Who is being divisive? [LWN.net]

Who is being divisive?

Posted Nov 29, 2006 10:20 UTC (Wed) by jhellan (subscriber, #17103)
In reply to: Who is being divisive? by BrucePerens
Parent article: Who is being divisive?

Ah, the open letter.

2450 people signed it, included the following:

"It is likely that there are many thousands of unlitigated potential infringements within the entire Novell system."

Bruce got 2450 people to say that they believe Linux is guilty of infringing patent rights on a vast scale. Tactically, that isn't the most clever move I've ever seen.

(Log in to post comments)

Unpleasant, but probably true

Posted Nov 29, 2006 14:00 UTC (Wed) by emk (subscriber, #1128) [Link]

Bruce got 2450 people to say that they believe Linux is guilty of infringing patent rights on a vast scale.

As a software author and signatory to Bruce's petition, I have every reason to believe that this statement is correct. The patent office has granted thousands of broad-reaching patents (despite clear prior art), and any program longer than a few thousand lines probably infringes several--even if it was written long before the patents were issued.

Who is being divisive?

Posted Nov 30, 2006 8:47 UTC (Thu) by jhellan (subscriber, #17103) [Link]

The number "many thousand" is based on pure guesswork.

The number 283 for the Linux kernel alone comes from a study commissioned by a company that wanted to sell insurance against that risk. We could certainly take it at face value, but in that case we have to do the same with the Microsoft sponsored studies about Linux.

Who is being divisive?

Posted Nov 30, 2006 15:24 UTC (Thu) by BrucePerens (subscriber, #2510) [Link]

The number 283 comes from Dan Ravicher, who was at the time a volunteer counsel for the Free Software foundation, and is now leader of the Public Patent Foundation and an attorney for the Software Freedom Law Center. OSRM paid for the study, but it didn't tell us anything we did not already know.

Bruce

Who is being divisive?

Posted Dec 1, 2006 8:48 UTC (Fri) by jhellan (subscriber, #17103) [Link]

Exactly. As we both have said, the study was paid for by a company which had a commercial interest in scaring developers. And the results can't be verified, since they are secret.

Ravicher report

Posted Dec 1, 2006 9:23 UTC (Fri) by BrucePerens (subscriber, #2510) [Link]

Actually, the results can't be published because if they were, the Linux programmers might be exposed to treble damages for knowing infringement rather than simple damages for unknowing. Indeed, I asked not to be informed of the specific patents in the study for that reason - it would have increased my legal risk as an occassional kernel programmer.

There is a parallel study, however, that is published, because Europe doesn't have that pernicious "you pay more if you look" law. It's at Your Webshop is Patented.

If you want to verify the Ravicher study, you can do so at www.uspto.gov . Just take a part of the kernel and start typing keywords into the patent search form. But please don't tell us of the specific patents you find. You'd be increasing our legal risk.

Bruce

Ravicher report

Posted Dec 2, 2006 13:56 UTC (Sat) by jhellan (subscriber, #17103) [Link]

So looking at the patents increases your legal risk. You know, I exected that response.

And somehow you seem to believe that signing a statement that you believe that you are infringing thousands of patents does *not* increase your legal risk. Strange.

Ravicher report

Posted Dec 2, 2006 15:45 UTC (Sat) by BrucePerens (subscriber, #2510) [Link]

Engineers everywhere are told "don't look at patents". The concern is that a specific patent text belonging to the plaintiff can be shown to have been known by the defendant. In contrast, knowing that software is generally infringing of patents does not pass the court test necessary for the award of treble damages.

Bruce