Walking With Elephants

Taking on a new challenge

2008-07-16T08:22:00.002-04:00

In my last blog I talked about the PeerToPatent project and how it is attempting to improve the U.S. patent system one patent at a time. As you may see in the press today, my interest in PeerToPatent is not benign. Starting back on June 1 I have joined New York Law School as a visiting professor and as executive director of the new Center for Patent Innovations, home of PeerToPatent.

Those who know me well know I am no fan of either software or business method patents. As covered thoroughly by Jim Bessen and Mike Meurer in their recent book, Patent Failure, the cost of software patents to industry players in the form of infringement litigation exceeds the income being derived from such patents. This is a nonsensical situation, and one of the chief culprits is patents that issue that shouldn't issue. That is what PeerToPatent is seeking to address.

In electing to work to reform the patent system by engaging with it, I am not giving up on the great need for broad patent reform. From my perspective, such changes as apportionment of damages and post-grant opposition remain critical. However, given the strong emotions on both sides of these issues and a nearly divided Congress, such broad reform will certainly not occur this year (given that it is an election year), and it unlikely that either political party will emerge from the elections with sufficient strength to push through the most contentious elements of patent reform.

When approached by Rick Matasar, Dean of New York Law School, and Beth Noveck, the law professor at NYLS who formulated the idea for PeerToPatent and, with the help of a number of corporate sponsors and foundations, garnered the funds to push it to fruition, I simply saw an opportunity to have an immediate impact on the patent system and for the better. PeerToPatent is not simply about improving the identification of prior art (preexisting inventions) for use by patent examiners, it is about citizens engaging with their government to improve decision-making.

Today the U.S. Patent and Trademark Office announced that the pilot program for PeerToPatent is being extended for a second year and that the areas of innovation it covers is being broadened to include e-commerce and business methods. This is a huge opportunity. The first -year success of PeerToPatent is well documented in its annual report. Now we have the opportunity to build from that successful first year to show that PeerToPatent has the ability to scale.

In the coming weeks I will talk about PeerToPatent, why it is important, what it accomplishes and what it doesn't do, and why open source developers should embrace it.

DIY Patent Reform

2008-06-26T08:38:00.002-04:00

As I noted in my last posting, there is little likelihood of legislative action on patent reform, certainly not comprehensive action, this year. While disappointing, especially to those in the software industry and, perhaps, those in the industries most impacted by so-called business method patents, that doesn't mean there is nothing to be done. In fact many in those industries can impact their own future through participation in the Peer-to-Patent pilot project run by New York Law School in cooperation with the U.S. Patent and Trademark Office.

Peer-to-Patent is a test program to determine whether individual citizens, expert in fields of technology, are willing to contribute their time and effort to identify prior art (existing patents, trade publications, academic papers, or other bibliographic information that either disclose the invention or some important aspect of the invention or suggest the invention) that may then be used by a patent examiner in the examination process. By assisting patent examiners in finding such prior art, these "peer reviewers" contribute to a narrowing of claims actually allowed or patents issued, i.e., reducing the number of poor quality patents that issue.

Poor quality patents - those that should never have been issued or should not have been issued with the broad claims they contain - result in unnecessary and costly litigation and an inefficient marketplace for ideas. While true innovation can and should be rewarded, issued patents that lack merit inhibit innovation by others.

Although Peer-to-Patent only addresses patents individually, the project's first annual report indicates it has improved the quantity and quality of information available to patent examiners, and the patent examiners clearly appreciate the help. A wide range of companies have subjected their patents to such "enhanced" examination, including IBM, Microsoft, General Electric, Hewlett Packard, Intel, and Red Hat. More than 2000 potential reviewers have signed up to keep abreast of the project, and of these almost 400 have been actively engaged in the review process.

There is no cost to patent applicants to include their applications in the program (see requirements). While some have suggested that such participation only contributes to a slowing of the application process, under the pilot program the reverse is true. Applicants who submit to the peer review process see their applications bumped to the head of the examination line, potentially shortening the time to issue by as much as a year.

There is also no obligation placed on peer reviewers. They can select which, if any, applications they wish to work on.

An extended and expanded Peer-to-Patent pilot is expected to be announced any day now. So while you wait for legislative action, how about chipping in to contribute to patent reform right now!

Where Now With Patent Reform

2008-06-20T09:28:00.002-04:00

Given the failure of pending patent reform legislation (S. 1145) to make it out of committee in the U.S. Senate this spring, it is unlikely that we will see any form of comprehensive patent reform legislation this year. This should not be surprising given that this is a major election year and the fact that there are a few other issues on the table (Iraq, Afghanistan, Iran, Middle East peace, soaring gasoline prices, soaring food prices, housing market in the ditch, Midwest flooding - shall I go on?). So what might we expect.

The major stumbling block for the comprehensive legislation seems to have been the issue of apportionment of damages. Technology companies have complained for some time that damages should not be assessed against the entire price of a device or piece of software where the patent covers only one small component, unless that component can be shown to have been the principal driver for the device or software. Other industries, including the pharmaceutical industry, are concerned that such a change in standards would drastically reduce damage awards against directly competing, infringing products. Don't expect any easy resolution of that issue.

However, there are a number of changes upon which all of the different factions seemed to be okay, and that fact may open the possibility to more limited, but non-controversial, reform legislation. Exactly what the scope of that legislation may be remains to be seem. Certainly the move to a first-to-file system would be included since such a change would conform U.S. practice to the rest of the world.

In the meantime, the USPTO continues to work on changes in their practices that can be implemented through rules changes, as opposed to legislation and that will potentially improve the quality of patents being issued. Among these initiatives are first-action interview, complex work unit, and peer review prior art (the Peer-to-Patent pilot run by New York Law School).

I'll talk more extensively about Peer-to-Patent in a follow-up.

No Longer Quiet - Red Hat and the Patent Settlement

2008-06-11T19:01:00.006-04:00

Sorry to have been gone so long. I really didn't expect to be quite so quiet. But between teaching at Duke Law and NCCU Law, a little bit of consulting work, and generally trying to enjoy retirement, well, time slipped away. But now I'm back, and I want to weigh in on Red Hat's settlement of the patent infringement lawsuits filed by FireStar and its successor DataTern. Under a confidentiality agreement with Red Hat I was given a chance to review the agreements. By the way, they won't remain confidential for all that long.

Although I had to repress my initial gag reflex at every settling these cases (the amount of prior art identified against the original patent asserted by FireStar was/is almost mind boggling), the settlement is a rational response to such claims. Red Hat disposed of the claims in a fiscally responsible manner given the cost of patent litigation. However, that is the far less interesting aspect of the settlement. The truly admirable part of the settlement were the terms that Red Hat and its legal team extracted from DataTern and its financial backers. Not only did Red Hat obtain license terms that protect its products, including Hibernate, it did so in a manner that I believe is entirely consistent with both versions of the GPL. That is no small feat. And we are not talking about a Microsoft/Novell style license. On top of that, Red Hat didn't stop with the asserted patents; they made sure that DataTern and its portfolio of patents aren't going to be a problem for Red Hat and its licensees for a long time to come.

Well done.

Half A Loaf

2007-10-22T15:53:00.000-04:00

“Compromise used to mean that half a loaf was better than no bread. Among modern statesmen it really seems to mean that half a loaf is better than a whole loaf.” - G.K. Chesterton

With the European Commission's announcement earlier today that it had reached a settlement with Microsoft in the long-running competition battle, many on both sides of the Atlantic are probably breathing a sigh of relief. Perhaps it is time to move on. While the public does not know exactly to what the Commission and Microsoft have agreed, there are hints in the Commission's statement.

Clearly, the Commission believes it has obtained free and unfettered access to the Microsoft work group server protocols that will benefit open source developers. Had the Commission used the terms open source or open source developers one more time in their press release, one would have come to the conclusion that open source was the sole focus of the case. A careful parsing of the Commission statement makes clear, however, that the Commission may have obtained access to the protocol documentation from a copyright and/or trade secret standpoint, but the same cannot be said on the patent front. Noticeably missing from the Commission statement is any affirmation that open source software developed to implement the protocols will be free from patent concerns. If anything, the Commission is clear that it has not solved the patent equation for those deploying open source solutions implementing the protocols when it states:

"Initially, Microsoft had demanded a royalty rate of 5.95 % of revenues for a combination of access to the secret interoperability information and for a patent licence and had refused to make the licence compatible with the open source business model.

In a statement of objections of 1 March 2007 the Commission warned Microsoft of penalty payments over its unreasonable pricing (IP/07/269). The Commission also stated that it would ensure that open source developers could have access to the non-innovative parts of the interoperability information (IP/05/673).

In addition to the two licences Microsoft will publish an irrevocable pledge not to assert any patents it may have over the interoperability information against non-commercial open source software development projects.

These measures will ensure that Microsoft’s competitors in the work group server market, including those following the open source business model, will have access to the interoperability information on reasonable terms which will in turn lead to more competition and innovation in this market."

Yes, open source will have access to the interoperability information on "reasonable" terms (ask independent open source developers how many can afford to cough up the $15,000 such access will cost), but nothing in this statement indicates the Commission has overcome Microsoft's "refusa[al] to make the [patent] licence compatible with the open source business model." In fact, we can expect that nothing about that patent license will be compatible with the most widely used open source license, the GNU General Public License.

So what has been accomplished. The Commission appears to have successful forced Microsoft to open its work group server protocols to viewable access by all, including open source developers. It also appears to have assured that such developers will be able to implement the protocols, at least from a copyright and/or trade secret perspective. At the same time the Commission appears to have been unable to change Microsoft's stance on the patent issue other than drastically reducing the royalty that Microsoft is to be paid. [N.B. - Having started out demanding 5.95% as a comprehensive royalty and finally agreeing to a low 0.4% comprehensive royalty, all of a sudden those "highly valuable" Microsoft patents don't look so valuable any more.]

So we are left with half a loaf, and perhaps that is the best the Commission could do under the law and under the circumstances. I, for one, just wish the Commission had gone a bit easier on the open source rhetoric. It is misleading from the standpoint that the settlement does not resolve all or even the biggest issues that open source will have with implementing these protocols. Why give Microsoft the public relations benefit of insisting it is open source friendly when it clearly isn't? I can't wait to see the limitations of the patent pledge alluded to in the press release and how Microsoft plays that one up.

We will know more when all of the agreements are out in the open. For now, I give the Commission an A for effort but a B- for follow through. Harsh grading? What the heck, I'm a law professor now.

Dear Steve

2007-10-11T14:17:00.000-04:00

Dear Steve,

I know you are quite concerned that Red Hat is not showing due respect for your (unidentified) patents, but it strikes me as a little rich that you or Microsoft should be lecturing anyone on the proper respect for the patents of others given all of the following parties that believe that Microsoft does not respect their patents:

Performance Pricing Sues Microsoft for Patent Infringement - October 2007
Premier International Associates - September 2007
Lo Sues Microsoft for Patent Infringement - August 2007
JuxtaComm Technologies Sues Microsoft for Patent Infringement - August 2007
Aloft Media Sues Microsoft for Patent Infringement - July 2007
Paone Sues Microsoft for Patent Infringement - July 2007
Achates Reference Publishing Sues Microsoft for Patent Infringement - June 2007
Veritical Computer Systems Sues Microsoft for Patent Infringement - April 2007
James Taylor Sues Microsoft for Patent Infringement - April 2007
Multimedia Patent Trust Sues Microsoft for Patent Infringement - April 2007
F&G Research Sues Microsoft for Patent Infringement - April 2007
i4i Limited Partnership Sues Microsoft for Patent Infringement - March 2007
VirnetX Sues Microsoft for Patent Infringement - February 2007
Fenner Investments Sues Microsoft for Patent Infringement - January 2007
Gobeli Research Sues Microsoft for Patent Infringement - December 2006
Alcatel Sues Microsoft for Patent Infringement - November 2006
PalTalk Holdings Sues Microsoft for Patent Infringement - September 2006
Computer Acceleration Corp. Sues Microsoft for Patent Infringement - July 2006
International Automated Systems, Inc. Sues Microsoft for Patent Infringement - July 2006
Anascape Sues Microsoft for Patent Infringement - July 2006
z4 Technologies Sues Microsoft for Patent Infringement - June 2006
Veritas Operating Corp. Sues Microsoft for Patent Infringement - May 2006
Triton IP Sues Microsoft for Patent Infringement - April 2006
Softvault Systems Sues Microsoft for Patent Infringement - January 2006
Mallory Ventures Sues Microsoft for Patent Infringement - January 2006
Sklar Sues Microsoft for Patent Infringement - January 2006
Visto Sues Microsoft for Patent Infringement - December 2005
Data Encryption Corp. Sues Microsoft for Patent Infringement - July 2005
Carlos Armando Amato Sues Microsoft for Patent Infringement - May 2005
Forgent Networks Sues Microsoft for Patent Infringement - April 2005
Compression Labs Sues Microsoft for Patent Infringement - April 2005
American Video Graphics Sues Microsoft for Patent Infingement - March 2005
Burst Sues Microsoft for Patent Infringement - March 2005
Spreadsheet Automation Corp. Sues Microsoft for Patent Infringement - March 2005
BTG Sues Microsoft for Patent Infringement - July 2004
Lextron Systems Sues Microsoft for Patent Infringement - February 2004
Motionless Keyboard Co. Sues Microsoft for Patent Infringement - Feburary 2004
Immersion Sues Microsoft for Patent Infringement - 2003
TC Interactive Data Sues Microsoft for Patent Infringement - May 2002
Network Commerce, Inc. Sues Microsoft for Patent Infringement - December 2001
Priceline.com Sues Microsoft for Patent Infringement - October 1999
Eolas Sues Microsoft for Patent Infringement - February 1999
Goldtouch Technologies Sues Microsoft for Patent Infringement - December 1998

By now you should be getting the picture. That isn't to say all of these complaints have or had merit, and I'm sure some have been resolved. It's just that someone who wants to talk so much about respecting intellectual property should do a little respecting himself, don't you think?

Logic From Another Planet

2007-09-19T06:55:00.000-04:00

Ever have one of those experiences where you wonder whether you are from the same planet as another person who has just uttered the unbelievable? That was my reaction when I read Assistant U.S. Attorney General Thomas O. Barnett's comment on Monday's decision by the European Court of First Instance upholding the European Commission's finding that Microsoft had acted in an anti-competitive manner. To quote Mr. Barnett in pertinent part:

"We are, however, concerned that the standard applied to unilateral conduct by the CFI, rather than helping consumers, may have the unfortunate consequence of harming consumers by chilling innovation and discouraging competition. In the United States, the antitrust laws are enforced to protect consumers by protecting competition, not competitors. In the absence of demonstrable consumer harm, all companies, including dominant firms, are encouraged to compete vigorously. U.S. courts recognize the potential benefits to consumers when a company, including a dominant company, makes unilateral business decisions, for example to add features to its popular products or license its intellectual property to rivals, or to refuse to do so."

Oh, really. Let's remember that this is the same Andrew Barnett who, to quote the NY Times:

"had until 2004 been a top antitrust partner at the law firm that has represented Microsoft in several antitrust disputes. At the firm, Justice Department officials said, he never worked on Microsoft matters. Still, for more than a year after arriving at the department, he removed himself from the case because of conflict of interest issues. Ethics lawyers ultimately cleared his involvement."

Mr. Barnett, according to the same Justice Department that claims torture is not torture, the Geneva Convention is antiquated, and firing of U.S. Attorneys for purely political reasons does not constitute a violation of law, is an unbiased source when it comes to Microsoft. Okay, I'm getting a bit catty, but you have to admit that the credibility of the Justice Department in the U.S. is not exactly at its apex today. Any way, enough of that.

Why did Barnett's comments strike a nerve with me? I suppose the first thing that comes to mind is Mr. Barnett's lack of presence in that court room in Luxembourg in April 2006. I was there, and to the best of my recollection, he wasn't. I heard the legal counsel for the European Commission and its intervenors present overwhelming, plausible evidence that Microsoft's actions in bundling products and denying interoperability information to others in the industry had little to do with Microsoft's desire to innovate. No, according to that evidence at each step of the way Microsoft's actions seemed designed to destroy or diminish the demand for a competing and often times innovative technology. True for Netscape. True for Real Networks. True for SAMBA.

In the end, the decision by the CFI does not pose a wide ranging threat to U.S. or other companies unless such companies are market dominant AND by their demonstrable actions have moved to destroy competition. To follow Mr. Barnett's logic competition doesn't come from competitors, it comes from consumers, and the only companies that innovate are super dominant companies, so let's leave them alone. If they are super dominant, they must be innovative. Didn't Microsoft invent the word processor, the spreadsheet, presentation graphics, the web browser, the media player, e-mail, identity management, and the world wide web?

No, they didn't.

Anyway, enough about Mr. Barnett's silliness. I am looking forward to the day he heads back to private practice.

In the meantime, we have a decision from the CFI that speaks to the facts in the Microsoft case and says, when those facts are examined in the light of the legitimate interests of the citizens of the European Union, they demonstrate a violation of law. Yes, it is precedent. No, it will not have the dire consequences that Mr. Barnett fantasizes because, you see, the decision seeks to restore competition unfairly lost.

A Major Step for Competition

2007-09-17T06:50:00.000-04:00

As you may have already seen, the European Court of First Instance has upheld the decision of the European Commission in the Microsoft matter. This is a major victory for competition and consumers. Having been present during the oral arguments in the case, I thought the Commission did an admirable job in presenting its case. But as one might expect, Microsoft also did an adequate job defending itself. As the decision was delayed for almost 18 months following the hearing, I became less certain of the likelihood of the outcome. Now that it is in, it is clear that the Court based its findings on established law and did not break new ground. This fact, in itself, stengthens the likelihood that this decision will ultimately be appeal proof.

Neelie Kroes, the European Commissioner responsible for competition was unquestionably pleased with the outcome, and rightfully so. The Commission has been under a constant barrage of negative publicity from Microsoft surrogates for several years now. Those surrogates all but proclaimed the downfall of the modern world if Microsoft was not sustained and complained bitterly that the Commission was trying to seize Microsoft's intellectual property.

It is interesting that Brad Smith, in his press conference, spoke about Microsoft's new relationships with both Sun and Novell as examples of how they are cooperating with other companies on interoperability issues. Perhaps true enough, but one should not lose sight of the fact that neither of those companies is as strong today as they were in 1998 or 2000. An image of crumbs from the table comes to mind.

Now we will wait to see if Microsoft will comply with the decision or whether they will choose to drag matters out for another 2+ years by appealing. While in my wildest dreams I did not imagine such complete success on the part of the Commission in this decision, I will now dream about the future in information technology that could be. In that dream Microsoft finally realizes that there is a huge benefit to true transparency and that it can share much more information with its competitors on much more reasonable terms and, as a result, become even more successful that it has been to date. In other words, let's see if Microsoft is ready to walk the walk instead of just talking the talk. This does not have to be a win/lose situation for all involved.

Another Fast One From SCO?

2007-09-15T09:43:00.001-04:00

My apologies for being missing in action for the last few weeks. Still trying to figure out this "retirement" thing. But the SCO Group's filing for bankruptcy protection yesterday was enough to shake me out of my revery.

I am not a bankruptcy lawyer, and I won't pretend to know all of the intricacies of Chapter 11. That said, I do know how to read a financial statement (I was a CFO before I ever went to law school). And that is the part of the SCO filing that has intrigued me the most.

Thanks to PJ and her supporters at Groklaw, we have access to the petition and supporting documents SCO has filed. As PJ points out, the one thing SCO doesn't appear to have filed with the court, which they are required to file, is a statement of assets and liabilities. So let's look at those.

In their most recent 10Q filed with the SEC for the period ended April 30, SCO reported total cash and receivables on hand of $17.5 million. Of that, they listed $5.4 million as "restricted cash. Let's focus on that number for a second. In SCO's most recent 10K filing (for the period ended October 31, 2006) they explain restricted cash as including the amount that is owed Novell under the 1995 Asset Purchase Agreement for licensing of the SVRx technology. That number has pretty consistently been just under $3 million from one period to the next. Here is the statement from the 10Q regarding the restricted cash:

"Pursuant to the 1995 Asset Purchase Agreement and the Company’s acquisition of assets and operations of The Santa Cruz Operation, the Company acts as an administrative agent in the collection of royalty payments from a limited number of pre-existing Novell customers who continue to deploy SVRx technology. Under the agency agreement, the Company collects payments from such customers and receives 5% as an administrative fee and remits the remaining 95% to Novell on a routine basis. The Company records the 5% administrative fee as revenue in its consolidated statements of operations. The accompanying consolidated balance sheets as of October 31, 2006 and 2005 reflect amounts collected related to this agency agreement as of each balance sheet date, but not yet remitted to Novell of $2,978,000 and $2,815,000, respectively, as restricted cash and payable to Novell."

What they don't explain is the difference between that $3 million and the $5.4 million. In other words, why do they list another $2.4 million as restricted.

Flipping back to the April 30 10Q, we see that SCO listed total liabilities of $12.65 million. In other words, but for the restricted cash, SCO had enough cash on hand to cover its obligations ($17.5 million, including restricted cash, to cover $12.65 million in obligations). So without the restricted cash, it would appear they would have been justified, back in April, to have filed for Chapter 11 protection. But wait a second, they also list that $3 million owed Novell as a liability. No fair double counting! If we also assume the other $2.4 million in restricted cash also pertains to one or more liability items, then the restricted cash should be included in determining solvency.

Isn't it possible that SCO was solvent in April but not now? Possible? Perhaps. Probable? I don't think so. If you look at the cash flow statement included in the April 10Q you will see that SCO, in fact, burned very little cash on operations during the first 6 months of their fiscal year, the period ending in April. In fact, SCO's cash actually increased by almost $2.4 million during that period! SCO would now have us believe that they burned through almost $5 million in cash since April 30 in becoming insolvent. ON WHAT?! Of course, given that we haven't had a chance to see their most recent quarterly report (that was due on Friday when they filed the bankruptcy petition) or a statement of assets and liabilities, it will be interesting to see how SCO explains this dramatic reversal of fortune.

Another tidbit. In its Chapter 11 petition SCO lists total assets of $14.8 million and total liabilities of $7.5 million. That is a reduction in total assets of $4.1 million and in total liabilities of $5.15 million since April 30. In other words, if those numbers are to be believe, SCO actually reduced its total liabilities by an amount greater than it reduced its total assets since April 30. That means shareholder equity will actually have gone up, not down, in the ensuing four and a half months. Again, that hardly paints a picture of a company that is insolvent.

Why does any of this matter? Well, filing for Chapter 11 does not automatically qualify the petitioner for Chapter 11. The bankruptcy court must still consider the petition and the actual facts. I'm suggesting that SCO may have a bit of trouble actually proving they are insolvent and can't meet their obligations. Would that still be the case a week from now if the trial with Novell were not stayed? There was certainly the possibility the Utah court would find that the licensing revenues SCO received from Microsoft and Sun were subject to the SVRx licensing arrangement and that the bulk of the license fees from those transactions were due Novell (perhaps as much as $25 million). If that were to happen, it would certainly have been within the realm of possibility that Novell could put SCO in Chapter 7 bankruptcy (liquidation). So by this fast (and loose?) move SCO has:

Stayed the Novell trial;
Stayed the IBM trial; and
Avoided possible (and perhaps likely) involuntary liquidation.

One has to ask, though, whether this is just one more vastly clever (or is it only half-vastly clever) move on SCO's part that in the end will make no difference. Staying a trial and dismissing an action are two very different things. In the end, the SCO Group will have to account of its actions. It's just that the accounting will now come a little later, assuming the bankruptcy court grants their petition. In the end I believe the result will be the same: no-SCO.

Why I Won't Lose Sleep Over Jacobsen v. Katzer

2007-08-27T14:58:00.000-04:00

Much ado has been made about the recent decision in the case of Jacobsen v. Katzer (N.D. Cal. 8/17/2007). Concern has been expressed that this decision could prove problematic in the enforcement of other open source licenses. For the record let me say that I am not going to lose sleep over this case.

There are a number of reasons I feel that way. First, the license involved is the Artistic License. This is not a widely used license. Out of some 2500 packages in Red Hat Enterprise Linux, only 9 are licensed under the Artistic license, and almost all of those relate to PERL. The Artistic License is one of those open source licenses we could have done without. It is not particularly well written. It contains a number of ambiguities ("place your modifications in the Public Domain or otherwise make them Freely Available, such as by posting said modifications to Usenet or an equivalent medium, or placing the modifications on a major archive site such as ftp.uu.net”; what has the method of posting got to do with the definition of Freely Available?). All in all, the Artistic License is not the gold standard of open source licenses.

Second, spend any time around open source licenses and you are likely to end up in a debate over whether they are permissions or contracts. A somewhat general rule in copyright is that a license agreement that is to be construed as a contract need have at least one added element beyond those elements (rights) granted under copyright. In other words, if the agreement only addresses copying, modifying and distribution, any so-called contract provisions are going to be preempted by copyright law. Examples of an added element would be the payment of a royalty for the rights granted; failure to pay the royalty would trigger a right in contract to recover the royalties. When you look at open source licenses, you will rarely find that added element, and that is certainly true of the Artistic License. In this case, however, the Court managed to somehow wangle an extra element out of the agreement. What that element is happens to escape me, but the Court seems to say it is the obligation not to remove copyright attribution notices and disclaimers. The Court appears to be saying, since the licensor granted such broad rights to copy, modify and distribute, there are no instances under which a licensee could exceed the scope of the copyright license. Hey, Court, wake up. If a licensee removes copyright notices and holds the work out as their own, and distributes it, I'm supposed to believe that's okay because the license provides them a broad right of redistribution. Are we supposed to gather from this holding that copyright notices and attribution requirements are always outside the bounds of copyright and will always sound in contract?

The truth be told, this whole area of copyright law (i.e., contract vs. license) is a bit murky, and there are conflicting interpretations by varying courts. Even the learned copyright scholar David Nimmer points to these inconsistencies in his treatise Nimmer on Copyright. So, if I believe the Court got this wrong, why am I not worried about it? Because it was such a minor aspect of the case, and quite frankly, it didn't appear that either party cared that much any more. The defendant had voluntarily ceased distribution, and there was no substantial on-going damage being done to the plaintiff.

Once again we have a case that nibbles at the fringes of open source, but I would be hard pressed to say it took a big bite.

OSi, Microsoft, and OSI-Certified Licenses

2007-08-21T15:25:00.000-04:00

The thermostat has been shoved up to about 84 degrees (that's Farenheit for those in the U.S., 29 degrees Celsius for most of the rest of the world) on the subject of whether the Open Source Initiative should certify the two licenses submitted by Microsoft, the Microsoft Permissive License and the Microsoft Community License.

Chris DiBona of Google has suggested that there be additional criteria imposed upon Microsoft by OSI before any license certification. Among Chris' suggestions:

Microsoft should stop referring to the licenses as "shared source"
Microsoft should stop spreading misinformation about the nature of open source software
Microsoft should be forbidden from making patent threats against open source software
Microsoft should stop using its desktop monopoly to interfere with the bundling of open source software on new computers.

While I can appreciate Chris' frustration with Microsoft, I disagree with this approach and for much the same reason as I disagreed with the anti-Microsoft, anti-Novell language added to version 3 of the GNU General Public License. This sort of one-off rule making sets a bad precedence when we are trying to establish free and open source licenses that have enduring value. Everyone knows that there are individuals at Microsoft who have been complete horse's a__es about open source for years. We don't need to educate folks on this fact, and we don't need to try and legislate them into open source advocates.

That said, I do believe Chris and others are right on point in questioning the need for two new licenses that may add little over existing OSI-certified licenses, if in fact that is the case. Let's take Microsoft at its word that open source license proliferation is not a good thing. Microsoft insists the Permissive License differs from the New BSD license because it contains a specific patent grant. Intel has previously taken a step in that direction by issuing a BSD+patent license. While Intel did not seek formal OSI certification for its combination license, such a license combination may be useful. And despite how one may feel about Microsoft and its various shenanigans, there is value in having some OSS licenses that are compatible with Free Software, Open Source Software, and closed-source software. The case for the Community License may be harder, but let's give the OSI board room to make a reasoned determination with the benefit of community feedback. Remember Criteria No. 5 of the the Open Source Definition.

As for Bill Hilf's digs at Google, let's be clear. Google, to my knowledge, has never violated the GPL in its practices. Google is not required to give anything back to the community because the GPL imposes no such requirement. In fact, the GPL expressly permits one to make modifications to GPL code and keep them solely for your own benefit. While some may feel such actions are not in the spirit of free software, I beg to differ. I cannot expect freedom if I am not willing to extend freedom, and that includes the freedom to not distribute modifications. Besides, as Chris has pointed out, Google has contributed more than a million lines of code to open source projects.

Connecting Dots

2007-08-17T07:19:00.000-04:00

Sometimes events transpire in the software industry that, when tied with other events, take on a much bigger meaning. Such is the case with three different announcements in the last ten days. The first two got a good bit of press; the last didn't. Let's see how they connect.

First, IBM and Novell announced a new partnership on the desktop and in relation to IBM's Websphere Community Edition. Aimed directly at Red Hat's JBoss Application Server, this move is fascinating in that it represents the prodigal son returning to the IBM fold, apparently with complete foregiveness for entering that deal with Microsoft. There is no question that this move by IBM will challenge Red Hat/JBoss. And as for the great offense that IBM took with Novell for cozying up to Microsoft, all you have to do is look at the next event.

The Massachusetts Technology Leadership Council sent out their invite for upcoming events this week. One of those events is entitled: Microsoft & Novell - Building Bridges. On its face you would expect this to be one of those Microsoft - Novell events trying to justify and promote their relationship. However, this invitation was more interesting given the session sponsors: IBM and the Choate law firm. (Sorry, that little tidbit is not available on the MTLC website; it was included in the e-mail invitation.) So here we have IBM sponsoring a session that attempts to rationalize and buy into the theory that the Microsoft - Novell deal is actually promoting interoperability. Of course, that doesn't mean that IBM supports that theory; such a shift would certainly be dramatic given IBM's support of the OpenDocumentFormat movement and its participation in the competition proceeding against Microsoft in Europe. Nevertheless, it catches one's eye.

I also mention that the Choate law firm is a sponsor of this MTLC event. Why is that significant? Because Karen Copenhaver, Choate partner, is legal counsel to the Linux Foundation. Coincidence?

The third event was the IBM - Sun announcement around shipping Solaris on IBM hardware (are you starting to see a common theme here?). This is a smart move for both companies. Sun is looking to shift from being a hardware/software company to being a software company alone, and key to that shift is burying the hatchet with IBM. For IBM it provides another Unix-based operating system vendor with which to challenge Microsoft and with which to keep pressure on the other Unix-based operating system vendors, Red Hat and Novell. Bill Zeitler of IBM called the move a "sea change." Jonathan Schwartz of Sun called it a "tectonic shift." Zeitler went on to talk about an ever expanding relationship. Don't be surprised if you see future announcements between these two regarding both JAVA and OpenOffice.

So there you have it. IBM put on the full court press in the last ten days. It is sending a message to both Microsoft and to the Linux/open source players that IBM will control the other side of the court.

It's The Community's Fault

2007-08-15T13:50:00.000-04:00

What is Canonical thinking? Have your servers go down, and then you tell the hand that feeds you that it's their fault?! It's bad enough that Canonical packages a bunch of proprietary code in with Ubuntu and yet proclaims it free, but to then blame the community because of its own inattention is ripe. Come on Canonical, you are better than that.

Novell Pipes Up

2007-08-15T11:00:00.000-04:00

Novell has piped up making it clear they will not be looking to enforce the Unix copyrights against Linux. Settles that issue. A smart move by Novell.

Some have speculated that it would be worthwhile to now take SCO off the market. Heck, their market cap is now under $10 million. The problem is that paying $10 million to buy SCO would not be the end of it. SCO is still embroiled in the IBM case and the Red Hat case, to say nothing of the on-going claims that Novell has. In addition, when the lights finally flicker out on SCO, look for some shareholder lawsuits based on violation of securities laws. If you go back to the press conferences that SCO repeatedly called back in 2003 and 2004, they never began those press conferences by making the standard disclaimers cautioning investors to take what they were saying with a grain of salt. As a consequence, investors had every right to take what Darl McBride and Chris Sontag were saying in public back then as the gospel truth. Like McBride stating publicly that SCO owned the copyrights to Unix in the spring of 2003 while he was privately corresponding with Novell begging them to transfer the copyrights to SCO.

We have come a long way from that day in 2003 when McBride suggested IBM buy SCO for $500 million.

Novell, SCO and Where Things Stand

2007-08-14T15:13:00.000-04:00

The public responses to the order issued by Judge Kimball in the Novell v. SCO case in U.S. District Court in Utah last week have been fascinating to watch. Clearly, the Judge's rulings did SCO a great deal of damage. Gone are the slander of title claims and the ability to assert any of the copyrights in Unix System V code against anyone. Moreover, SCO now faces a claim from Novell with respect to the $26+ million in license fees it received for Unix code from Sun and Microsoft. The judge has already ruled that some amount is due; the only issue left is how much. If the amount approaches the 95% cut called for in the Asset Purchase Agreement between Novell and SCO, SCO, which only has $13 million in the bank of which $5 million is subject to restrictions, is effectively bankrupt.

Some have suggested that this ruling should raise new concerns among Linux users about Novell's intent with respect to the Unix copyrights. Is it theoretically possible that Novell could become the new SCO? Not likely. First, despite almost four years of looking, SCO was never able to identify a single bit of Unix System V code infringed by Linux. Are we to expect Novell will find what SCO has so diligently looked for? Second, Novell has the same problem SCO did, only in spades. Novell has been distributing Linux code for years now and continues to do so. Under any of the reciprocal licenses covering components of Linux distributions, Novell would be hard pressed to now bring a copyright infringement claim against another party for distributing those same components. Such a claim would almost certainly breach the open source licenses upon which Novell has depended, but it would also likely be estopped by Novell's own behavior in encouraging such redistributions.

While it would be nice to see all of this ultimately adjudicated, that is not likely to happen principally because SCO will run out of money first. In the meantime, perhaps Novell could make nice with the FOSS community by burying these copyright claims once and for all.