FYI, Savio - Hyperic does sell a proprietary version of Hyperic HQ - named, Hyperic HQ Enterprise. This version is built on the OSS version and contains more sophisticated features that provide greater automation for users at scale, and reporting. That said, our mission is to have a fully functional OSS product that absolutely gets the job done, and done well. However, for those managing at scale - they have additional complexity that includes both the physical environments they are managing as well as the business organization (this is why they need reporting beyond graphs and charts). This does encourage enterprise customers who are more likely to pay to engage with us. And, this in turn funds development just like you said. Its good for all the vendors out there.

Also, I think almost every OSS vendor - apart from maybe OpenNMS - has turned this way. I even heard something about Alfresco finally doing it late last year - although I couldn't say what was commercially licensed over there. I'll leave that to Matt.

Regardless, dual licenses work. And the OSS versions only help the vendor if they are complete and usable. Not crippleware. Otherwise - why do OSS at all?

Hi Savio,

Favor repaid. :-)

As far as this post goes, you make clear(er) points, but I still disagree with many of the points.

I think Matt and I are also arguing that if we assume that OSS is an agent of change for driving customer value. We actually feel you are stuck reiterating the legacy dogma of the license/maintenance model. We absolutely understand that the legacy model has "trained IT buyers to value software support and maintenance at ~15% of the initial price of the software product."...gated by steep initial license costs.

The traditional/legacy software model will still be applicable for many years....especially in software markets where OSS is not present as an agent of change.

Before JBoss, I worked at Princeton Softech (traditional database archiving software vendor acquired in Sep by IBM). I fully understand both models. For the markets I am focused on, an OSS approach makes a lot of sense....on many dimensions.

Finally, I agree with Stacey. It is important that OSS alternatives are complete and usable...not crippleware. Otherwise a mixed model will fail.

I don't understand your argument about software revenues and license and the requirement that we need to sell both service and licenses. It makes sense for early products in build-out mode when alternatives are scarce and one is recouping the initial development costs.

The problem is that the data shows a clear problem with software license sales in a commoditizing market. Look at the SG Cowen or Goldman Sachs studies that show (a) vendors in these markets making more from services and less from licenses (b) ~80% of license revenue going to feed the sales process

Can you explain your position that selling licenses is necessary in the long term with community-developed software? Or are you only referring to single-vendor controlled projects?

I have been following the ongoing debate and thinking very carefully about the validity of Savio's arguments. His core premise is based on "free riders". Savio argues that OSS free riders make up a very large percentage of those who use and derive value from open source software, yet contribute nothing back.

These free riders do not participate in the community and contribute to quality or innovation. They do not tell others about the software and contribute to awareness and distribution. They do not value support and will not pay for it. They have no aversion to using software under a copyleft license and may even cheat on its terms by not opening dependent code they create. They are shortsighted and narrow minded enough to fail to recognize that not paying for code or support and not contributing to the community deprives everyone, including themselves of continued and potential innovation of the software.

That is probably true.

Savio argues that the commercial software license model is better because it eliminates these free riders access to software.

It is true that the commercial software model protects against free riders, but not true that this is necessarily better - it is a matter of perspective.

If there is some very large demand for software, possibly demand that is currently being fulfilled by some companies offering commercially licensed software, and then some competitive company determines to enter this market by offering open source software - fully recognizing that it will disrupt the market, create free riders and reducing the total amount of revenue that can be extracted from the existing demand - and yet determining that this situation is of great value to them because a smaller slice of the existing pie is large enough to be appealing to them. A smaller market with a bunch of free riders in which the new entrant gets to participate is potentially better than a bigger market without free riders in which this potential new entrant cannot economically enter. Said another way, I would rather have a small pie than let you have a big pie which you will not share with me!

The willingness to shrink the software pie so that the new little guy can get a piece of it is good for software consumers and for this new competitor in the marke while it is clearly bad for the exiting market participants.

Open source software models may not be as effective as commercial license models at extracting money from potential free riders, but may still offer good and lucrative opportunities for new market entrants using commercial open source models - especially those entering, disrupting, and taking market share from commercial license competitors. This is the simply free market at work. Welcome Savio to the free market!

I grew up in the rust belt (outside Chicago) during the time when US manufacturing industry was being disrupted by less expensive overseas labor. It was painful for highly paid US manufacturing workers, but greatly benefited global consumers of manufactured products, not to mention willing workers in lesser developed countries and their families. Savio's lament about the disruptive economic effects sounds a lot like the lament of those displaced manufacturing workers and their unions.

Bil Miller, XAware, www.xaware.org