Dealing With a Copyright Takedown Request?

George Maschke writes "I recently received a takedown notice from a corporate lawyer demanding that I remove a post on my Web site's message board. It purportedly lists the first 75 of 567 questions on the MMPI-2 paper-and-pencil psychological test. It seems to me that such posting of a limited amount copyrighted material for discussion purposes on a public-interest, non-profit Web site falls within the scope of the fair use exemption of US copyright law. I have thus declined to remove the post. I believe that the corporation in question is seeking to chill public discussion of its test, which applicants for employment with many governmental agencies are required to complete. I would be interested in this community's thoughts on the matter."

Comments

Advice

[MightyMartian]

I'm sure most people will agree that what you've done falls well within the realm of fair use. But you know very well that you're going to need to talk to legal counsel with expertise in copyright matters, and that means money. Maybe someone with contacts in the Electronic Frontier Foundation could give you a hand. Sometimes having a lawyer responding to the guy making the threat is enough to make them back down.

Re:Advice

[yali]

Fair use is an affirmative defense, you can't just claim it as a right but have to prove your use was "fair" in court.

Mod parent up -- this quote is key. Popular folklore on the Internets holds that there is a certain percentage of material you can post that qualifies as fair use. That is bogus. Fair use is a judgment call based on a balance of four factors. From the linked Stanford Law site: "The only way to get a definitive answer on whether a particular use is a fair use is to have it resolved in federal court."

The copyright holders probably would argue that the amount of quoted material is excessive (one of the 4 factors) and that simply posting the items for discussion does not have adequate "transformative value" (another factor). Furthermore, they would probably argue that copying those test items will have a significant detrimental effect on the market for their product (yet another of the factors). On this point, they may well have the American Psychological Association backing them up. The official position of the APA is that psychological tests require careful protection because disclosing their content can invalidate the tests. If this went to court, judge would probably be strongly influenced by a friend-of-the-court brief from the world's largest professional society of psychologists.

Disclaimer: I am not a lawyer either.

75 of 567 is *NOT* "limited amount"

[Anonymous Coward]

Sorry, but 13% of the content is not "limited". If I reprint 85 pages of "Harry Potter and the Half-Blood Prince", would you consider that a "limited amount"? I don't think the publisher would... Not to mention you say "the first 75"... If it had been a 'random' 75, with commentary and discussion for each one; that would be one thing. Even then, it would be dubious. Even if I split it into one-paragraph sections, with commentary and discussion between each paragraph, 85 pages worth of Harry Potter would be difficult to claim under fair use.

For example, the usage of 400 words out of a 500 page book was considered infringement: Harper Row vs. Nation Enterprises.

Now, according to the wonderful DMCA, if you take the material down now, *YOU* are safe from prosecution. If you present the user of your board the opportunity to protest, then if they want to put it back up, *THEY* become the responsible party.

Re:Well, It Seems You Have Already Taken It Down

[fractoid]

And this is precisely the problem with the DMCA. A large company who retain a team of lawyers on salary will have no problem firing off a few dozen such takedown notices. The cost of hiring a lawyer to screen them all for actual, valid complaints becomes rapidly prohibitive for the private citizen on the other end.

This is why the DMCA needs strongly punitive measures for repeat posting of unwarranted takedown notices for economic purposes.

EXPLANATION: How the MMPI test works

[Markmarkmark]

I noticed several posts pointing out some of the seemingly silly questions (ie "My hands and feet get cold"). They may in fact be silly but there is reasoning behind them. I went and actually did some reading up on how the test is supposed to work. There are 8 different major scales measured and several other more minor ones too. For example Scale 1 is essentially looking at Hypochondria, a person's tendency to be really focused on (and maybe whiny) about every little ache and pain. The test understands that everyone has some stuff wrong with them and certain physical peeves too, so you're supposed to mark some of the stuff "T". But if you look at the questions, there are a bunch about this physical stuff and they are all over the place. If you put a "T" by a whole lot of them, then the test scores you higher on this scale. If you put an "F" by all of them, the test basically scores your "truthfulness on test questions" lower because these are things that everyone should complain about a few of.

The issue of test validity is a big deal and dealt with in different ways including checking for truthfulness by asking the same thing in a different way in different parts of the test. There are a bunch of these question pairs and there are some set up for consistent answers being "T/T", "T/F", and "F/F". There are also question sequences in the back half of the test designed to detect if the user is just starting to mostly randomly check or barely skim questions. Too high on this and the test is reflected as invalid.

Gaming the test is not as easy as it might seem at first glance. Some questions can be taken at face value, like "I sometimes think about killing myself". If you check that one "T" along with some other similar questions then you may well be suicidal. However, there are other questions that state mildly negative personality traits that most people have. If you refuse to admit to any of them then the test scores you as either trying to present an unrealistically positive image or as having an unrealistic self-image/ego. Answering some of those type questions with a "T" will get the test to paint you as a self-confident personality with a healthy self-image that feels no need to hide common human foibles.

Personally, I'm a skeptic of these kinds of tests. I think they may work to some degree in some scenarios with some people but there will be other scenarios or people for which the test will largely fail. This particular test is also susceptible to interpretation error. Some evaluators tend to focus in on individual scales but what I read says that that over-simple approach almost always yields skewed results. To get an accurate scoring the evaluator must consider the scales together. In large scale testing of different populations, the experts in this claim to have identified different groupings, for example two particular scales elevated while a third specific scale is lowered may represent a certain personality trait (ie rebelliousness or conformity). It's also said that the evaluator *must* have accurate background info on the subject (ie record of physical violence, manic behavior, etc). These factors can apparently change the assessment significantly.