Linux Today: Linux News On Internet Time.

More on LinuxToday

Editor's Note: Free Software is Not Anti-Business

Apr 22, 2005, 23:30 (9 Talkback[s])
(Other stories by Brian Proffitt)


Desktop-as-a-Service Designed for Any Cloud ? Nutanix Frame

By Brian Proffitt
Managing Editor

Just back from LinuxWorld Canada in Toronto this week, and I pretty much enjoyed the show. Great city, very nice attendees, and the show organizers were really on top of things.

The vendors were the usual crew: Novell, IBM, and Red Hat; with a notable lack of Sun (go figure). There smaller vendors were all less familiar to me. Part of this was because of the location, and part because LinuxWorld was co-located with NetworkWorld this year in Canada.

(Though it was hard to tell. I mentioned to Jeremy Garcia of LinuxQuestions.org that I din't know why Canon was there, except for NetworkWorld, and he reminded me that Canon was at LW in Boston earlier this year, too.)

All of the commercial players are really making an effort to define their role in the Linux community, and I received a good lesson in why it's not good to paint all vendors with the same brush.

Over the past few weeks, I have been examining the relationship between software that is free and software that is open source. This has not been done with any particular end goal in mind; I have no holy crusade to make, and no axe to grind.

What prompted this examination was the recent activities of several entities, both commercial and not, which seemed to signal a new movement within the open source community away from Linux.

Some have criticsized this observation, citing that Linux is open source, and I must be spreading FUD for detracting open source. Let me be clear: Linux is a part of open source, and open source is more than just Linux. It is possible to separate elements within the open source community away from Linux, made easier by the fact that some elements, such as Sun or perhaps Computer Associates, are doing a darn good job of separating themselves.

My position on this is straightforward: I want Linux the operating system to succeed. Up until very recently, I wanted open source as a whole to succeed, too. Now, from my perspective, it seems this may be a conflicting set of priorities.

This is a tricky area of discussion, since it is full of semantic landmines that could easily blow up in my face and make my point less clear. I will try my best and hope for constructive suggestions. One more clarification: I am talking about commercial Linux, not the technical aspects of Linux, or the non-commercial distros. Those are alive and well, and I have no doubt will continue to thrive no matter what happens in the commercial arena.

What makes this topic all that more difficult is trying to figure out the motivations of all the players involved. Sun... they're pretty clear. Open Source Institute... I think their ducks are all in a row. But than we get to other big players in the Linux arena, and it becomes more of a prognosticating effort. To be honest, with all of the negative aspersions I have been assigning to the commercial players lately, it was very easy to assign the same suspicion to companies like Red Hat, IBM, and Novell--even though they haven't made any moves that signify problems for Linux.

I got caught out by this during the Canada show this week, and I felt pretty embarassed after the fact. Red Hat, whom I have chastised in the past for their agressive vendor relationships, was frankly an easy target for commercial == bad for Linux. Now, I think I need to need to stop overgeneralizing.

What brought me to this realization was a session at the conference delivered by Red Hat's Michael Tiemann. He and I have spoken in the past, usually at the LinuxWorld shows, usually for just a few minutes at a time. I have only heard him speak once before, and that was before the Fedora Project was launched.

His talk, "Mastering the Open Source Triple Play," was a fairly non-marketing oriented explanation of why Linux is a good fit for any given organization. He delivered this message in high-level business analyst mode, and made a very good case for how Linux implementation is not only good for total cost of ownership, but it also can start a chain reaction within an organization--lower TCO means more money for collaboration efforts, more collaboration means more money for market expansion and customer relationships. Much of this is based on user-driven innovation, a concept that's making the buzzword circuit these days.

User-driven innovation, simply put, means taking the innovative process away from the business and putting it in the hands of the customer. One historial example of this cited by Tiemann was James Watts' steam engine, which, while Watt held the original patent, only saw an improvement to the 20% efficiency level, based on the Detroit engine standard. But, after Watt's death and the expiration of the patent, the user community managed to improve the engine up to the 60% efficiency level.

This approach is ultimately what drove Red Hat to their enterprise Linux sales model. By shifting the innovative focus for RHEL out to the community via Fedora, Red Hat not only saved a lot of time and money, but they were able to tap into the resources of the community to get things like SELinux and soon stateless Linux up and running.

Of course, this move was not without its detractors, myself included. Many felt that by turning their support away from Joe User and to James Suit, Red Hat was effectively abandoning the community. While I cannot change the feelings that people had when the move to the Fedora/RHEL model was announced, I am beginning to come to the realization that this may have been one of the best moves Red Hat could have made to keep themseleves alive.

Certainly Tiemann thinks so, as he cited Geoffrey Moore's Crossing the Chasm model as being the roadblock to Red Hat's continued success. By slowing down the RHEL release cycle and implementing a seven-year support model on the enterprise product line, Red Hat effectively jumped the chasm from the early adopters to the much more numerous and much more wealthy mainstream adoptors.

Others may be following suit. While they have not separated into separate community/enterprise product lines, Mandriva did announce a slow down in their release cycle--something that will save them a lot of resources and ideally improve their bottom line that much faster.

Red Hat's is not the only commercial model a Linux distribution can take, but right now it seems to be working well for them. And, given Tiemann's staunch advocacy of the free and open aspects of Linux, I do not think that this commercial involvement is going to detract from the community aspects of Linux.

Like any sweeping judgement, saying that commercial is bad for Linux is just not going to work. When handled properly, it seems, commercial interaction can be a positive thing for Linux, and can offer proof against the notion that free software is anti-business.