The announcement of the end of commercial versions of GlassFish has a lot of people pronouncing GlassFish, in general, dead. While GlassFish is definitely not dead, the open source version will live, there are definitely going to be some challenges.
This move closely follows IBM’s similar actions in slowly backing away from Apache Geronimo, the basis of IBM WebSphere Community Edition for which IBM discontinued commercial support in May this year. The two giant software companies seem to be following each other lock and step.
Who is to Blame?
While it’s tempting to wag our fingers at these two giants for their decisions to divest in any commercial interests in open source Java EE, we might want to save a few wags for ourselves.
If you are a GlassFish user, how would you compare Oracle’s contribution to GlassFish to your own contribution to GlassFish?
If you had any answer to that question, good for you. If you find yourself shocked and not knowing how to respond, you are probably in the majority. What this says to me is that we as an industry still do not fully understand Open Source.
Open Source Isn’t Free.
Anything involving humans is never free. Software costs money to both consume and create. There is both a cost of ownership and a cost of creation.
The concept of total cost of ownership of Open Source is something we as an industry are learning and accepting. We’re getting good at supporting ourselves. The concept that Open Source costs money to make, however, is almost completely unnoticed. We are not good at supporting the people and communities that create the software we use. Someone must be doing it or it wouldn’t exist, right?
Simple Economics
Using Open Source and contributing nothing in return is unwise. You are letting someone else decide your fate in blind faith that all people, except you, are supporting the creators of the software you use. It’s never safe to assume other people are doing the thing you are not doing. If you’re wrong, your cost savings go out the window.
Supporting your open source communities isn’t charity, it’s good business.
Even if a project appears to be doing fine without your support, it’s a safe bet that if the project creators did have your support the software you get would be better. Giving it further thought, even if the software is already “perfect”, you are enabling people you know to be smart to solve more problems for you at an attractive price. Win-win. Supporting your open source communities isn’t charity, it’s good business.
Industry Change
For as much as we throw money around in our companies, it’s shocking we do not think to direct some of it at the communities of Open Source software we use. There has to be a middle ground between paying extreme prices for proprietary software and nothing for Open Source software.
This is the lesson we need to learn as an industry over the next 10 years. We must find this balance.
Not even IBM or Oracle can pick up the bill for Open Source forever. All Open Source communities need your support.
This is something that really needed to be said, and broadcast widely throughout the industry. Congrats for getting it out there.
Jan
Well said.
+1 on this…It is definitely good business
I totally agree. We have wrong misconception of open source leading to the neglect of the community and the services they offer.
Well said.
A good post but perhaps I could throw in something to the discussion. Typically, a lot of open source projects do not originate in commercial aspects but more from a more pragmatic, practical one – the developer has a passion and starts coding. For those that want to understand this better, you only need to read Eric Raymonds Cathederal and the Bazaar.
What we are seeing here is just two examples of corporations who have tried to pollute a “pure” process (open source) with commercial motives (they both had reasons for implementing geronimo and glassfish) and ultimately this is the reason why it has failed as you quite rightly point out – no community, no contributions, no passion or developer backing.
I know many passionate GlassFish developers who could have had far better careers in Oracle working on WebLogic, but stuck with GlassFish out of a genuine love for open source and community. In terms of Geronimo, IBM didn’t come around till 2 years later. As a Geronimo co-founder and someone who benefitted from IBM deciding to support the project, I can say it was nice to not have to starve while working on open source for a change.
There are plenty of other examples out there. Jan and Greg of Jetty have worked on that software for almost 20 years. Many of them not so good. Greg Luck struggled with Ehcache for years before finally joining Terracotta where it became a small part of what they did. Then of course there is the famous Heartbleed and the four, overloaded developers of OpenSSL.
There is no immunity for the assumptions people make. Seems either you’re doing fine without their help and so they don’t, or you’re not doing well enough and there’s no point in helping. Whichever way you add it up, the constant seems to be people always doing the math so they are not in the equation.
We need to further understand and educate ourselves about open source. There’s no situation, big project or small, where using it and contributing nothing is wise.