Integration: I’m sexy and I know it
Integration: I’m sexy and I know it
Join the DZone community and get the full member experience.Join For Free
Discover how you can get APIs and microservices to work at true enterprise scale.
I’m a technology guy who tries to keep up with every booming tech startup, acquisition by a major player, and new version of top-selling software. I’ve watched waves of technology crash on the shore of commerce and government.
Why am I telling you this?
The conversation we’re not having
I’m confident I’ve been left with a pretty good view of how it all fits together, but am continually surprised by how little we talk about the biggest challenge of yesterday, today and future: how rapidly increasing amounts of data is found, understood and used seamlessly. It feels like this problem, known by the un-sexy term ‘integration’ had its heyday in the late 90′s as client/server exploded and ‘middle tier’ and ‘EAI’ were hyper hot out of operational necessity. It stayed at the front through ‘SOA’ but then quietly took a back seat as the world turned its attention to mobile, social, analytics and cloud.
What happened? Did we simply get distracted? My gut tells me the problem not only didn’t go away and there can only be a couple of good reasons for the integration conversation going ‘dark’:
- Integrating systems and data isn’t the challenge it used to be — We only have to think of mobile platforms and their need to touch many systems incredibly quickly to know this one can’t be true. With so much change afoot in technology and globalization, it will only get worse, not better.
- The cost of integrating outweighs the value — This one is a deceptive argument. On a case-by-case basis, it would be easy to show that taking a step back to integrate could be more expensive than a quick fix. The problem is that those quick fixes quickly become ‘death by a thousand paper cuts.’ Quick fixes won’t scale and become complexity that’s unmanageable.
- Platforms (vendor ‘stacks’) solve the problem for us — Really? Anyone spending enough time with any major vendor knows that the hodgepodge of applications sold by the biggest players are rarely integrated.
- Software as a Service makes this someone else’s problem — Forget it. SaaS simply creates cheaper but far more silo’d applications. Only recently, with applications like Cloud Bus, are SaaS integration solutions becoming viable.
So, if none of these reasons hold up (they don’t), we have a problem that will only continue to get worse unless we address it. Business processes will break, data analytics will be inaccurate, customers will be unhappy and money will be lost.
Fight Club and Project Mayhem
In a great comparison to the movie Fight Club, Lance Speck in First Rule of Integration: Do Not Talk About Integration, says that trying to charge ahead while bypassing the foundational work to make technologies talk results in total chaos:
With the explosion of data and technology, and the rise of various IT infrastructures with millions of endpoint application systems, formats and databases, maintaining connectivity can be hell. Most endpoints are now incompatible with each other as developers continue to build on top of them, transforming them into unique applications – puzzle pieces that no longer fit together seamlessly. IT has to weave these systems together, but frankly, it’s struggling. That’s why you need integration, which at a basic level is a set of program systems designed to connect your disparate sources for a coherent view of your business and your customers.
His dire words are more accurate than anyone seems willing to talk about. Get most CIO’s alone for a few minutes and they’ll admit this, but they’ll also say in a world of declining technology budgets, they’re doing what they can and facing an impossible customer. But are they really?
A success story
While some companies fight for budget and point fingers at evil SaaS usage, others are taking steps to solve the bigger problem. One of my favorite examples is Mercy Health in St. Louis, Missouri. In a healthcare network operating at a 1 to 2% margin, they took the time and spent the money to proactively join up their systems, knowing that Big Data, analytics, and healthcare overhaul were the new reality.
A byproduct of their work was Mercy SafeWatch, an integration and business events solution they actively market in the expanding role of telehealth. SafeWatch, among other monitoring, helps solve one of the most pernicious of challenges…systemic infections (sepsis) that occur within the hospital and have a high mortality rate. They integrated and as a result, saved lives.
It doesn’t get better than that but it took organizational ‘guts’ to make the investment and stick it out. There need to be more gutsy decisions like Mercy’s before integration is seen as sexy by more than just technology people.
If you’re in the Paris area tomorrow or London on Thursday, join me at TIBCO Transform 2013 Conference. I’ll be happy to talk about this and other challenges facing business.
Opinions expressed by DZone contributors are their own.