I think it will be hard to gain traction in the much larger accounts where IT management departments are heavily focused on ITIL processes and supporting products including CMDB's. "Inventory" might be acceptable for smaller companies with very little experience of the standards in this domain but it will not pull its weight with people looking for federated and model based approaches to change and configuration management.

I believe that the code offered is more than two years old compared to the latest version shipped with the commercial (subscription based) distribution by Hyperic which sold off its source code before it did an around about turn (after gaining very little traction) and adopting open source like they invented it along with holding wii tennis tournaments and rewriting their code so that there product could actual scale beyond its initial limitations.

JBoss ON 1.0 never made any impact but this is not entirely to do with the dev or sales team that took over this project.

I am sure the offering might help kick start the development of extensions by JBoss/RedHat customers but I doubt it will have any impact elsewhere in the IT management world.

It will be interesting to see what is delivered by the JBoss ON team after having nearly two years to develop something on top of the initial source code donation that was quickly repackaged but the opening of the source code just before the 2.0 release and going back to Hyperic for assistance in alignment does bode well in my book - but I could be wrong as it has been off the radar for so long now.

William

"It'll be interesting to see if RHQ gets traction beyond RH/JBoss."

Don't gloss over Hyperic who is generating additional traction. For example SpringSource is a Hyperic partner who can leverage RHQ if they want:
http://www.hyperic.com/news/releases/12_12_2007springsource.html

RHQ from JBoss perspective is a step towards alignment of the JBoss ON codebase with the Hyperic codebase it was originally built on. This alignment can now happen in open...where it should be.

The fact that it's in open enables others to leverage the effort if they choose to do so.

While it is true that this code has diverged from the initial cooperation with Hyperic there have been great strides in capability and functionality from both sides. We think this collaboration is going to be a great way to combine our efforts and also to build a consensus and community for compatible management.

The integration of generically applicable configuration and software management capabilities as well as new enhancements in monitoring, grouping and event tracking makes for compelling technology. JON 2.0 will just be the first evidence of this progress.

Greg sorry to go straight to the point but do you honestly think a large customer with mature IT management practices (ITIL?) is going to consider adopting (or extending) a product purporting to be a broad IT management solution that has a pretty dismal change and configuration management story. There is no notion of an extensible configuration management model (hint: CI, Relations) never mind even a change management (hint: change requests, work orders, delta analysis,..) not to mention no event correlation with automatic root cause analysis offered by the big 4 admittedly across various products and product lines.

One last question when will Hyperic implement the interfaces and revised signatures in its own product line which themselves seem to be still a work in progress judging by the forum discussions.

William

We're not trying to replace mature products, just provide an open platform where functionality can be developed and hopefully integrated into processes managed by existing infrastructure. ITIL is not necessarily about a single central CMDB that integrates directly to every product. Federation and integration is important. But the infrastructure to build even a little bit of direct configuration management is so painful as to keep community efforts from making much broad progress. We'll give you some of the basics and make it easy to write some custom support with features like CIs, versioning and differencing. Also, don't mistake what we have today with where we may be headed. I hope people are able to utilize our capabilities to make their CMDB more accurate and that is just part of what we're trying to do. I do think that we won't have much trouble getting tighter integration to some of the open source products for example.

-Greg

Hi Greg I spent 3 years working on ITIl/CMDB products you do not have to tell me that "ITIL is not necessarily about a single central CMDB". A customer could still implement many of the ITIL processes without a CMDB (federated or not) but they would be terribly ineffective and inefficient. The JBoss ON team are not a customer so I assume you are planning on offering a solution that does actually make this effective and more efficient so naturally I assume you would tackle one of the bottlenecks - information management model and change management. Maybe JBoss DNA could help here?

"Also, don't mistake what we have today with where we may be headed." If you had said that 2-3 years ago maybe it would not be so funny. You are now starting to sound like Microsoft of yesteryear with such future promises.

I am not honestly starting to think that we (the industry) are going back in time with the push by some open source vendors to provide limited solutions that met some basic needs at a very affordable initial price (zero) with a blatant disregard for standardization across other vendor products. Leading a customer to the "promised land" and then taxing them as soon as they have settled (and realized that all is not well) so that you can actually re-build the land that has the same facilitates from where they could have come from or gone to. Wow you are Microsoft of yesteryear.