Archive for August, 2010

5 Minute FAQ: License Agreements vs. SaaS Agreements

Thursday, August 26th, 2010


Photo Courtesy NASA

Q: Computing used to be based on the model of the end user running another party’s software on its local computer through a client-server or fat client architecture. Now the business model is moving increasingly towards cloud computing and software as a service. If the end-user no longer has someone else’s software on his computer, does that mean the license agreements are no longer necessary?

First, let’s start off with some definitions. Cloud computing means computing where data is input and stored on a remote third party’s computer, on the Internet or “cloud.” Think Facebook. Software as a service or “SaaS” is a subset of cloud computing where not just data is stored, but also processed via a specific application in the cloud, which is often being used concurrently by many other users. Think Salesforce.com and Google Apps. The end user of SaaS pays a fee over the time of usage, using a subscription or utility model, in contrast to local software where the cost is usually paid all upfront. Also, since the SaaS user does not possess SaaS intellectual property, but accesses it remotely through a Web browser, it pays for access to, rather than use of, the IP.

Q: Well, that was my question. If there is no license of third-party IP, then is a license agreement even necessary?

I was not ignoring your question. You are correct– although there is technically no license involved in an SaaS business model, a user agreement most definitely is necessary, call it what you will. At the end of the day, the difference between an SaaS agreement and a typical end user software license agreement is more a matter of degree than kind. The meta-theme of a software end user license agreement is, “We allow you to use our intellectual property as long as ____ .” The meta-theme of an SaaS agreement is, “We allow you to access our service as long as ____ .” Despite the difference business models, large chunks of both kinds of agreements are interchangeable, because the software licensor and the SaaS service offeror are addressing many of the same business risks. Also, note that some SaaS offerings require installation of software on the local client in combination with access through a web browser, in which case the agreement would be a hybrid even closer to a traditional software end user license agreement.

Q: In what ways is an SaaS agreement different from an EULA?

Obviously, the SaaS agreement would not contain license grant language, except if it were a hybrid, as noted above. Also, an SaaS agreement needs to address business risks that are specific to a cloud environment, such as:

  • Performance and uptime guarantees and/or SLA’s
  • Data privacy and security
  • Data backups and disaster contingency
  • Data portability, especially in case of nonrenewal of the SaaS agreement
  • Term, termination, and renewal provisions, given the periodic nature of the SaaS subscription model

Besides the license grant provision, a typical SaaS agreement also often omits common EULA provisions relating to maintenance, support, and updates, because it is usually done behind the scenes by the service offeror, on a multi-user basis.

Has Pay for Delay Seen Its Day?

Thursday, August 12th, 2010


Photo by Sage Ross, CC by-sa

What do you call it when a patent holder pays royalties to a possible infringer, and not the other way around?

In the pharmaceutical industry, it is called “reverse payment” or “pay for delay,” a practice in which the inventor of a patented drug pays the developer of a generic substitute not to produce until the patent is nearly expired.

The practice has withstood multiple court challenges, but is coming under increasing fire from the executive and legislative branches of the federal government as anti-competitive.

And a three judge panel of the influential federal Second Circuit Court of Appeals, which recently upheld a reverse payment arrangement, has even taken the unusual step of inviting the challengers to appeal their loss to the entire 10 judge circuit court panel.

The origins of pay for delay are in the Drug Price Competition and Patent Term Restoration Act of 1984, popularly known as the Hatch-Waxman Act. Hatch-Waxman provides that when a developer of a patented drug applies to the Food and Drug Administration for permission to market it, the maker of a generic version of the drug may file a certification that the generic does not infringe the new drug’s patents, or that the new drug’s patents are invalid. The certification itself is deemed an infringement of the new drug’s patents, and sets the stage for the new drug’s inventor to sue the makers of the generic version for patent infringement.

The original logic of Hatch-Waxman was to encourage development of generics without undue cost to patented drugs by resolving infringement issues at an early stage, before either party had invested huge amounts of money in production and marketing. But it did not work out that way. Many innovators chose instead to pay the generics a “reverse payment” not to produce, usually until shortly before the new drug’s patents expire. Often payment to the generic is in the form of a percentage of sales on the patented drug, in effect, a “reverse royalty.”

Reverse payments have been widely criticized as an unreasonable restraint of trade in violation of Section 1 of the Sherman Antitrust Act.

The Federal Trade Commission estimates that pay for delay settlements cost consumers $3.5 billion per year because of the unavailability of cheaper generic drugs.

But pharmaceutical manufacturers reply that the agreements are reasonable commercial agreements to avoid the expense and uncertainty of litigation, especially where the prospective generic manufacturer has not risked significant market entry expenses or infringement damages, and has little to lose in filing a challenge.

And courts for the most part have agreed. The Eleventh Circuit and the Federal Circuit have upheld the practice, while the Sixth Circuit has held it unlawful.

Two of the biggest class-action antitrust challenges to pay for delay were lengthy class action lawsuits decided by the Second Circuit Court of Appeals.

In 2005, the Second Circuit upheld the practice in In Re Tamoxifen Citrate Antitrust Litigation, in which it ruled that reverse payment agreements do not violate antitrust laws, “[u]nless and until the patent is shown to have been procured by fraud, or a suit for its enforcement is shown to be objectively baseless,” or the settlement agreement extends the patent beyond its original scope.

In Re Ciprofloxacin Hydrochloride Antitrust Litigation challenged a settlement of a 1991 patent infringement lawsuit in which Bayer Corporation paid Barr Laboratories reverse payment royalties estimated at nearly $400 million not to manufacture a generic version of Bayer’s patented drug Cipro. A three-judge panel of the court ruled in April that in light of the circuit’s earlier ruling in the Tamoxifen case, it was compelled to dismiss the plaintiffs’ case, because they had not properly alleged that Bayer’s patent was procured by fraud, nor that the underlying patent lawsuit was objectively baseless.

But the appellate panel had significant reservations about the practice of reverse payments as a whole, and invited the plaintiffs to challenge the decision before an en banc hearing of the circuit’s 10 active judges. The plaintiffs accepted the panel’s invitation, and filed a petition for an en banc hearing, which was joined by briefs submitted by the Federal Trade Commission and the Department of Justice criticizing reverse payment settlements. A ruling on the petition is expected shortly.

Sentiment is also mounting against pay for delay in the legislative branch. Senator Orrin Hatch, who lent his name to the Hatch-Waxman Act, has criticized the practice as increasingly anti-competitive. A bill pending in the Senate would greatly curtail pay for delay by presuming that such agreements are anti-competitive, while permitting the parties to rebut that presumption in court. The bill is similar to a measure that has already passed the House of Representatives.

Pay for delay may have seen its day.

Update, 9/17/10: But maybe just not yet. The Second Circuit declined by 9-1 to rehear the Cipro case en banc, which means that pay for delay has withstood yet another legal challenge.

IP & Licensing Trends in Japan

Monday, August 9th, 2010

“It is often said that Japan wins at technology, but loses at business,” said Kimikazu Noumi, chairman of the Innovation Network Corp. of Japan (“INCJ”) at a recent press conference announcing INCJ’s initiative to monetize Japanese intellectual property assets in the life sciences.

This is a welcome reversal of a trend that we have previously noted, in which Japan has not effectively monetized its abundant intellectual property assets, through licensing and similar means.

INCJ is collecting billions of yen from Japanese public and private investors to launch the “Rising Sun Fund” to purchase dormant life sciences patents from universities and public research institutions, package them, and license them to domestic and foreign companies so they can develop new medicines and treatments without fear of infringement lawsuits. The initial areas of concentration will be embryonic stem cells; cancer; Alzheimer’s disease; and biomarkers.

According to the Japanese newspaper Asahi Shimbun, a main purchase target of the fund will be universities, because they use only an estimated 20% of their patents.

A Japanese version of the story from the same newspaper notes that the impetus for the fund was the 2007 arrival in Japan of the American bioventure fund, Intellectual Ventures, and its subsequent purchase of Japanese university research patents for relatively low license fees. The fund founders feared that the technologies would be developed overseas, and that Japan would not share in the fruits of medical innovations developed from patents developed at its own taxpayer-supported universities.

INCJ is expected to invest up to 1 billion yen (approximately $11.7 million), and several private companies, such as pharmaceutical heavyweight Takeda Pharmaceutical, are expected to invest millions of yen more.

On the other side of the coin, Japanese book publishers failure to license e-books has created a phenomenon known in Japan as 自炊 (jisui), or “cooking for oneself,” in which readers purchase hardcopy versions of books, slice out the pages with razor knives, and scan them into PDF form to read on their iPads. Japanese are avid book readers and big fans of compact technology, yet publishers have made only about 50,000 book titles available in e-book form, with few current best-sellers, compared to about 630,000 English language titles available for the Amazon Kindle.


How to Cook Your Own E-Books, Japanese Style

In fact, consumer demand is so great that new businesses have arisen to fill it– companies that will save readers the hassle of slicing and scanning by performing the service for about 200 yen per volume (approximately $2.38). The reader mails in the hard copy, the company e-mails back the digitized book. Between the service fee and the postage, the reader is probably paying a $4 or $5 premium (or 15 to 20 minutes of his time), over the hard copy price, to have his e-book.

It appears that the delay in making more e-book titles available through official channels is due to publishers’ fears that e-books will be sold at discounted prices relative to paper versions (as in the United States), thereby shrinking profits. But it appears that at least some Japanese readers want e-books so badly, they are willing to pay a premium over the hard copy price. It is a shame that publishers have not adapted their business model, and are leaving money on the table.

Update, 8/24/10: for a good survey of the jisui trend, see this article in the Mainichi Daily News (includes link to original Japanese article). It appears that competition in the jisui service bureau industry has pushed the price down to near 100 yen per volume.

Speaking of the iPad, the Washington Post recently ran an interesting article on how the iPad has become a hit with Japanese senior citizens who were not previously computer adept, because of its sleek design and intuitive, user friendly interface. The iPhone is also a smash hit in Japan– when I was in Osaka in late June, there were literally lines around the block of people waiting to purchase the iPhone 4. Many thought that Japanese would never become enthusiastic about a non-Japanese made electronic device, much less a smartphone, a product for which Japan has long had its own specialized, highly competitive market.

Why has Apple succeeded where so many others have failed? As noted in the link above, Japanese smartphones tend to be feature-centric, but somewhat clunky to navigate, while the Apple devices have been designed from the ground up to provide an elegant, seamless user interface among hardware, software, and abundant content. The usually compact size and sleek design of Apple devices also appeal to Japanese tastes.