Do you have questions about the site? Write to bruce at perens.com .
List of all articles we've ever published.
Source Code
Developer's Documentation
RSS
XML Sitemap
XML News Sitemap
![[Valid RSS]](/images/valid-rss.png "Validate my RSS feed")
Novell Apologies Debunked, 2700 People Sign Protest Letter
Two weeks ago I wrote a letter protesting Novell's patent agreement with Microsoft and made it available for other people to sign. I expected 200-300 signatures for this rather technical matter about patents and licensing, but there are 2700 signatures as I write this, and the number keeps increasing. Many of the signers attached notes directed at Novell, filled with emotion. Obviously, the Free Software community feels very strongly about this issue. But Free Software is not the only party represented. Many of the signers identify themselves as recent Novell VARs and institutional customers who will now turn to another Linux distribution.
If you haven't read the letter, you should do so now, as the arguments in it still stand. Since I wrote the letter, Novell and others have raised some arguments of their own, most of which are faulty. I'll deal with them here.
Various people asserted that Novell's patent protection is comparable to the indemnification that Red Hat and HP offer. But nothing could be farther from the truth. Companies that offer indemnification don't have a business relationship with the aggressor, just as insurance companies don't make a contract with the local burglars to deter them from robbing certain houses. Indemnities protect against all potential aggressors, in a similar way to insurance. But the Novell-Microsoft relationship is more like making a contract with the local burglars - Microsoft is both the party making the threat, and the party offering "protection", much as an extortionist would elicit a fee to "protect" a business from his own violence.
I assert that Novell has cheated the GPL programmers. Some point to an out-of-context quote of Richard Stallman that the current GPL section 7 text doesn't rule out Novell's action - as evidence that the GPL programmers have not been cheated. This is like saying that if there's a law against punching someone in the nose, it would be OK to kick noses. The harm still exists, whether the law catches it or not. Stallman has hardly "absolved" Novell. He concurs that the spirit of Novell's agreement with the GPL programmers was violated. He is at work crafting GPL 3 text to poison the Novell-Microsoft patent agreement.
How will that poison work? It will require that parties that offer, or arrange for the offering of, protection from patents regarding the software must grant that same protection to everyone. It's intended to make the patents involved unenforcible against anyone if Microsoft offers patent protection to any party, but it will also stop Novell from taking advantage of that protection if it wishes to incorporate any GPL 3 software. As time goes on, the gap between Novell and other distributions that accept GPL 3 would widen.
Why is GPL section 7 important? It places everyone who would use the software in the same boat regarding risk from patents, and thus aligns their interest in protecting everyone from those patents. Circumventing it, as Novell has done with Microsoft, strikes at the motivation that developers have to produce Free Software. The vast majority of developers prefer the sharing-with-rules of the GPL to making an outright gift of their code using BSD and similar licenses. If it becomes clear that corporations can violate the GPL with impunity, those developers will not be motivated to contribute more code.
Some assert that customers just want a solution that works and won't get them sued. Customers, however, are responsible to make sure that their business can operate in 10 years, not just this year. Thus, the customer's interest is in lasting protection from software patents, not just protection from a single patent holder out of many for only five years. It's in the customer's interest for developers to be protected so that they can develop, and for a healthy market of Open Source - rather than a patent-driven monopoly - to exist so that economic forces will continue to drive prices down, quality and choice up.
Some see Novell's involvement in Open Invention Network as a redeeming factor. OIN holds only 13 patents, they are on various e-commerce techniques. This is a small portfolio that might deter some patent aggressors and not others, and would be entirely ineffective against the patent "trolls", who produce no products other than litigation and thus are not themselves vulnerable to patent enforcement. And yet Novell promotes its involvement in OIN heavily.
Microsoft has already invested in "patent troll" company, Intellectual Ventures. That company could operate as Microsoft's proxy in bringing suit against other businesses and Open Source developers, without placing Microsoft at risk, much as SCO acted as Microsoft's proxy while Microsoft directed captial to it through BayStar Capital.
We must also look askance at Novell's partners in OIN. With only one exception, they are the companies that lobbied most strongly for the European Community to implement greater enforcibility of software patents. It appears that they use OIN as a token effort to provide some very limited protection to Free Software in order to muster justification for implementing increased software patent protection in the face of protests from Free Software advocates.
Ad-hominem criticism of the Free Software community who objected to the deal, or directed at me, was seen in online bulletin boards for discussion of the Novell stock. Ad-hominem criticism usually means your opponent can't muster a logical argument about the topic. Nat Friedman of Novell tried to connect my argument with OSRM, a risk-management company that I was associated with for a year or so. For the record, I am no longer connected with OSRM, have made no money from it, and never will. OSRM's dream was to offer an indemnity that would be independent of any Linux distribution. It might have been a help to the Free Software community if they had been able to do that, but I am not aware that they ever sold any indemnities.
Others criticized the Daniel Ravicher study because OSRM commissioned it. Ravicher was at the time a volunteer counsel for the Free Software Foundation, and has dedicated years of his life to fighting the threat of software patents as leader of the Public Patent Foundation and an attorney for the Software Freedom Law Center. Ravicher has never been able to release the particular patent numbers he found because of a pernicious provision of U.S. patent law: the penalty for looking. Knowing infringement brings a triple damage penalty, while unknowing infringement brings only simple damages. For this reason, most engineers are instructed not to look at patents, and Ravicher chose not to increase the liability of the Linux kernel developers by exposing them to that information. If you wish to see an independent study from Europe, where there is not a penalty for looking, see Your Web Shop is Patented.
Finally, some suggest that we should not shun Novell for their actions as this would hurt the SuSE Linux distribution. This led me to quip that SuSE is the "human shield" of Novell, like the hapless civilians who are herded around some war target. If SuSE is worthy of continued existence, it will continue as Open Source software with or without Novell. The vast majority of the authors of the software in the SuSE distribution do not work for Novell - indeed, I'm one of them. And of course there are several other Linux distributions of similar or higher quality.