The Linux marketplace has been growing exponentially over the past several years. Concurrently the demographics of the Linux customer have shifted from highly technical to the more commercially focused. We are at a critical point that will determine whether the Linux market will continue its rapid expansion into the broad commercial marketplace, or whether it will remain limited to the more technical markets where it is currently strong. The Linux Standard Base (LSB) project is one component that will help propel Linux growth into those broader commercial markets, by making it easier for Independent Software Vendors (ISV's) to support the Linux marketplace with their products.
I am interested in promoting the growth of the Linux market, and that is the perspective that will serve as the background for this commentary. My byline states that I am a Director of Linux International (LI), and President/CEO of Enhanced Software Technologies, Inc. (EST), and I am wearing both hats as I write this editorial. You may not have heard of EST, but you may have heard of our backup product BRU (now BRU 2000). BRU has been ported to virtually every flavor of UNIX, and has been available for Linux since 1994. At one point, we supported BRU on about 30 different UNIX platforms. As an ISV, we would prefer that there be just one Linux port required. When Tim Jones, EST's VP, alerted me to the glibc issue and the problems it was causing for our customers, I brought it up to the LI Board. Further discussions ensued during a small Board meeting at Cebit. On June 28, at a meeting of the Board of Directors of Linux International, a discussion was held on Bruce Perens' proposal for a Linux Standard Base. The meeting was attended by representatives of most of the member companies of Linux International, as well as several non-LI participants, including a representative of Debian. Without rehashing the entire meeting, these were the key outcomes:
There has been a great deal of conversation about the LSB, mostly from a technical perspective regarding what will or won't be included, and who is or isn't participating. With regards to the technical issues, I don't have much to contribute. If the goals stated above are met, then the LSB project will be a success. I trust in Bruce's ability to lead the project and to provide a technical solution that will fulfill the stated goals.
With regards to which distributions do or do not participate, I would like to clarify the situation. The project as a whole, and the discussion surrounding it has been blown out of proportion. As the project stands right now, it is nothing more than the above stated goals. None of the companies who have indicated support for the goals of the project have committed to using it when it is completed. I have read Patrick Volkerding's opinion on the project, and I understand his concerns. I invite him to reconsider. Can Slackware align on the goals of the project? If so, I invite Patrick to contribute some resources to the project. If Slackware participates, they will have as much input as anyone else in determining the final product. The team working on this project will include ISV's as well as representatives from the various distributions, and, I believe, a significant number of individuals not aligned with any particular distribution. The invitation will remain open to any distribution to make use of the LSB when it is completed, regardless of whether they participated on the team at the start. It will be a community project.
A successful LSB project will accelerate the decision making process for ISV's who are considering porting their applications to Linux. As more commercial applications become available, the market for Linux will continue to expand. The recent announcements from Corel regarding Linux ports of their products are just the beginning of what is possible. Another announcement or two from major software vendors of Linux ports (especially if a major DBMS vendor steps up to the plate) will create the momentum to bring additional ISV's into the fold. These ISV's will choose to certify their software to be compatible with Linux distributions that meet the LSB. Commercial customers will choose distributions based on the availability of supported applications that meet their needs. Distributions that choose not to support the LSB may find themselves with a limited market niche of their core highly technical constituencies.
The LSB is not the only component necessary to aid commercial acceptance of Linux, but it is a critical component, and it is doable now. Linus speaks of "World Domination" for Linux, and it is only partly tongue in cheek. Great things are accomplished by people who are willing to take a stand, to throw their hat over the fence, and say what will be in the future. Microsoft is where they are now because, long before it seemed possible to most, Bill Gates said there would be a computer on everyone's desk running Microsoft software. As a community, we can take the stand that there will be a choice in operating systems, not just for the technical elite, but for commercial customers as well.