Yesterday, Progeny joined Conectiva, Mandrakesoft and Turbolinux in announcing the Linux Core Consortium, a development partnership to build a reference implementation of the Linux Standard Base to be incorporated in our respective Linux distributions. Details can be found in the press release as well as in media coverage from CNET News.com, eWEEK, Techworld and others.
The goal of the LCC is simple: To build a common core (i.e., a single set of binaries that constitute an implementation of the LSB 2.0 standard) in an open, collaborative fashion, and to get as many distributions as possible to base on that common core. The initiative is supported by such industry heavyweights as Computer Associates, the Free Standards Group, HP, Novell, OSDL, Red Hat and Sun.
Why is the LCC important? The name of the game today in enterprise Linux are ISV and IHV certifications, and the name of the game in attracting these certifications is volume. Individually, the volume leaders today are Red Hat and Novell/SuSE, which is why they attract the lion’s share of the ISV and IHV certifications. However, one of the main attractions of Linux is freedom of choice and diversity, and there are literally hundreds of other distributions that target geographies or markets Red Hat and Novell/SuSE don’t (or can’t) adequately address.
Individually, most of these specialized distributions don’t have the volume to attract ISV and IHV interest. However, as a group, these distributions represents a powerful, global force. It could be said, then, that the goal of the LCC is to unify these distributions around a common core to realize that powerful, global force. Furthermore, because that common core will be created in an open, collaborative fashion between the participating distributions, the result ensures an open, inclusive process rather than one defined solely by one or two vendors.
Of course, a large number of specialized Linux distributions are based on Debian, which is why the LCC is committed to “increasing interoperability between Debian and RPM-based technologies and [working] toward a common binary core that can form the basis of both Debian and RPM-based distributions.”
One open issue, though, is how to facilitate this. How can the LCC produce a single set of binaries that can form the basis of both Debian and RPM-based distributions? One possibility is simple enough: to produce the common core in both package formats (Debian and RPM), but this would essentially mean providing a different set of core packages than Debian provides, which doesn’t make it a particularly attractive option. If your company or organization supports Debian or produces a Debian-based distribution, I’d love to hear from you. This could finally represent an opportunity for us to provide ISVs and IHVs with a way to support Debian. There’s plenty of demand out there–the ISVs and IHVs just don’t know how to engage us.
[OT]
You should upload discover.
It took me a while to realize the point of all this, but now that I finally get it I find it incredibly pointless. The only purpose of the LSB, LCC, and similar “binary standards” is to make it easier for third-parties to provide binary software that works on more than one distribution. Given that all Free Software is available in source form, it can either be packaged for a distribution or put in an appropriate location via ./configure or a similar process. Therefore, these standards only really benefit proprietary software vendors. Hence, I’ve personally ceased caring about them in the least.
If there’s free software out there that does everything you need, then you’re right–this doesn’t really impact you. However, there are a lot of people that don’t fall into this category, particularly in the business world, and for them, this kind of initiative is absolutely essential. For some additional perspective, please see the “Looking Ahead” part of my Debian retrospective (http://www.linuxplanet.com/linuxplanet/editorials/4959/1/):
“The commercialization of Linux is at a crossroads, and it is much the same crossroads that UNIX reached in 1993. The delicate balance of the Linux ecosystem, which is what makes Linux valuable, is being threatened by a certain few who would like nothing more than to own it. To the extent that any company is successful at ‘productizing the ecosystem,’ the ecosystem will cease to be an ecosystem, and the very thing that is different about Linux, the very reason that we are all here, will be gone.
“Some say that Linux will never suffer the fate of UNIX because of the GPL–and from a community perspective, they’re right. No company will be able to ‘own’ Linux, because Linux is not ownable.
“My concern here is not the Linux community, which will do just fine either way, but rather Linux in the commercial sector. There are more ways to lock in commercial users than just intellectual property, and we’re seeing this strategy play out today. The Linux opportunity is enormous, and the opportunity is in the ecosystem around Linux, where any number of companies, large and small, can benefit and coexist. It would be a shame if this ecosystem were to be destroyed.”