1996 Cyberspace Law Sample Final Exam Answer by Eric Goldman


USF School of Law/Cyberspace Law/Spring 1996

Professor Eric Schlachter (eschlach@netcom.com)

This document discusses some of the points I was looking for in my Spring 1996 exam.  There were many paths to success, although most of the best answers addressed these points at least in passing.  In general, I was very pleased with your performance on the exam and I felt that most of you identified and properly weighted the issues of each question.  In terms of grading, there were 11 As, 18 Bs and 3 Cs.

Question 1

Most of you correctly identified this question as pertaining to sysop liability for defamation.  Those of you who treated this question primarily as a copyright question missed the call of the question.  I would have organized an answer as follows:

1. Is the file actually defamatory?  Defamation is a defamatory statement that is published and injurious to reputation.  Imputing a sexually transmitted disease to a person is defamatory per se, is published (in the defamation sense) at least to the sysop and to any others who downloaded the file, and would tend to be injurious to reputation.  Some additional information would have been helpful on this point:

* was it true?  (absolute defense to defamation action)

* was it a matter of public concern?  (doubtful but possible)

* was it about a public figure?  (recognizing that the public figure definition might stretch in online communities)

* was it a parody?  was it an assertion of opinion rather than fact?

If the BBS can show that there is no defamation action, no further action would be required.

2. Is the BBS a publisher as a matter of law?  If the BBS is a publisher, then the BBS is probably already liable and the best it can do is mitigate damages.  If the BBS is a distributor, then the law of republication applies and the BBS will be liable as a republisher if it knows or should have known that the file was defamatory and does not remove the file within a reasonable period of time.  To determine whether the BBS was a publisher as a matter of law, the facts should have been compared to the facts in the Stratton Oakmont and Cubby cases.

* The fact that the BBS deployed an automatic virus check was meant to hearken back to the dumb word filters deployed by Prodigy.  Note that unlike a word filter, a virus checker is almost an absolute necessity for the proper functioning of a file exchange, and I personally would consider a BBS negligent if it failed to deploy a virus checker.

* The removal of pirated software is not really editorial control (see below), although it does hint at the “damned if you do, damned if you don’t” problem of sysop maintenance.

* The moving of files by category is the most problematic and the most like editorial control in my mind.  Personally, I would like to have known if this review and moving was ad hoc or part of a systematic plan.

As you might expect, the decision could go either way.  Personally, I think that Susan’s behavior should not  make the BBS into a publisher as a matter of law, but then again I disagree with the result in Stratton Oakmont.

One point to consider: all of Susan’s behavior applied to the software exchange and not the GIF area.  Could Susan argue that the GIF area was unfiltered but abutted the filtered area of the software exchange?

3. Even if the BBS was a distributor, now that the BBS had “knowledge” that the file may be defamatory, the BBS must promptly remove the file to avoid liability as a republisher.  Other useful actions would be to offer a “retraction” or to encourage the person to exercise their own right of reply.

Many of you rightly pointed out that removal of the file here would create future evidence that the BBS was a publisher as a matter of law.  I believe, however, that this should not be the state of the law and I would cite the court’s analysis in the Netcom case for the proposition that removal of liability-creating material does not otherwise indicate the exercise of control.  More to the point, although we did not discuss this in class, the CDA may offer some hope here–Section 230(c)(1) says “No provider or user of an interactive computer service shall be treated as the publisher or speaker of any information provided by another information content provider.”  The legislative history (such that it is) indicates that this provision was inserted to overturn the result in Stratton Oakmont.  Until I get more definitive analysis, I am not prepared to rely on this provision globally, but all of you should consider this language if the issue arises.

4. Other issues worth a quick mention were invasion of privacy (which we did not really discuss in class but could definitely be implicated by the file), obscenity (a stretch), CDA/indecent material (possible although we did not get into too many parameters of a CDA action) and copyright (is Susan in danger of a copyright action generally?).

5. Some common mistakes:

* many of you tried to apply to copyright analytical structure to the “republisher” situation.  Although they are analogous, the bodies of law are completely different.

* a number of you thought disclaimers might help here.  In addition to the difficulties of properly forming binding disclaimers, these will only be effective as between the parties involved and not third parties.  A user agreement could have an indemnity clause in it, where the user indemnifies the BBS for the tortious posting, but this is unlikely in practice to amount to much.

* Finally, a few of you suggested that Susan close down the file exchange altogether.  This advice is aggressive given the call of the question and a very conservative approach from a legal standpoint.

Question 2

As I indicated, Question 2 is a real life story.  Indeed, the story was even worse, in that Soundtouch was truly knocking on the door to bankruptcy and Javelin was a key creditor that gave Soundtouch an option: either enter into the license agreement or Javelin would throw Soundtouch into bankruptcy the very next day (which was a real threat).  I was completely new to the client when I was given the Javelin license (and knew nothing about how Soundtouch had developed its IP) and given less than 24 hours to develop a response.  If I was too aggressive, Javelin would charge Soundtouch with not being serious about the negotiations and file the bankruptcy paperwork.  If I was too soft, Soundtouch would blow all of its assets by entering the license and have nothing left for its stakeholders and stockholders.  Ah, the joys of being a lawyer.  Fortunately (at least for my malpractice insurer), Soundtouch decided to nix the license for other reasons and is plodding along in bankruptcy as we speak (after stiffing our firm for $60k…).

In part I, I really only intended you to discuss whether Soundtouch’s use is fair use.  In the question, Soundtouch freely admitted that it infringed the CDs to build its database, so there was little danger by Soundtouch conceding infringement and arguing fair use.  I should note that multiple 106 rights were implicated–reproduction, distribution, derivative works and public performance/digital performance (some of you improperly asserted only one or two of the rights were implicated).  The fair use factors:

* Commercial v. educational/nonprofit use.  This use tends to look like commercial use, although the courts have said that this factor is not a per se presumption against use by for-profit companies.  This factor weighs against Soundtouch.

* Fact/fiction and published/unpublished.  Many of you misapplied this factor, but it’s a stupid one anyway.  Songs are published (more like fair use) but are fiction (less like fair use).  This factor is a toss-up.

* Amount and substantiality of the use.  Soundtouch used 30-45 seconds of 3-5 minute long songs, or 10-25%.  However, because Soundtouch took the “best” portions of the long, the substantiality rises. Personally, I believe this factor weighs against Soundtouch.

* Effect on the market.  This is the most important factor and also the toughest.  Soundtouch believed that the database would actually promote the sale of CDs and therefore increase the market. However, the relevant market may be the market for soundbytes, not the downstream market for songs.  Further, some courts misapply the factor by making it tautological–if someone thinks the IP is important enough to use, they must value it and would have been willing to pay for it.  Finally, and most importantly, there are licensing schemes for music, even portions of songs, so the court would have little problem quantifying how much the use was costing the copyright holders.  Just ask any movie or TV producers how often they rely on fair use for even snippets of songs they use in their productions.

Clearly, I believe fair use is precarious at best.  Most importantly, if the fair use argument actually works, any attempts to license the database are extremely difficult (see below).

Some of you mentioned that the record companies had to be aware of this scheme and their failure to act presumably could be construed as some sort of implied license.  This argument is not a bad one, but if even a few of the soundbytes turned out to be infringing, Soundtouch would still be on the hook for huge dollars.  Again, not the best way to build your key corporate asset.

Part II was really a difficult question and I was impressed how many of you got at least to the right issue.  Some of you even shocked me by coming up with the answer I was looking for, which I thought might be so obscure as to be impossible.

If Soundtouch’s use was not fair use, clearly the license is a huge problem since Soundtouch would be liable for direct infringement (reproduction/distribution to Javelin) and contributory infringement for Javelin’s use.  Javelin did ask for an indemnity from Soundtouch, but query how useful that indemnity was given that Soundtouch was sounding its death knell.  Some of you tried to get Javelin to indemnify Soundtouch for infringement–nice try, but in context extremely unlikely.

Some of you suggested that Soundtouch procure the appropriate licenses.  This is a good suggestion except that administratively the cost would be huge.  Alternatively, Soundtouch could pay statutory fees to BMI/ASCAP, but query if these statutory fees make sense given the Net economics (i.e., will Soundtouch have to pay for every download of a promotional clip?).

If Soundtouch’s use was fair use, then Soundtouch’s intellectual property was actually quite thin.  It clearly had good intellectual property in the software, but the database starts to look like the databases in Feist and ProCD–chunks of data which the possessor could not copyright but which had some selection/coordination/arrangement aspects that created a thin compilation copyright.

As many of you properly deduced, giving Javelin the right to make derivative works which Javelin owned meant that Javelin would be able to quickly evade the scope of the license by modifying the database.  Even though the general rule is that a derivative work may not be used without a license to the underlying work, Javelin could avoid using the copyrighted portion of the “underlying work” without much effort and then would have complete ownership of the resulting product.  At that point, the license is meaningless and Soundtouch has to compete with Javelin.

Many of you suggested that Javelin should assign ownership of the derivative works back to Soundtouch.  In addition to being a change in the parties’ intent (watch the call of the question!), query whether this works–is a modification to a largely uncopyrighted work a “derivative work”?  The same analysis applies to the suggestion that Soundtouch not permit Javelin to make derivative works.  In other words, I believe Javelin’s mere possession of a database built on fair use would create an unstable situation.

Many of you tried to get control over the database by relying on Soundtouch’s copyright in the software.  For exam purposes, this was clear thinking; however, in practice, the software was rather simple and it would have been an easy task for Javelin to develop its own software.

A number of you tried to trot out a laundry list of technologies from my lecture on technology supplements to copyright.  This was OK but many of the technologies simply did not apply.

The clear winning answer was for Soundtouch to keep the database on its server and allow Javelin customers to access the database via Javelin-specific hypertext links over the Net–in other words, converting the license agreement into a database access agreement.  This way, Soundtouch never has to part with a complete copy of the database and yet could provide Javelin and its customers with all the benefits of possessing the database.

Question 3

Question 3 is also a true story.  Since this was not really my client, I do not know if the deal was ever consummated (but I hope not!).  Ironically, when I pointed out to George that gambling on the Net was presumptively illegal, his initial response was “well, then we really need to make sure that we get some good warranties and indemnities…”!  Also, High Rollers was represented by a large and generally reputable Palo Alto firm (let’s call it “Billsome Obscenely”), which led George to believe that surely Billsome must have done some due diligence about gambling on the Net as part of representing High Rollers.  As I stressed in class, this kind of thinking can create all kinds of problems, especially since I later discovered that Billsome had only been representing High Rollers for only 1 week and the attorney working on the account was a general corporate lawyer more known for his biotech practice…  In this situation, I for one would not bet my client’s personal freedom on Billsome’s analysis.

Some questions worth asking:

* Was High Rollers going to use the software for gambling or just for play/fun?  If it is the latter, there may not be that many legal issues–but then again, why the heavy indemnity and the 3% royalty…?

* Did High Rollers have a scheme to ensure that only gamblers who could legally gamble on the site would have access?  This would make me more comfortable if true, but I would be very suspicious that this could be implemented at a reasonable cost.  Note that disclaimers (even those purporting to “deny” access to people who are not entitled to access under their local laws) are very likely to be insufficient to prevent criminal liability, and those of you who recommended this approach should reconsider the validity of this recommendation.

* Where were High Rollers and Online Motion located?  If both are located outside of the reach of any extradition efforts, maybe my attitude would change.  However, remember that it is generally against professional ethics to advise your clients to break the law.

As almost all of you reasoned, gambling on the Net creates sticky legal problems.  In addition to federal laws that probably prohibit gambling on the Net, states (including but not limited to Minnesota) have laws that prohibit gambling by their residents.  Even advertising about gambling is restricted.  In other words, successfully implementing online gambling requires extreme caution–if it even can be done at all.

As many of you pointed out, our client is not engaging in gambling on the Net but providing a tool or device for gambling.  You should note that gambling devices are extremely regulated even in Nevada (basically to ensure that they are fair to the gambler).  In any event, many of you properly noted the potential liability as an aider or abetter or accomplice and perhaps co-conspirator (those of you who used “vicarious” or “contributory” liability were on the right point but again were using copyright language).  In addition to generally participating in the scheme, note that Online Motion was designing the device to High Roller’s specifications and participating in the profits over time (the 3% royalty).  Therefore, if High Rollers is breaking the law, Online Motion is likely to go down with it.

Most of you also pointed out that the indemnity is worthless.  First, there is no meaningful indemnity for jail time.  Second, it is probably void against public policy to indemnify for criminal fines. Finally, High Rollers is unlikely to be around long enough to provide a meaningful indemnity.

Many of you talked about venue.  You may recall that criminal venue can attach where ever the crime is committed; hence, Minnesota will prosecute people who violate Minnesota law in Minnesota courts so long as they can extradite.  A lot of you talked about civil procedure and personal jurisdiction limits under the constitution, which was fine but not the major issue.

A number of you recommended that Online Motion “sell” the software or otherwise convert the fee into a one-time payment.  Although this is generally not bad advice given the precariousness of the future cash flows, I do not believe this advice minimizes the likelihood that Online Motion would be liable criminally (particularly since a gambling device is subject to its own set of regulation).  It reminds me of the cases in criminal law where the vendor sells a gun to someone knowing of their criminal intent–the vendor can still be criminally liable for aiding and abetting in such circumstances.