Views on the Industry

why the cloud scares Oracle

It takes some effort to look below the surface and additionally look past the comedic routines of its CEO on stage, but i am sure there are more people in the industry that ask, besides me, why the purveyor of the most robust database appliance in the marketplace offers nothing towards the cloud-based movement of apps running on virtualized hardware: in other words, why is Oracle running hard from hosted infrastructure?....

Even considering security and Identity management webinars coming up, based around cloud-infrastructure and the little salesforce.com competitive hedge that Ellison is invested in: why would oracle not be in the driver's seat when it comes to offering a competitive cloud based offering around MySQL and Glassfish, along with the ESB, BPM, and tools, not to mention Amberpoint?....

That is the question i have for the roadmap determinants, when they finally release a clustering schedule for glassfish v. 3, and it is a question i have had for Sun for some time now, why is MySQL and Glassfish not reserved for scale-out deployments, when the core Fusion/Oracle DB can handle enterprise accounts?....

Is it resources, that keeping two competing application platforms within the same company would be too distracting, too costly, too cumbersome, too cannibalizing, or too threatening to the money makers of WebLogic and Oracle DB, to just build a cloud org., in the form of MySQL and Glassfish, and put to rest the roadmap questions that will come if Oracle offers anything other than full enterprise-wide, web-wide support for the Sun software assets?....

My only answer can come from the open source model of the Sun infrastructure product-lines that never had a chance in a hardware scenario, but could become true deal makers in a software sales force's hands, and that is Oracle refuses to abide by a competitive affront to their proprietary model of selling high license cost, and even higher maintenance cost products, that do not have competing organizations, whether they be channel, SI, or ISV to under-cut them on price....

They have spent a lot of money on Sun, for a dying company that was doing great things in OSS terms so they have to get their investment back, through more than just hardware margins, and that means pushing Fusion out the door with little to no impediments to competitive pricing, and honestly where does that pricing pressure come from: not from IBM and WebSphere with Global Services offerings, only Red Hat with JBoss stands in the way of WebLogic pricing, now that Glassfish is safely behind closed doors, or so it seems.....

So, I respect the Glassfish people at Oracle, and will listen to them to be patient, and wait for the roadmap, but it is pretty tough now to back-down from what was said two weeks ago with the web event to announce some details of the Glassfish and MySQL purposes within a new Oracle software organization, and wonder why Oracle continues to deride something that everyone else is planning on, in the form of cloud-scale deployments....

It takes integration of apps and data, and it takes open standards, in the form of web services, and honestly it takes Open Source Software in order for those things to happen, and i am not so convinced we are going to see an investment beyond Fusion for engineering resources to be applied to scale-out deployments, like what it would take in the form of Reference Architectures and the like in order for non-techy sales people to translate for customers....

OSS is not what Oracle likes to do, except unless it means hurting or at least attempting to hurt Red Hat's model for Enterprise Linux by doing their own fork job, and though Mike Lehmann, Thomas Kurian, and Ted Ferrell are saying that OSS will continue, can they promise or agree to definitive support for the open source communities that were built around java.net at Sun?...i dont think they can or are willing to do that, considering the immense under-taking of getting three world-class ERP systems on one app server platform, along with all of the ancillary products to accompany it, in the form of Fusion....

I wonder, and then stop to wonder why there is no cloud future for Glassfish and MySQL because honestly, that was the major threat to Oracle Fusion in the form of technology parity and pricing pressure before the merger, so what is the incentive now to continue them now that they dont have to?.....lots of questions still remain but it becomes clearer as to what Oracle will ultimately prioritize in the face of hard choices on resource allocation, that will come regardless of current promises, and irrespective of amount of money available beyond what Sun had....

The reality will be a hard lesson for anyone trusting altruism from a company specialized to turn a profit at the expense of what is best for its customers....thats not all i am saying, but it begs the question, Oracle: what is so bad about the cloud to your business?.....

posted by douglas dooley @ 9:05 AM 0 comments

"forking" Glassfish

i dont buy it, i just refuse to capitulate, and i dont have to as i am not an employee of the world's largest business software provider, and say that Glassfish is a 'departmental' application server, to be upsold to expensive WebLogic licenses, that frankly offers an inferior product to what has been built in the OSS community through the long-gone independent java.net....someone needs to explain to me the art of forking because the only company or organization that i have heard do it have been failures in the form of Apache's attempts with openJDK, and Oracle's attempt to copy Red hat's Enterprise Linux....but to me, it is time to explore the option of forking glassfish and continuing to invest in it as a platform for cloud-based, high availability, enterprise-wide and web-wide deployments, that scale beyond a developer's studio....

so, i am calling on those in the industry with the means to look at code, namely the Java developer ranks to get glassfish on to google's Code site, and begin the process of forking the OSS content that has been built over the past four years, since Glassfish's announcement and inception...there is no logical play for Glassfish within Oracle's stated product strategy with Fusion, and therefore there is a significant market opportunity to take the assets under GPL license and form a new entity that will support enterprise-wide deployments, and even build within Google's cloud offerings to make the Reference Implementation the premier app server in the marketplace....makes sense, doesnt it?....

I would say that the most natural person to do the rallying for this is Marc Fleury, himself, though i know it would be a tough sell, as this is not his code base baby, but JBoss has morphed in to something else beyond what he originally started, and perhaps his non-compete is coming up to expiration....just think of the developers he would attract worldwide, if he were to sign-on to a fork of Glassfish, and just think of the resources that could be potentially available from Google, if we were to make an app server cloud oriented, and on their Code site, its like a built-in acquisition guarantee, from the company with the deepest pockets....there is even a scenario where Google will need Glassfish to do business, as they move in to the enterprise, so with some luck and lots of execution, it could be a high, high premium paid for the work of a fork job....

I have also proposed portable applications across JEE servers, perhaps even components for vertical implementations for HIPAA, SWIFT, FMEA, and other industry standards, so the project could grow beyond just an app server platform fork, and include applications that would run on JBoss, and perhaps even WebLogic and WebSphere with some work....but the real test is in finding Glassfish a home, and getting the write-once-run-anywhere promise going with some apps that run in the cloud that can be deployed on compliant platforms from any vendor or company....forking Glassfish means taking on Oracle's stated product strategy, and it is about time someone does it, as Fusion toils along, with no word whatsoever on its delivery schedule....all they do is buy more companies to divert attention from their timelines for release, and now they have kicked the most valuable Sun asset in to the ranks of 'departmental': Fail.....

i ask you all in the industry to consider a world where JBoss is the only JEE 6 app server left for deployments, as WebLogic and WebSphere have basically not even supported JEE 5 yet, within their product-lines, so now we have a uni-tier app server market on standards, with Spring Source being the natural beneficiary of this splintering....it is time for the Java developer ranks to get paid for OSS work and not let companies like Oracle take all the work in-house and mis-market it, with a term that make me cringe: 'departmental'....

take a look around, what do you see in the marketplace other than a lot of invested companies looking for answers on what is next for JEE, and the answer had been coming from the Glassfish org., and now that org. is essentially dead, so forks dont usually work, they are seldom ever attempted, but this one is worth it, this one could introduce Google in to the enterprise and could make the developers on the project a lot of money that Oracle is leaving on the table with their positioning that basically says we will not sell Glassfish, and we will only sell WebLogic....i say to all those not satisfied with the decision, to rise up, and begin to build the ranks of those willing to fork glassfish for the industry to stay afloat of portability....it only takes some momentum to really build something unique, and only you are one of many who could make it possible, so take a look at my site, and tell me i am not on to something....

it is time to fork Glassfish:
http://code.google.com/p/astrocloud/wiki/PageName

posted by douglas dooley @ 8:49 AM 1 comments