What if SCO wins?

COMMENTARY--I was all set to do another article on .NET, but themedia turbulence surrounding SCO's recent raising ofthe stakes in its apparent battle with the entireLinux world scotched that idea (it'll appear at alater date, an event I'm sure all of you will awaitwith bated breath).

The SCO suit isn't the firstlegal challenge of an open source product by supposedowners of intellectual property used therein, but itis arguably the most important one. Linux is rapidlybecoming the center of the Unix programming domain,and the number of companies adopting Linux and/ordeveloping products for the platform grows by the day. Billions of dollars are at stake, as an entireindustry has arisen around an operating system thatseemingly came from nowhere to challenge the businessmodels of proprietary companies, Unix-based orotherwise.

The merits, or lack thereof, of this case have beenanalyzed to death by a number of people. So, let'signore all the legal splitting of hairs and peer a fewyears into the future. We all know what would happenif SCO loses. What would happen, however, if SCO won?

Can an IP encumbered Linux be distributed?
This is the first issue which would need to beresolved. The GPL explicitly forbids distribution of GPLed code if it is found to be IP encumbered.

If, as a consequence of a court judgment orallegation of patent infringement or for any otherreason (not limited to patent issues), conditions areimposed on you (whether by court order, agreement orotherwise) that contradict the conditions of thisLicense, they do not excuse you from the conditions ofthis License. If you cannot distribute so as tosatisfy simultaneously your obligations under thisLicense and any other pertinent obligations, then as a consequence you may not distribute the Program at all. For example, if a patent license would not permit royalty-free redistribution of the Program by all those who receive copies directly or indirectly through you, then the only way you could satisfy both it and this License would be to refrain entirely from distribution of the Program.

Note that it doesn't say that existing users have tostop using IP-encumbered GPLed code, just that no onewould be legally allowed to distribute the GPLed codeto new users.

		News Focus Microsoft-SCO battle Linux Intellectual property

There are two possible solutions to this problem. Theopen source community can pull out the portions foundto be owned by SCO. This may or may not take verylong, dependent on what IP SCO is found to own (as SCOis keeping the details close to the vest, no one canreally estimate what this would involve). Alternatively, the open source community can choose toallow the code to be distributed, anyway, withend-users expected to pay a licensing fee for theportions found to be owned by SCO. This is really upto the people considered to hold the copyright onLinux. Who, exactly, would legally be allowed tochange the licensing terms of Linux? Linus Torvalds? (that IS a question, and I hope someone has an answerin the Talkbacks)

Assuming there is a solution to the IP issue, and thatsolution doesn't take too much time to implement, whateffects would a successful outcome for SCO have on the industry?

1. A temporary slowdown in the growth of Linux
Those most leery of rolling out Linux would be largecompanies that make more interesting targets forowners of intellectual property contained in Linux. Smaller companies would be less leery, as they wouldfall beneath the radar screen, and chasing them downwould be like trying to catch a swarm of gnats. Those1500 threatening letters SCO sent out went to largeLinux licensees, which excludes hundreds of thousandsof companies and end users around the world that SCO,or anyone else for that matter, would be unable tocatch.

As noted, this slowdown would be temporary, even ifthe end result is that Linux licensees must pay alicensing fee for the right to use the product. Linuxwould still be a completely open source product withall the flexibility that entails. It would still costless than most proprietary operating systems. Itwould still have a large pool of knowledgeabletechnology personnel from which to draw (what computerscience major exits university without exposure toLinux?).

A possible snag would arise if SCO's success causesmore companies to pop out of the woodwork with claimson IP contained in Linux. As many have noted, the SCOsuit is not over patents, but supposed "trade secrets"to which IBM was granted access after signing anon-disclosure agreement. This doesn't resolve thequestion, in other words, of patents on logic used,consciously or not, in Linux. Even people at IBM haveadmitted this possibility, which explains why IBMisn't in the business of shipping an IBM-branded Linux distribution.

Were this to happen, SCO might be only the first actin a longer battle over the future of Linux.

2. No change in the rollout schedules of Linux products:
Linux has long since passed the market share threshold which makes it interesting as a target of software development. That will remain the case, even if there is a bit of confusion over intellectual property in Linux.

Of course, that assumes my worst case scenariooutlined in item 1 doesn't come to pass. Mostcompanies, however, will assume Linux's IP-relatedissues to be short term, and will act accordingly.

3. Don't expect a boost in sales of proprietary UNIX:
Companies aren't going to decide to ditch a Linux rollout in favor of a proprietary Unix unless there is evidence that Linux IP difficulties will be long-lasting. Linux is still the future of the Unixprogramming domain, and will remain so as long asthe full source code remains available to everyone.

As for companies suddenly deciding to pull up theirtent stakes and move to the Microsoft side of thetracks, I doubt it. Making a shift from UNIX toWindows would be a large investment in terms ofsoftware and personnel, and most companies won't do itin the current weak economic environment. UNIX is aunique programming domain with its own set of skills,software and programming conventions, and thosecompanies currently operating in that domain are liketo remain inside of it.

4. Expect more IP litigation:
IP litigation has ballooned over the past decade asmore and more companies endeavor to patent everythingthey can get their hands on. This is partly due tothe need to make money in a market threatened bycompetition from free products, and partly a result ofan industry taught that courts can be a usefulcompetitive tool as well as a source of revenue. Asuccessful SCO will only compound this problem,providing yet another reason to sink more money intolegal instruments of competition and revenuegeneration.

This is bad news for small companies, which don't havethe resources to defend themselves against IPlitigation, and more important, don't have the moneyto generate the IP larger companies use as a defensiveshield (for more on that, see my article on patent trading). It'salso bad news for software developers, who will findit much harder to write code that is not IPencumbered. This is good news to lawyers, though. Onthat note, current trends in America seem to benothing but good news for lawyers.

5. Expect more proprietary companies to releasemore source code.
This is wildly speculative, but if it comes to pass,would be the most unusual outcome of a SCO win. AT&T's experience with Unix taught a generation oftechnology executives the pitfalls of releasing thesource code to a product that a company has any hopeof commercializing. AT&T essentially lost control ofUnix (a fact which may affect the outcome of the SCOcase), and companies who wish to avoid that fate tendto keep their source code close to the vest.

If SCO manages to prove, however, that access tosource code does not require losing control ofintellectual property contained in that source, thenmore might be willing to consider releasingmore source code. The code, of course, would have tobe covered by some form of licensing arrangement, butthe history of Unix shows that an awful lot of peoplecan see code that is distributed in this fashion. This might appeal to those whose interest in opensource is less ideological so much as utilitarian.

Microsoft talked about the "viral" nature of opensource software. Oddly enough, if SCO succeeds, thatviral nature could be put to use by proprietarysoftware companies.

Conclusion:
Open source companies are growing up and becoming the profit-oriented entities that companies at their core truly are. Unfortunately, as the SCO case has shown, this can be like watching Dr. Jeckyll turn into Mr. Hyde.

Corporations have one simple motivation, and that isto make money for their stockholders. This means theywill not necessarily adhere to the core beliefs andprinciples that motivate the open source community tocreate software. This conflict can only become moreand more frequent, as more companies attempting theopen source route run across rough patches typical ofany company's hoped-for evolution towardsprofitability.

biography
John Carroll is a software engineer living in Ireland. He specializes in the design and development of distributed systems using Java and .Net. He is also the founder of Turtleneck Software.