If open source on its own doesn't prevent vendor lock-in, what's to stop Linux from ending up like Unix?
Jim Zemlin, executive director of the Free Standards Group , is on a mission. He wants to prevent Linux from falling prey to the fragmentation fate that befell Unix.
Zemlin was among those at the LinuxWorld Canada show talking about the Free Standards Group and its latest Linux Standard Base (LSB) certification, version 3.1. The LSB specification is intended to provide interoperability standards via a base set of APIs and libraries. Ideally, ISVs will develop and port applications that will work on LSB-certified Linux distributions.
Its last major release, version 3.0, came out in September 2005. The 3.1 specification adds components that extend the standard to the Linux desktop. Open source in and of itself is not a standard, Zemlin argued, and it will not necessarily prevent vendor lock-in.
"Open source is a development methodology, and choice is not guaranteed by a development methodology," Zemlin told the LinuxWorld Canada audience.
Zemlin argued that the choice of open source provides brings risk. For example, what if a user picks the wrong version or the wrong project? Users are looking for "one throat to choke" as the adage goes. Commercial Linux vendors Red Hat and Novell want to be that throat.
But the desire for a throat to choke and the need for choice aren't mutually exclusive.
"There is a real fundamental paradox that I think the open source movement and end users need to solve, and that's the paradox of throats to choke vs. vendor lock-in," Zemlin said emphatically. "That paradox is that the throat you are choking is the vendor who is locking you in.
"Open source as a development methodology in and of itself cannot solve the problem," Zemlin continued. "You can get just as locked into an open source implementation of something as you can a proprietary version."
The solution is open standards, which is what the LSB is attempting to achieve. The LSB and the open standard that it professes provide a degree of assured interoperability that will allow users to move their data and applications to another platform should they want to.
Unix also had a standard that was supposed to promote a degree of standardization. But according to Zemlin, POSIX covers far less than the LSB and never achieved the degree of interoperability that was necessary for application vendors to deliver their solutions across platforms. LSB is POSIX-compliant.
Zemlin also argued that there are fundamental economic reasons why the LSB effort will not end up like efforts on Unix in the past.
"Why are we not going to end up with Red Hat and Novell ending up like AIX and Solaris? These are different economic times," Zemlin said. "System sales are no longer tightly tied to the hardware systems themselves. Operating systems software does not produce big software-license margins anymore."
The interests of the systems vendors, software vendors and end users are all aligned in Zemlin's view.
"IBM, HP and the big systems vendors' interests are aligned with the software vendors' interests who just want to target the most users at the lowest cost," Zemlin said.
"Those interests are aligned with the distribution vendors who just want to grow the overall market who are aligned with end users who just want choice.
"For the first time historically you've got all these interests that are aligned."