In what will be a major, precedent-setting case, the state of New York sued Network Associates earlier this month to end the restrictions that the company claims over its users, restrictions that block publication of benchmark results as well as unauthorized product reviews.
New Yorks attorney general attacks these censorship clauses on several grounds (the lawsuit can be found via www.eweek.com/links), the most significant (in my view) being that they are an illegal limitation of fair comment rights.
According to the lawsuit, Network Associates has exercised this clause on at least one occasion, in 1999, to demand a "correction/retraction" of a Gauntlet firewall review printed in the publication Network World.
I can vouch for the fact that this is not an isolated case. On several occasions, various vendors have threatened us at eWeek with this type of clause in an effort to prevent us from measuring and then describing the performance of their products (a database case, a mail server case and a high-end router case come immediately to mind).
However, we at eWeek are committed to comparative benchmarking because it provides our readers with information that helps them make informed purchases.
For example, as far as Im aware, the comparative database benchmark Special Report published in this issue is the first time this type of testing has been done since 1993, a personally embarrassing gap in my coverage of this beat and in consumer knowledge—and one due directly to legal threats by database vendors.
There are good arguments that copyright fair-use rights and First Amendment free speech rights supersede contract law in this area (even if shrink-wrapped licenses are found to be binding contracts, which itself is not a sure thing).
The usual vendor justification for a media gag clause is that it exists to prevent consumers from being exposed to poorly executed tests. Please, spare us the paternalism.
The right way for benchmarks to be judged is in the public sphere, where consumers can decide for themselves (amid free and open debate) the credibility of each source and the relevance of a benchmark to their workloads. The more data published, the better. Information should be free.
How have software license clauses affected you? E-mail me at email@example.com.