Jon’s Radio Comments

September 24, 2006

How translucency could defuse the Turnitin/McClean High controversy

Filed under: Uncategorized — jonsradiocomments @ 6:03 pm

The original item is here.



  1. A hash as you propose would find only the most literal of all copying. Unless the text is canonicalized for white space (not hard, admittedly), something as little as adding a space will lead to a different hash. Of course, for students lazy enough to turn in a verbatim download from the internet deserve what they get. On the flip side, a quoted and properly cited phrase might also be flagged. It falls to the teacher to understand that this is a legitimate usage and not rely solely on the software.

    Comment by David S — September 25, 2006 @ 3:29 am | Reply

  2. A one-way transform is not necessarily distant enough to avoid the problem of a claim that it is a “derived work”.

    There’s probably a good case to be made that it is a “fair use” of a copyrighted work. (Pro: transformative, not published, no effect on original’s market. Con: uses entire original work.)

    But until the case is made and tested in some court, it’s only hypothetical.

    I am not a lawyer.

    Comment by Liudvikas Bukys — September 25, 2006 @ 3:07 pm | Reply

  3. The idea is interesting but the problem is that one way hashes are, purposefully so, very sensitive to the slightest variations in content. Only someone who steals wholesale unaltered content would get caught by such a mechanism. Finally, when a hit is made, without the original sample text to compare to you have no evidence that the accused may challange which seems far less fair.

    FWIW – I think the claims of copyright violation and proof of innocence is without merit. You’re submitting your work to the teacher to be judged on its merits in regards to the assignment and its originality so that you may claim rightful credit for the work. Clearly such an application should be considered fair use of the content for testing purposes and, ironically, also ensure your own copyright is protected against those who would plagurize your work in the future. If the testing company’s use of your original material goes beyond that then I’d say you could persue that action seperately.

    Comment by Ben Scherrey — September 26, 2006 @ 1:51 pm | Reply

  4. > Only someone who steals wholesale unaltered content
    > would get caught by such a mechanism. Finally, when
    > a hit is made without the original sample text to
    > compare to you have no evidence that the accused
    > may challange which seems far less fair.

    Agreed. I addressed these points in a couple of follow-ups appended to the entry. In general I was trying to provoke discussion about translucency as a design principle. Of course when you try to work out the implications in any specific case, like this one, it becomes clear why we rarely apply that principle. It’s hard!

    > Clearly such an application should be considered
    > fair use of the content for testing purposes

    On the copyright point, I agree with you. However it would be nice to arrive at an architecture that did not require the service to retain the students’ content. Partly because of how that will feel to the students, and partly because it’s a liability to the service. TurnItIn will hate to read the headline: “5 million student papers stolen from website”!

    On the question of presumption of guilt, and the adversarial relationship that may or may not be created by use of the service, there’s an interesting discussion to be had. In one of my follow-ups, Liz Lawley from RIT and Daniel Freudberg from McClean High represent to different perspectives.

    Comment by jonsradiocomments — September 27, 2006 @ 8:29 am | Reply

  5. I just came across an absolutely eye-opening article with tons of proof. I had no idea how much Turnitin violates students’ rights.

    The Well-Known Secret about

    Comment by Dan — October 19, 2006 @ 7:18 am | Reply

  6. “To meet the business requirement, they need only be machine-readable versions derived from the human-readable originals.”

    While that statement may be technically correct, Turnitin would never choose to use one-way hash. Why? Because Turnitin would no longer be able to email students’ complete essays to third parties. Yes, Turnitin does that illegal act, but they don’t want the public to know about it. I got the information below from the “Well-Known Secret about” article linked above.

    Despite what Turnitin representatives falsely claim, Turnitin does not simply make a “digital fingerprint” that is indistinguishable and unavailable to the public. To the extreme contrary, Turnitin profits immensely from providing students’ actual work directly to other human beings (without the student authors’ direct permission). In fact, the following statement appears in Turnitin’s user guide for professors:

    “If the paper is from another instructor’s class, we cannot provide direct access to the paper. To view the paper, you must first request permission from the instructor in possession of the paper by clicking the permission request button 4. We will then auto-generate an e-mail detailing your request. If permission to view the paper is granted, a copy of the paper will be sent back to you via e-mail.” (

    Now, “the instructor in possession of the paper” has absolutely no authority whatsoever to grant license for anyone else to duplicate, transmit, or read a student’s intellectual property.

    You may be asking, “How does providing to someone a word-for-word copy of a student’s original paper benefit monetarily?” Well, we’re glad you asked! If you were a responsible professor considering paying for Turnitin, you would not commit to the purchase unless you received a POSITIVE answer to the following question:

    “If Turnitin flags one of my student’s papers as having been plagiarized from a student paper in your database, will I be able to read the original paper from which you claim that the student plagiarized? I need to see the original source myself if I am to confront the student with plagiarism charges, risking the student’s reputation, my job, and a lawsuit against the school district.”

    If Turnitin staff were to rightfully and responsibly answer “No” to all such requests, thereby refusing access altogether to students’ original, word-for-word writing, countless professors would never pay for Turnitin’s uncorroborated results (or not renew existing contracts). After all, a professor can’t convict a student of plagiarism and tarnish the student’s life-long reputation based on hearsay. Turnitin’s revenue would plummet. So, the people at Turnitin apparently choose to roll the “legal dice” by disseminating students’ intellectual property to third parties, hoping that a lawsuit never arises. In the meantime, Turnitin generates at least $10,000,000 in revenue per year by violating the IP rights of the millions of students for whom Turnitin hypocrytically claims to “protect copyright.” This practice of blatant copyright infringement via third-party, public viewing may also violate FERPA regulations because a student’s personal writing is an integral part of his/her educational record.

    Comment by Enrique — October 23, 2006 @ 8:01 pm | Reply

  7. Mclean is spelled, well Mclean, not McClean

    Comment by Anonymous — December 19, 2006 @ 11:54 pm | Reply

RSS feed for comments on this post. TrackBack URI

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

Create a free website or blog at

%d bloggers like this: