Slashdot Mirror

njcoder writes "C|net's news.com.com has reported that Sun Microsystems is releasing parts of the OpenSolaris code today licensed under the OSI-approved CDDL . The release consistes of over 5 million lines of code for the base system OS/Net (kernel and networking). OpenSolaris is based on Solaris 10, the current version of Sun's Unix Operating System. Back in January, Sun released the code for DTrace, a dynamic tracing tool for analyzing and debugging kernel and userland events. DTrace is one of the big features in Solaris 10. Some other highlights include the GRUB bootloader, SMF (Service Management Facility) which replaces init.d scripts, it starts up processes in parallel for faster boots (7 second boot on a dual opteron workstation I think that was the setup) as well as providing features for automatically restarting. OpenSolaris provides support for x86/x86-64 processors as well as Sparc. The Blastware guys are working on Polaris which is an OpenSolaris port to PowerPC. Sun has been working on opening Solaris for over a year now. The OpenSolaris project started with a pilot group of Sun and non-Sun users. During the pilot program a lot of info including screenshots could be found on various OpenSolaris member blogs. (My favorite is Ben Rockwood's blog). Teamware is the source code management system Sun uses for Solaris and OpenSolaris. Which was designed by Larry McVoy (now of BitKeeper) while he was at sun. No word yet on if Teamware will be available for OpenSolaris developers or not. Sun also uses CollabNet for it's Open Source project websites so that might be a possibility as well."

Mark Stone writes with a thoughtful look back at the year 2004 in open source, pointing out both major gains and inevitable uncertainties. He writes "2004 stands out as a year in which open source consolidated its position as a valuable and accepted approach to business and technology policy. A less obvious but significant trend underlies all of this: even as open source business models join the mainstream, the open source development model remains a mysterious process on which large technology companies struggle to capitalize. Key issues and developments have played out in four areas: legal, policy, business, and technology." Read on for the rest. Legal

The biggest non-story of the year was SCO's legal efforts. So far SCO has not been able to make substantial headway with a single one of its legal claims, and indeed has suffered a number of significant setbacks in court.

This is certainly good news for Linux and open source. Going back five or six years, clearly one of the major obstacles to widespread adoption of open source software was the uncertain legal status of both the software and the licenses. While this aspect of open source is still an unfinished saga -- more on that shortly -- the inability of SCO, through either legal or PR channels, to undermine Linux gives reason for confidence about the future.

The real story about SCO in 2004 has in fact been the telling of that story. While mainstream media coverage of SCO has varied widely -- sometimes accurate, sometimes resembling coverage of the OJ Simpson trial -- Groklaw has emerged as a steady voice of reason and objectivity adeptly defusing all attempts at "FUD" PR around the case.

2004 has been, especially as an election year, a controversial year for the phenomenon of blogging. Whether blogging will provide a sustainable alternate voice in journalism is very much an open question. A few blog sites, however, have shown what a handful of dedicated individuals can do in the face of much larger, and better funded PR machines. Groklaw is an outstanding example of the positive journalism effect that blogging can have.

The legal front brought other good news for the open source community. Norway's Supreme Court acquitted Jon Johansen, and the Norwegian Economic Crime Unit opted not to appeal the decision. In the United States the Digital Millenium Copyright Act still remains the law of the land, but the Recording Industry Association of America has made little progress in forcing ISPs to disclose the identities of alleged file swappers.

A more troubling legal trend is the shift in debate about the intellectual property status of open source software. The principles behind the "copyleft" approach have gained continued acceptance, and have even been leveraged as an integral part of some business models. The debate now, however, centers more around patents that copyright.

IBM has been out in front of the patent issue. Their open source license was the first to explicitly address patent licensing as an issue above and beyond copyright, and they've taken steps, even recent steps, to see that open source development is unencumbered by patent concerns. IBM is not the only company putting patents in the open source domain. Sun Microsystems recently announced they will make patents available under their recently approved Common Development and Distribution open source license (CDDL).

All of this would seem to be good news for the open source community, especially given that Poland's objections have put a temporary halt to the Europan Union software patent initiative. Appearances can be deceiving, however. IBM is a supporter of software patents. Sun's gesture is in fact intended to create a competitive advantage for OpenSolaris over Linux, since the patent protection Sun offers applies only to work licensed under the CDDL -- in other words, not Linux. In a recent News.com commentary, Bruce Parens said, "So while claiming to make the patents available to open-source developers, Sun can sue folks who work on Linux rather than Solaris."

The biggest patent concern comes from Microsoft. In a speech in Australia, Microsoft CEO Steve Ballmer claimed that Linux violated more than 200 patents. While this may be more hype -- or hope -- than fact, it does tip Microsoft's hand in terms of what tactics they are willing to use to meet the Linux competitive threat.

Policy

All other things being equal, customers prefer an open system to a closed one, and vendor choice over vendor lock-in. In the IT world in general, and between Windows and Linux in particular, all other things are not equal, which makes platform choice complicated. More and more, however, organizations are seeing Linux as a viable platform choice that

• Lowers up-front licensing fees
• Has the support and backing of significant technology vendors, whether small, medium (Red Hat), or large (IBM, Novell)
• Avoids vendor lock-in at both the platform and application level

These claims are independent of the more controversial claims about improving security and lowering total cost of ownership. 2004 has added an interesting additional element to the mix: the desire of government organizations outside the United States to not be dependent on a large, American technology company whose revenues exceed the gross national product of most nations.

This software declaration of independence has taken several forms. Sometimes it seems simply to be a negotiating tactic to force Microsoft to lower prices. India may be an example.

Sometimes, however, price is not the issue. Munich, for example, committed to making the switch to Linux despite direct lobbying efforts by Microsoft CEO Steve Ballmer. In the case of a high tech country like Germany, this decision is probably influenced by the reluctance to be dependent on an American company guilty of monopoly practices.

The situation in the developing world is somewhat different. Unshackled by significant requirements of backward compatibility, emerging economies like Venezuela's have a chance to make a clean start and avoid what they perceive as the pitfalls and inefficiencies in older IT infrastructures.

The policy approach in China is even more alarming to traditional technology vendors. China clearly does not want to build an economy dependent on outside production or services, whether it's factories or satellite launches. In the software world China has made it clear that it can and will build its own platform and application stack leveraging open source components, if that is what it has to do to maintain control of its software destiny.

Business

The North American market for computer technology has, in many ways, reached the saturation point. A Pentium 4, to say nothing of a 64-bit processor, is already overkill for most office desktop applications. Older versions of the Microsoft Office suite, and older versions of Microsoft Windows, are often quite adequate for business productivity needs. The problem for traditional technology vendors is aggravated by the fact that Linux, Open Office, and other open source software may now be good enough.

On the one hand this accounts for why policy issues and the international technology market have become so important: this is where technology vendors see the biggest opportunity to grow new business. On the other hand, open source is forcing some significant changes in the software market domestically.

The most visible effect of open source has been the commoditization effect. Microsoft, as we've seen, has been forced to acknowledge the competitive impact Linux is having, and to cut prices overseas in response to this competition. Yet even companies like BEA acknowledge that open source will have an increasing commoditizing effect, meaning that they will cede lower levels of the application stack to freely available open source software and seek to add value further up the stack.

The most dramatic concession to commoditization in 2004 has been the announcement that Sun is open sourcing Solaris. Said one Sun executive who asked to remain anonymous, "Do you think we'd be open sourcing Solaris if we had any other way to compete with Linux on price? Of course not."

If anything, the opening of Solaris reinforces that Sun has been unable to find a business model built around Linux. Given that competitors like IBM and HP have, with varying degrees of success, been able to integrate Linux into their business models, one suspects that there are deeper problems at Sun than the opening of Solaris can solve.

The bottom line is that Sun is still trying to compete with, rather than embrace Linux. The CDDL doesn't extend patent protection to anyone working under a different open source License, and the CDDL is incompatible with the GPL, meaning none of the Solaris code can be used to benefit Linux.

This move, of using a license as a competitive tool, is one of the more subtle but more important business trends to emerge from open source in 2004.

The most common approach is a dual-licensing scheme, utilized by Trolltech (for Qt), Sleepycat (for Berkeley DB), MySQL, and newcomer db4objects, among others.

In each case the company makes its core product available under the GPL, or else under a similar viral-type license. Since each of these software products is intended to be embedded within or combined with other software to create a derivative product, companies are forced to make their own product available as open source, or to approach the originating company about separate licensing under proprietary terms.

The result is a very low-cost distribution mechanism for the open source companies, as well as a cheap in-bound sales channel of pre-qualified leads.

Of course, to be able to dual-license, you must have created all the code in question, or have full rights granted to you for all the code in question. Thus this very successful open source business model is incompatible with the open source development model; each of the companies using the dual-license approach does all, or nearly all of their software development in-house.

Technology

What then of the open source development model? Has it enjoyed the growth and widespread acceptance that open source business models have?

Certainly 2004 saw a number of significant releases for open source projects. GIMP 2.0 was finally released, as was Gnome 2.6. Large companies as well as individual projects made strides. IBM announced the release of its Java database, Cloudscape, as open source. Novell released SUSE Enterprise Server 9.

The year's most significant releases were the 2.6 series of Linux kernels, and the 1.0 release of Mono. With 2.6, Linux now has many of the features needed to compete as an enterprise-class server: better multiprocessor support, failover and hot-swap support, better journaling file system support.

Mono is absolutely critical if the open source community is to compete in the application development market. C# and .Net will be important application building blocks for the forseeable future, and Linux and open source need to be viable approaches.

The Debian Project has undergone an interesting evolution in the last year. Long-time Debian users have often complained about the slow pace at which Debian moves, favoring security and stability over feature growth. The result is a very solid server system, but one that, for the end user, often lacks support for advanced hardware.

The solution, which seems so obvious now, is independent distributions that leverage Debian as a base but target the end user with ease-of-use features and hardware-support features that have yet to make it into Debian. Two successful projects heading down this path are Ubuntu, which follows the Gnome approach to usability, and Mepis, which follows the KDE approach to usability. Either distribution will give you an easy install, access to Debian packages and apt-based network updates, but with more advanced hardware support and an improved UI over stock Debian.

By far the biggest development story of the year, however, has been Firefox, the browser component of the Mozilla project.

Timing is everything. Security, privacy, and spyware have become major concerns in 2004. Microsoft has refused to significantly update Internet Explorer (IE) until Longhorn is released, which could be in 2006 (as in "Santa Claus could be real"). The Mozilla Foundation capitalized on this opportunity with a major fundraising blitz for the foundation and PR blitz around Firefox; this included a full-page New York Times ad.

In November, Firefox 1.0 was released, and to date downloads exceed 10 million. Mozilla has raised over $250,000 in its fundraising campaign. While IE's market share still hovers around 90%, Firefox has rapidly grown to 5% market share, and put a dent in IE's market share for the first time in years. Industry analyst Gartner Group has looked at the results of 2004 and declared the browser war open again.

Looking ahead to 2005, it's interesting to ponder the tech sector's differing response to open source business and open source development models. The business models are reasonably well understood and generally accepted now. Not everyone is leveraging open source as a business play, but everyone understands it is one viable strategy to pursue.

On the development side, however, the results of open source continue to confound the establishment. Why did no one see the Firefox phenomenon coming? Equally important, why isn't anyone (AOL) attempting to leverage Firefox's market success and technology advantages?

With Solaris, it's interesting to note that even supporters of OpenSolaris admit it sees no real development savings to opening Solaris; the benefits are all on the marketing side. Ben Rockwood blogs "It's going to take Sun more work to maintain it open source than it will to just leave it closed."

Yes, open source has become mainstream. But that mainstream presence needs to be more than a commodity benefit to companies willing to leverage the results of open source. Will mainstream technology companies figure out how to anticipate and collaborate with open source development as a deep part of their technology strategy? That's a big question that 2005 may answer.

Mark Stone is an open source consultant and freelance writer living in the Sierra Nevada region of Northern California. He can be reached at mark.stone@gmail.com.