t tested the Linux method as I don’t have a Linux distro. installed). I will try it sometime because I bought a 65 dollar book and now I can’t even use it at all.

    This is crap…

  3. be sure to try the Linux methods as they will work out the best.

  4. Jake says:

    For the record, the XML method still works for Adobe Digital Editions files that you can print but does not work for Oracle’s “Sealed Media”.

    Check this out though.

    I was able to make a fake printer and tell it to save the .spl (spool) files that it sends to the printer. So, I “print” the ebook and save the .spl file. Now, there is only one program I have found that converts .spl data to .pdf and image files, however, when I do that the text comes out all garbled:

    ¾¼
    ¿ ¾ ½ ¿¾ ¿½
    ï
    Î ÛÚÛÎ ÛÒÝÛ ÐßÙÛÍ
    ݱ°§®·¹ ¸ ¬ î ððé Ì ¸ ±³ ±
    ² Ô»¿®² ·² ¹ ô ײ½ò ß´´ Î ·¹ ¸ ¬Î
    »
    »®ª»¼ò Ó¿§ ² ±¬ ¾» ½±° ·»¼ô
    ½¿² ² »¼ô ±® ¼«° ´·½¿¬»¼ô ·² ©¸ ±´» ±® ·² °¿®¬ò
    Ô·½»²
    »¼ ¬±æ³ ¿¹½±ò
    «°°à ¹ ³ ¿·´ò½±³

    This is supposed to be some Calculus equations and some text. How could I go about ‘fixing’ this I wonder? Perhaps I’ll just try the Linux method…

  5. I’ve been thinking about this from the writer’s side. How does on protect an ebook?
    The answer is that one can’t.
    I’ve been using OCR to read protected text for years (and that for what I consider to be legitimate personnal use thereof).
    Naturally the producers of text have caught up on this by blocking screen shots as described in the article.

    Anybody can use their numeric camera and put the images through OCR to recreate the text. Somewhat cumbersome for 600 pages mentionned above! So this will be improved upon by some kind of hardware converter reading from the video screen output and sending the image to a usb input. Then the OCR program will interpret the image to recreate the text.

    This argument can be transposed to any form of medias, sound or video (which would need three recordings and remasterising) . Recent events suggest that the “protectors” know that they have lost so they are going to stop using drm.
    Victory?
    No, to survive they are going to have to put publicity in everything published. So we are going to have to find a method of eliminating publicity from documents.
    It’s a “cat and mouse game”.

    Paul_wi11iams

  6. Jake says:

    Okay, I figured it out. I sent the book to a ‘fake’ postscript printer and told it to save the spool file. Then I just used Adobe Distiller to convert the postscript file to a .pdf. It worked great and I got a perfect copy of the original document. Unfortunately, this doesn’t seem to work on documents with actual images instead of ‘postscript rendering’. With these, you can send them to XPS printer or, if all else fails, just printscreen them (I haven’t found one yet that blocks printscreens, even sealedmedia).

    The last straw for me was when I paid 120 dollars for an ebook that I was allowed to print. After I printed 20 pages it started printing blank pages. I called them and they told me “due to copyright restrictions we have disabled printing on all ebooks indefinitely”. What the hell? How is that not stealing (from me)? I paid for a product and then they stole part of it back (the most useful part).

    Now the only way for me to use this 100+ dollar book is to take my laptop to class but, guess what? This professor doesn’t allow laptops in class!

    If only I had enough money to sue because they won’t give me a refund. Should I consult the BBB?

  7. Brittany says:

    I can’t even get mine to show up on my computer let alone print it….I don’t understand how it is legal to sell a book that no one can open the files for…..any help would be appreciated. I can’t figure out how to change the spdf to a .doc and read it in a program that I have.

  8. Tiffany says:

    I tried to use xps and onenote 2007 methods. However, I got the message said “A required resource cannot be loaded”. Please advise.

  9. Drew says:

    DRM SUCKS!

    …I love how most the websites that “sell” this crap (ichapters etc) offer the first chapter free of charge which arrives on your hard drive in a clean editable .pdf format, just bait for the fish…yuck.

  10. Noah says:

    Ok guys and gals, I figured out a much easier way to do this if you are stuck on a windows computer. Also, the only additional software you need to download is called SPLview, which is free. Google it.

    So, with any windows computer, whenever you print something, windows creates a image file to send to the printer. That img file is a .SPL file, which is temporarily stored in your “C:\Windows\System32\spool\PRINTERS” folder. The problem with copying this temporary file is just that, it’s temporary. As soon as the file is complete, it’s sent to the printer and then deleted automatically.

    This is the simply part. Take the paper out of your printer. Windows will compile the SPL file and try to push it to your printer, but if your printer is out of paper, it will wait for you to load more paper. While it is waiting, windows will not delete that file until your printer is error free.

    So, all you have to do is remove your paper, print the .spdf using a regular printer, navigate to your “C:\Windows\System32\spool\PRINTERS” folder, wait for the file to finish compiling it will take a couple seconds to 20 seconds depending on how big your document is), and copy the .SPL file to a different folder somewhere on your computer.

    Now you can use SPLview to open up the document, with no quality loss, and save as a .PDF or whatever the hell format it is you like.

    Screw you Harvard, you just got semi-owned by a slow community college student.

Leave a Reply

You must be logged in to post a comment.