The Open Road

Brian Behlendorf of Apache fame once declared the freedom to fork the cardinal rule of open source. He is right, though it's a freedom that is rarely exercised, and even less rarely exercised to good effect.

But on Wednesday a group of developers announced ICINGA, a fork of Nagios, the popular open-source network monitoring tool.

While it's too early to tell whether the fork will succeed, the action already demonstrates both the health and disease of the Nagios community.

Health, because a fork or spin-off of the original project, demonstrates that there is an active community of users and developers that cares enough about the project to ensure it's done "right" (i.e., according to their preferences).

Disease, because clearly the core Nagios developers weren't serving the broad Nagios community well enough. In fact, the ICINGA developers write:

This independent project strives to be more responsive to user requests and faster in software development through the support of a broader developer community.

While there have been few successful forks, ICINGA can learn from those few. Joomla!, for example, has done marvelously well outside the Mambo project, and Openbravo (Disclosure: I am an advisor to Openbravo) and Adempiere have both thrived beyond Compiere. So, it can be done.

I tend to view forks as a sign of strength, because they suggest a broad-based community that cares passionately about the project. With this in mind, I wish both that Nagios and ICINGA projects the best of luck. (And I hope ICINGA will stop capitalizing all of its letters.)

Follow me on Twitter @mjasay.

The Register paints a very unflattering picture of Sun Microsystems' alleged mismanagement of its hardware and software assets.

Unfortunately, there's likely a lot of truth to the argument, though it's easy to point fingers from the outside and tell others what to do.

But this is precisely why Sun should be grateful for Oracle's acquisition of its assets: Oracle needn't appease internal or customer lobbies. It just needs to determine what pays the bills, and shutter or sell everything that doesn't.

The one open question for me, however, remains MySQL. Oracle could do much with the technology, but I'm skeptical that it has much incentive in ensuring a long and prosperous future for MySQL.

Regardless, it may not get the chance. As reported by OStatic, MySQL co-founder Monty Widenius has suggested that MySQLers' life under Sun was rocky enough, but Oracle may convince them to bolt:

Sun's acquisition of MySQL did not go smoothly; most of the MySQL leaders (both commercial and project) have left Sun, and the people who are left are sitting with their CV and ready to press send. Oracle, not having the best possible reputation in the open-source space, will have a hard time keeping the remaining MySQL people in the company or even working on the MySQL project.

Given the fracturing we've already seen with MySQL, what with OurDelta, Drizzle, MariaDB, and other variants on the MySQL theme emerging in the past year, I suspect that we may be in for several more forks of the MySQL code base. There's simply too much at stake in the database layer of computing to allow MySQL to be submerged by Oracle's other database priorities.

So here's a thought: could Red Hat fork MySQL, hire some key developers, and effectively assume the mantle of MySQL leadership?

I doubt that it has that ambition, as it would end up hurting its still-strong partnership with Oracle. It is more likely that Red Hat would offer to buy MySQL, if it made a move for MySQL at all, and I doubt that the two could find a mutually agreeable valuation.

Regardless, unless Red Hat could replace MySQL's dependence on InnoDB, it would lack the means to truly create an independent fork of MySQL. By controlling InnoDB, the primary storage engine for MySQL, Oracle effectively controls MySQL, regardless of whether it owns the MySQL code.

I'd like to see MySQL in Red Hat's hands. But Red Hat hasn't shown much near-term desire to get far beyond Linux. We're going to have to wait to see Red Hat become the full-stack competitor to Microsoft that some of us would like to see.

Follow me on Twitter @mjasay.

For anyone interested in seeing just how different and game-changing open source can be, there's really no need to look beyond MySQL, the open-source database leader. Jeremy Zawodny, formerly of Yahoo, and now of Craigslist, takes a hard look at the changing face of MySQL, reaching some surprising conclusions about MySQL in the process:

Nowadays MySQL has a much slower release cycle than it used to. It's still available in "commercial" and free ("community") releases. There's still a company behind it--a much larger one in fact. But one that also has a vested interest in showing how it works better on their storage appliances or 256 "core" computers and whatnot...

Meanwhile, all the cutting edge stuff (at least from the point of view of scaling) is happening outside Sun/MySQL and being integrated by OurDelta and even Drizzle.

Zawodny details the importance of these forks to MySQL ("The single most interesting and surprising thing to me is both the number and necessity of third-party patches for enhancing various aspects of MySQL and InnoDB"), and it's here that one sees the strength of the open-source model, but also the potential fragility of open source as a business, as I've written before. These forks provide a robust MySQL database...for free.

This is good, right? Well, it is, but perhaps not if you're MySQL (or, rather, Sun), the company. For all the benefits such forks and additions provide to MySQL, they absolutely depend on Sun doing the core development on the MySQL database, core development which becomes ever more difficult to fund if such peripheral projects siphon away Sun's return on the MySQL investment.

It would seem to me that the best way for this vibrant community around MySQL to become good for the corporate MySQL would be for the community to become so active and diverse that the MySQL database begs for standardization at the core again. Sun can provide that, making enterprise customers happy and, in turn, making Sun happy.

One thing is clear: Sun needs to immediately start releasing its own "fork" of the MySQL database, one that is tuned to enterprise requirements, and one that includes functionality/tools that customers can't find elsewhere. If it's fair for Drizzle, OurDelta, Percona, etc. to enhance and extend the MySQL experience, then it's fair that Sun do this, too. Only as Sun creates differentiated value will it ensure an ongoing, rising revenue stream that will enable it to fund MySQL development, development upon which these forks critically depend.

In most companies, there are prohibitions against creating competitive, derivative works of the company's intellectual property. At MySQL (now Sun), well, things may be a little different.

As announced at the O'Reilly Open Source Convention, Brian Aker, MySQL's director of architecture, has launched Drizzle, an optimized and trimmed-down version of the popular open-source MySQL database.

In other words, MySQL has forked itself. "The right to fork" is, of course, a cardinal right of open source.

But forking is usually driven by rival factions on a project (e.g., the Adempiere developers forking Compiere). In MySQL's case, its own employees created the fork, a fork that has the blessing of Sun's senior management, according to MySQL co-founder Monty Widenius.

Personally, I find it a bit odd. If the fork was needed, why not work within the company to offer it as a separate product? But then, for those who have worked with passionate open-source developers like MySQL employs, sometimes the best policy is simply to step back and let the magic happen, even if it initially appears not to be in the company's interests. Perhaps this could end up being a supported database for Sun?

Martin Peacock sent me a link to this fascinating study of ant behavior in The International Herald Tribune. The article tracks the research of Iain Couzin on ants, locusts, and even humans and their instinct and ability to swarm.

While it doesn't call out open source specifically, I found the "follow-the-leader" behavior corresponds nicely to the forking of open-source projects. Despite the talk about the importance of the fork to open source, we actually rarely see it happen. Why? Probably because the group inertia is such a strong force:

Couzin and his colleagues have built a model of the flow of information through swarms. Each individual has to balance two instincts: to stay with the group and to move in a desired direction. The scientists found that just a few leaders can guide a swarm effectively. They do not even need to send any special signals to the animals around them. They create a bias in the swarm's movement that steers it in a particular direction....

... Read more

Proprietary vendors sometimes point to the possibility of open-source fragmentation as one of its great weaknesses believing, as they do, that the vendor should always control its own destiny. What such vendors fail to see is that a community's right to fork a project is actually its greatest strength. The fork is the community's most valuable tool for ensuring the ongoing health of the project.

OpenOffice.org is feeling this now, as a cumbersome community process has led prominent members of the OpenOffice community to lay down the law (er...the fork) and declare independence, as Kohei Yoshida does:

If Sun insists on rewriting all the work I've already done just to ensure that they own all the code in OO.o, even though it is legally permissible to integrate my code under a pure LGPL license as an external component, then perhaps I need to re-think my relationship with the project. ... Read more

If you think you're avoiding open source by buying a "commercial" (read: proprietary) product, think again. Gartner predicts that >80% of all "commercial" products will include open-source software by 2010.

So, you can buy open source outright, or you can pretend it doesn't exist and buy it in the form of proprietary layers on top of open-source projects.

Either way, you're going to be using a heck of a lot of open-source software over the next 20 years. Everything you use will be a product of open source, in some way. Better get used to it.

Gartner writes: