Quinton Wall is a Sr. Product Marketing Manager for Integration at BEA
where he is responsible for articulating the strategic vision and direction
of the products such as WebLogic Integration and AquaLogic Integration.
Quinton, a certified Enterprise Architect and member of BEA's internal SOA
Practice prides himself with looking at technology and the business of IT 'a
bit different from the rest'. He firmly believes that tomorrow's problems
can not be solved by today's thinking; rather a fresh perspective is often
needed.
Originally from Australia, Quinton has been involved in the Java community
since early 1996 where a number of colleagues formed Australia's first Java
Users Group. Quinton now resides in Pacific Grove, California with his wife
and 5 year old daughter (and about 50 Barbies, My Little Ponies etc..).
My fingers were sore from typing yesterday so I wanted to try something a bit different. Sit back and watch a short overview of the need for a revolutionary solution to packaged application connectivity called BEA SmartConnect.
I have said it before and I will say it again, in my opinion Java is the cobol of the 90's! Yes thats pretty harsh words especially for someone who works for a product company that build their applications in Java but I believe it is true. So as the sting begins to subside from that statement let me explain the shades of gray in the statement and premise the conversation with the note that Java is still fundamentally important but in a different context.
I was in a session this morning presented by Yefim Natis, Gartner VP where he was speaking about the future of application platforms. During this session Yefim much more eloquently than my statement above how application servers as we know it are becoming a commodity and we will begin to see parallel product offerings that support more agile frameworks such as spring (just in case you didnt know WLS 10 already supports Spring!) and going forward there will be a delineation between lightweight java (Spring etc), JEE, and Java Extreme Processing. This is where we get into the meat of where my statement above starts to ring true. Cobol traditionally has been reliable, stable and around for years, Java is moving that way too. This is evidenced by the declining release schedules of J2EE and JEE. I was around when we got a new version every few months. JavaEE is now a commodity no doubt about it. Lightweight Java; Spring, Hibernate etc will continue to drive enterprise innovation but I believe even those technologies will soon be swallowed by dynamic languages such as Rudy and PHP and way cool frameworks like Flex and Apollo for what they are good at; agile web applications.
Now there is the rub. Java is great at certain things which are especially critical to higher transaction and concurrent applications which happen to be BEA's traditional sweetspot. With the evolving SaaS environment growing rapidly languages such as Java will likely play a heavy lifting role behind the scenes and will continue to use products like WebLogic Integration, ALSB etc but users will be sheltered from the complexity of using them (IMO through Rudy, PHP, AJAX and Web2.0 related technologies).
Sounds kind of similar to cobol still running on mainframes and new innovation being done in more modern technologies doesn't it?
Perhaps the Gartner conference has me thinking of what technologies and circumstances have brought us to where were are now or perhaps it is the late hour but I got to thinking if I had an IT crystal ball that could not only look into the future but allow me to go back into the past and change something technology related what would it be?
Perhaps my thoughts will be biased in the fact that I no longer am responsible for maintaining applications and as a result may not look to change all those systems I implemented with EJB1 or AWT (remember that fun little toolkit before Swing) or something similar but one area that kept coming to mind was how pervasive the use of databases were (and still are) and the fact that these often resulted in the most complex areas that required continual change management to keep them in sync.
I think perhaps I would change how I used databases and instead I would have looked more at differential development and maybe even Jini (Yes I am still bummed about this one never taking off!)....but I have so many things I would change I am sure these are just the tip of the iceberg.
In regards to what I would like to see what the future holds for technology would be to see where the convergence of social computing and mobile computing is going to take us and what opportunities are just beginning to germinate today.
So many possibilities I really can not just pick one. What about you?
Day 1 - Gartner Application Architecture Development & Integration Summit
I am in Las Vegas attending the Gartner Application Architecture, Development and Integration Summit and it seems that there are more and more acronyms and buzz terms than every before...most good, some bad but all very interesting. Here is but a few:
WOA - Web Orientated Architecture
APAAS - Application Platform as a Service (ala Salesforce.com)
XAAS - Everything As A Service
Yes its kind of terminology overload here but in the midst of it all there were some great nuggets of information in particular around the true opportunities for innovation where there is an intersection of technologies such as Cloud Computing and Social Computing. With so much changing and the fact that no one has a blank slate the need to leverage existing assets to align your environment to mirror the business and adapt seems to be pushing us into, as Gartner calls it, the Zone of Competitive Leadership.
So what does this mean to your daily demands of working with legacy (or the new politically correct term, Heritage)? It means thinking of applications differently with as much design and composition as close to run time as possible and understanding that you should now start looking at applications as either bought or an endless combination of mashups the landscape looks decidedly different, according to Gartner
I had to put my brakes on here for a moment.... legacy and run time composition in the same sentence? Something smells odd here but the more I thought about it the trick comes down to granularity. If I have a tool such as WLI that is designed to integrate systems in a highly performant and transactional way then I combine it with the runtime composition ability of ALSB I get a mashup of functionality (in a product BEA calls AquaLogic Integrator!) that lets me control the granularity of processes and services exposed within my environment. I will always need some heavy lifting and long running transactions but I want to make these atomic business activities at a level of granularity relevant to the business so they can achieve this run time composition into tools such as ALBPM, mashups or whatever.
I thought about this for a while and threw in a helping of Gartner research which stated that in 2008, organizations must be doing SOA just to keep up with the competition and came to the conclusion that yes I concur with this statement but many of us have been doing composition of some sort for a long time now, the difference is, as I mentioned at the start of this rant, that convergent opportunities need products, processes and people to take full advantage of them. Traditional software products just may not cut it, at least not without mashing up the old and new and creating something that addresses the new use case presented to us.
Perhaps the next trend we will see is convergent products like AquaLogic Integrator that address the convergent opportunities Gartner alluded to today. Who knows? Perhaps tomorrow will yield an answer or perhaps more questions. We will see...
