This work is licensed under a Creative Commons Attribution 4.0 International License.
by
Andy Oram
August 20, 1999
The spat between America Online and Microsoft over instant messaging has led many to unfurl banners in support of open standards, product compatibility, and competition. People concerned with these policies should take a look at the proposed Uniform Computer Information Transaction Act, which would determine rules for the distribution of software and computerized information. Given final form earlier this month, it will now be submitted to all 50 state legislatures for approval.
In the AOL/Microsoft controversy, Microsoft started innocently enough by using the open, published AOL standard to create a messaging service that could interoperate with AOL’s to permit conversations between MSN and AOL users. But since then, AOL has repeatedly made changes to its service—and Microsoft has almost as quickly put out a new version of its product that restores compatibility.
Although the July 24 New York Times said Microsoft used reverse engineering to achieve compatibility, a Microsoft spokesperson simply told me, “we used standard development procedures, lots of testing, trial and error.” That was a smart answer to stave off charges of hypocrisy, because like many software companies (including Netscape) Microsoft licenses contain a clause saying “You may not reverse engineer, decompile, or disassemble the software product…” Whatever they did, the way Microsoft chased the AOL protocol round and round the mulberry bush remains a good reminder of how important it is to permit engineering practices that contribute to compatibility and interoperability.
And thus one of the central controversies over UCITA. The goal of this 356-page draft is to improve the legal environment for software development on several levels:
It decrees that software and information products are licensed, not sold. We are all familiar with the “shrink-wrap licenses” that fall out of software packages or confront us when we download a plug-in; UCITA claims to eliminate ambiguities in the ways courts have accepted the licenses.
It provides a legal framework for technologies that enforce licenses, like digital signatures, ways to identify license violations, and “self-help” measures. (That last, innocuous-looking term refers to time bombs and other technical means a vendor can use to shut down software and perhaps even corrupt the data you’ve created with the software.)
It lays out a number of rules assigning responsibility for bugs or incorrect information.
Mark Nebergall, Vice President and Counsel for the Software and Information Industry Association, lauds UCITA and predicts, “Some states will rush to enact this: states that want to attract information industries.” Not so fast. UCITA has developed one of the broadest opposition coalitions I have seen in the history of intellectual property. The list of opponents reads like a directory of organizations in consumer rights, software development, entertainment, and information. UCITA didn’t even make the Federal Trade Commission happy.
Originally the provisions were meant to be part of the Uniform Commercial Code that governs business regulation throughout the country, and were called UCC Article 2B. But one of the organizations responsible for developing the UCC, the American Law Institute, withdrew its support. So the other organization, the National Conference of Commissioners on Uniform State Laws, completed the work and brought it to the states under the name UCITA.
What gives all these opponents the shivers? UCITA has been criticized for:
Shielding software manufacturers from liability for bugs and misrepresentation.
Hiding important license information from users, and requiring customers to bear costs of returning software if they don’t like the license.
Keeping people from transferring software to someone else, the way people commonly trade books or records. (This would also affect companies that sell business units or merge.)
Shifting the burden of accuracy from people contracting for technical work onto the writers performing the work.
Suppressing reverse engineering—the topic of this article.
The NCCUSL made changes in response to criticism, so that UCITA now states the importance of free speech, competition, and fair use. It also states that courts could overturn provisions that they see violating these principles or other compelling public policies. But many details are left up to precedents that will be developed over years of court cases. That’s how the situation would be if UCITA didn’t pass, too. The drafters of UCITA, of course, did not claim to be making social policy. But personally, I can’t understand why a 356-page document had to leave the status of fundamental rights ambiguous, when the authors could spend dozens upon dozens of pages specifying what it means for a customer to “agree” to terms or how a company can prove that an agreement was broken.
In defending UCITA, Nebergall claims it actually offers consumers more protections than current precedent. For instance:
If you don’t see the license until you’ve paid for the product, you can get your money back.
Businesses purchasing single copies of mass-market software get protection similar to individual customers.
The much-criticized self-help measures are hemmed in with lots of restrictions not found in current law.
But when it comes to reverse engineering, Nebergall says a prohibition in a shrink-wrap license would probably be binding under UCITA. In the absence of UCITA, according to software engineer and lawyer Cem Kaner, “No court has ever upheld a ban on reverse engineering for mass-market software.”
Like other provisions in UCITA, a prohibition could be overturned by law. But current law is not too strong. The flagship copyright law of the decade, the 1998 Digital Millennium Copyright Act, permits reverse engineering “for achieving interoperability.” Sounds good; if Corel wants to create a word processing product that accepts .DOC files, that’s all for the benefit of interoperability, isn’t it?
But an aggrieved company could plausibly claim that reverse-engineered products constitute really competition, not just interoperability —so a prohibition on reverse-engineering might stand. And as Kaner points out, reverse engineering has many legitimate purposes that might be squelched by a shrink-wrap license.
Reverse engineering raises many of the same questions as the user interface or “look-and-feel” copyright suits ten years ago. Both issues raise the questions of what is the true intellectual property in software, and how important it is to promote new innovations or competition in comparison to protecting earlier innovations. Where your sympathies fall will determine whether you think UCITA is fair.
This work is licensed under a Creative Commons Attribution 4.0 International License.
Andy Oram is an editor at O’Reilly Media. This article represents his views only. It was originally published in the online magazine Web Review.