John Reynolds's Blog

Java EE app servers: Why pay for support?

Apparently this blog entry was perceived as a threat to JBoss Inc. and got the attention of JBoss Chief Architect Bill Burke.

Bill made some really good points, but he also offered the following:

"John made some good points, but I think he is a little naive and has failed to grasp and factor in open source projects that have mass adoption."

Read in context, Bill's response also implies that I am clueless.

This reaction is understandable... but it's also a bit on the rude side.

Bill obviously needs to take note of my Rudeness Checker^TM project ;-)

With Sun's decision to "Open Source" their Java EE app server, it's likely that all Java EE app servers will soon be free. JBoss, JOnAS, Geronimo and Sun's GlassFish are going to exert huge pressure on the holdouts IBM, BEA and Oracle. IBM and BEA are already formulating responses, as is witnessed by the announcements of IBM support and BEA support for Geronimo, but it's hard to believe that anything short of total capitulation to FOSS is going to keep their app servers viable.

When all Java EE app servers are free, many companies will turn to selling support and maintenance contracts, but will anyone pay for ongoing support?

Comments

• Most of the time when I was working as JBoss Consultant was not to support the companies' application server but to help them designing their applications or find problems in their application. In addition from time to time I had to find a bug in the code of JBoss and then fix it.
  So I think that there is still enough money to be made through application servers but that it is rather education, consulting and bug fixing than plain support.
  -Andy

  Posted by: schaefa on July 05, 2005 at 04:22 PM

• John, your thoughts mirror mine. Many open source zealots say OS won't destroy the industry because in their worldview people will just shift from being paid for development to being paid for support but as you state support just doesn't sell (at least not in the long term). And once all software is free it'll be cheaper to just move to a competing product (though in the world of the OS zealots that's not needed because all OS software is perfect by definition) than to pay someone to help you out with the (non-existent, see above) imperfections with the OS software you're using. My prediction is that things go one of either two ways: either there's no working revenue model in software in the long term or companies will revert from off the shelf (possibly open source) products (which are unsupported because there's no revenue in support) back to using custom created software and as a result a new market for custom development will flourish where it's been more or less drained over the last several years as more and more effort has been shifted towards "integrated solutions" based on OS or commercially available systems. The two aren't mutually exclusive of course, the second situation likely following the first as companies find they can no longer find the support they need to implement their integrated systems and are forced to go it on their own.

  Posted by: jwenting on July 06, 2005 at 12:35 AM

• I will not pay for support - we're a small ISV and even when developing for commercial platforms, we usually resort to trials or developer programs. On the other hand, once stuff goes to production, most of our clients(*) will demand a contract support for every piece of the system. In fact, support is usually included in early project budgets and for the ISV it's a very good move because we don't want to be accountable for mysterious crashes, security bugs, or anything that's not a defect in our own code but rather in some opensource server or component.

  (*) granted, medium-to-large size companies. Smallish clients won't likely pay for an appserver, but they won't also pay for their OS, database, etc. so in these cases the final architecture is something like Linux+MySQL+Tomcat/JBoss/whatever without a single cent being paid for the software infrastructure, either in license fees or in support.

  Posted by: opinali on July 06, 2005 at 06:26 AM

• Reply from John to Opinali:

  The company that I work for is one of those clients who up until now has demanded contract support for every piece of the system. The fact that an extremely risk-averse company (like mine) is even considering dropping support for its app servers is a huge change from the standard operating procedures of the past.

  It will take time, but I think the writing is on the wall. The medium-to-large companies are learning what the smallish companies already know.

  FOSS is shaking up everything.

  Posted by: johnreynolds on July 06, 2005 at 07:53 AM

• Nice posting. I think there will be plenty of support revenue out there. I think why this is a tough issue is because support is a business problem, not a technology problem. There are all kinds of viable service businesses out there that I don't understand the need for. Will there be support revenue for FOSS stuff? Certainly. Will the code committers be able to figure out a target market, advertise, close deals, and manage an appropriate cost structure? I doubt it. Perhaps most commercial apps won't need much support. But some will. That support may be application support in the context of the application server instead of supporting the server itself. It may resemble consulting work. Who knows? But you are correct. If the people behind a piece of FOSS believe that support is just this obvious thing you start doing to make money, no way. It will be much harder than that.

  Posted by: edfactor on July 06, 2005 at 11:29 AM

• I would argue that you're perhaps looking at support in the wrong way. Support can also be viewed as a form of insurance. You probably don't complain about the fact you haven't used your car or home-owner insurance in the past year. I'm not saying that all support organizations are truly doing enough value-add to justify their prices, but one benefit of that contract is that it increases the odds of getting the proper person to actually fix the issue. At the very least that piece of paper gives your non-technical manager and/or CFO somebody else to blaim if the site goes down (it's called spreading the risk). Like many things in life - it's socio-political in nature - not technically driven.

  Posted by: mwilcox on July 07, 2005 at 07:03 AM

• Excellent observation mwilcox:

  "that piece of paper gives your non-technical manager and/or CFO somebody else to blaim if the site goes down "

  As the CFO's of the world become more tech savvy... and they are becoming more tech savvy... I think that they will be more astute in the insurance that they buy. Insurance is a good thing if you are buying the right insurance.

  Often, you don't realize that you bought the wrong insurance until you need to file a claim. Insurance needs change over time, so you need to review your insurance portfolio periodically.

  In the case of support for a Java EE app server... This is no longer a bleeding edge technology. App servers have proven to be stable and reliable. Perhaps we still need insurance, but the premium should be lower given the low risk to the insurer.

  --John

  Posted by: johnreynolds on July 07, 2005 at 08:42 AM

This work is licensed under a Creative Commons License.