Red Hat's Project Spacewalk could make it the hub in the open-source wheel

Back in early 2007 Red Hat let slip that it was planning to release its Red Hat Network code as an open-source project. In June of 2008, Red Hat officially announced that Red Hat Network Satellite would be open sourced.

Last week, Red Hat posted an update on the project, now called Project Spacewalk.

In the nine weeks or so since the debut of Spacewalk, we've been blown away by the level of interest, the contributions, and the excitement generated by the project...

• [e-mail] : currently has over 250 members...
• [e-mail]: currently has about 120 members...
• The first patch from inside Red Hat came within three days of the opening of the mailing list.
• The first patch from the community came within eight days.

I've suggested before that the company that owns the heart of open-source monetization would be sitting on a massive opportunity. Yes, there are alternative ways to monetize open source (e.g., Google's advertising model), but for many years to come vendors will make money by distributing software, not merely advertising around that software.

As such, a community effort around a network service, such as Red Hat's Project Spacewalk, is hugely important. It's important because it provides Red Hat a way to corral the growing commercial open-source ecosystem.

To achieve this more effectively, however, Red Hat needs to reach out to the commercial open-source ecosystem and evangelize the benefits of building on Project Spacewalk, rather than creating silo'd "Red Hat Network-esque" offerings. To date, Red Hat seems to have taken an "If we build it, they might come" approach to Spacewalk. It needs to be a bit more proactive.

[Philips]

It's as good as useless if it supports only RH systems.

Outside of VM support, people were doing the same with Debian (w/o UIs) for decade now. RH was always behind because their software management system (RPM) was always deficient and biased to corporate needs of RH itself rather than its customers.

[lordmorgul]

RPM is not deficient, the method of distributing a mirrored package database was (by nature of being deficient) until the development of YUM reached some maturity and thats in all practicality the only advantage Debian's packaging system has. The RHN method of distributing packages actually works quite well for RHEL and used to work well for RHL... by open sourcing it the code CAN support more systems than just RH so all people need to do is take advantage of the code and make it support their systems (why should Red Hat do that for them?). What RHN provides is not just a neat way to download and install packages on the machine (apt is nothing more), but an admin-level view of the machines being maintained and a way to coordinate that update process smartly. I'm aware of nothing but makeshift scripts that would do this for Debian on an enterprise scale.

[gtewallace]

"I'm aware of nothing but makeshift scripts that would do this for Debian on an enterprise scale."

www.netdirector.org

release notes: http://sourceforge.net/project/shownotes.php?release_id=609239