With so much compatibility among distributions, the LSB (Linux Standard Base) might not seem that necessary. Linux already apparently enjoys a level of cross-vendor compatibility the UNIX community never achieved. But, going forward, the Linux community might not be so lucky. If successful, the LSB will ensure that the various distributions retain their current high degree of compatibility and prevent any one vendor from hijacking the Linux standard.
The red threat?
Even though large companies such as IBM, Hewlett-Packard, Compaq, and Dell have become Linux enthusiasts, they aren't seen as posing a significant threat to retaining a single Linux kernel.
"I'm more worried about some small company that gets off to a quick start, then puts a lot of resources into development, and absconds with the direction of the OS to its advantage," says TurboLinux's Lonn Johnston, vice president of North American operations.
In the Linux community, that "small company" everyone is concerned about is Red Hat, which is generally acknowledged to have the largest market share of commercial Linux shipments in North America.
Bob Young, CEO and chairman of Red Hat, however, says his company is as committed to the LSB as any other Linux distribution company, if not more so. He says Red Hat has committed more resources, including staff members, to the LSB effort than just about any other commercial Linux distributor, a claim backed up by LSB chair Dan Quinlan.
This is the second article in a series. In the first report, the history of the LSB was outlined. This content originally appeared in the September issue of Wiesner Publishing's Software Magazine and appears on TechRepublic under a special arrangement with the publisher.
Red Hat, says Young, hasn't been as vocal in its support of the LSB because "it doesn't exist yet." When it does, Young says, Red Hat will be one of the first to submit its operating system for testing and verification, and to loudly endorse it.
Young says Red Hat is working quietly behind the scenes on the LSB, and doesn't want to give the impression that the success or failure of Linux depends on the success or failure of the LSB. "Linux will succeed without the LSB," says Young. "Will the LSB accelerate the growth of the open source software movement? Yes, but it doesn't depend on the LSB."
On that point, Caldera's Tamang disagrees. He says it's vital for the LSB to be a success, apparently if for no other reason than to convince Linux's many potential users and Independent Software Vendors (ISVs) that the operating system will retain what he calls its "port-once" philosophy, a reference to the popular "write once, run anywhere" theme of Sun Microsystems' Java programming language.
"Efforts to standardize, to make sure Linux is Linux is Linux, is a very good move," says Dan Kuznetsky, director of operating systems research at IDC.
The problem for the Linux community, he adds, is that the standardization wasn't built into Linux at the beginning. "The problem is that it has to be applied retroactively," he says.
Current Linux distributions will have to be reworked to comply with the new standard, and that means work, "and people often don't want to do that," says Kuznetsky. "They'd rather concentrate their resources on something that might advance their distribution ahead of others."
Still, he says, the lessons learned from the fragmentation of UNIX have not been lost on the Linux community. "Everybody's life will be much better," says Kuznetsky. "They know that if there's a standard, and they all adhere to that standard, the overall market does much better."
If they all keep to the same road, says Kuznetsky, "they know that they can all have a slice of a bigger pie."
Philip J. Gill is a former editor-in-chief of UnixWorld Magazine. He writes regular monthly columns for The Java Report and is a frequent contributor to InformationWeek, Oracle, and Profit magazines.What can the IT community do to ensure the future of Linux? What have we learned from UNIX? Post your comment below or send us a story idea .