Post 1 – The future of application architecture is to revisit the past.
I don’t have a specific article to point to regarding this topic, but spend 5 minutes researching new application trends and architectures and you’ll be inundated with topics like the cloud, docker, machine learning, microservices, and the list goes on an on.
All of these concepts are tangentially the same thing. They’re enhancements to a digitized manual business process. I think people tend to forget where they came from, so to speak. That neat application you’re making is most likely a fancy digitized version of a manual process. Don’t get me wrong, there are services that just can’t be done manually in the span of an average human life. However, the concept of what we’re doing with applications if fundamentally the same. We’re digitizing the business process. And now we’re spending all this time enhancing the previous digitized versions to make a more efficient digitized version. Knock knock, it’s the law of diminishing returns!
Has anyone actually thought to stop and look around? Is the cloud going to make you more money? Perhaps it reduces development time and allows you to ship code faster or load balance more efficiently to take on more visitors during a surge. So perhaps it can make you more money. But are those “easy wins” compared to significant gains?
If you did stop and look around, I think you’d find a shift in comparison to the yesteryear of how businesses are run. 50 years ago the accountants ruled the world of business. A new idea came about and the accountants determined if the numbers worked. Marketing, sales, everyone went through accounting. That made business sense.
Now everything goes through technology in some way, much like they used to do with accounting. However, rather than focusing on the business side, they’re focused on discussions of technical feasibility or whether or not something will show a single digit improvement in speed. Business opportunities are measured in terms of “can we do this with our existing technology or do we have to upgrade?” Why didn’t you stop to think about potential business ventures BEFORE you built your stack? Or why are you waiting till now to evaluate a business venture that you’re probably too late for anyway.
I think the future of application architecture is a much more dynamic set of rules for how applications are built. Building for today is easy. Heck, building for tomorrow is not that hard either. But building for next month, regardless of what happens, is the future. If you can figure a way to do that, you can write your own check.
Topic 2 – Agile for the business
The concept of agile development isn’t new, but the adoption hasn’t permeated every corner of the business world either. Agile development makes a lot of sense to people that have experience with it. You provide iterative changes to a product or service along with the consumer who provides feedback along the entire process. This makes it nearly impossible to ship something the customer or consumer didn’t want.
What if we took this iterative feedback loop and adopted it toward the business itself? Instead of the business coming up with iterative changes to their existing business processes, the iterative changes were to the business strategy. What if one day you were working on an application that did A, but the business was interested in testing out a completely different method altogether?
That idea is not crazy, it happens all the time in the startup world. Something called the MVP or minimum viable product is built to test a business theory. For example, will people buy shoes online? Before Zappos, no one knew. So Nick Swinmurn got this idea of putting shoes online. Did he pay for an ecomerce website to be built and line up suppliers and customer service representatives and a sales team? What about inventory? Did he have to research which shoes sold the best and stock up? What about sizes? What if people buy his online shoes. Did any of this happen?
Nope. Nick built an MVP. He went to a local shoe store and asked if he could take pictures of the shoes in the store, which he would then post online. If a shoe sold, Nick would buy the shoe from the store at full price and ship the shoe to the customer.
Nick simultaneously built a business before it was even a vetted concept. What if the same happened within an existing company? One morning the business partners approach IT and tell them they’d like to build an MVP. In fact, they’d like to start building lots of MVP’s. Some will work out some will fail and die. Once everyone stopped laughing, it might interesting to talk it through.
I think the idea would be to start trying MVPs with the existing architecture and application stack. But the ultimate future would be to provide a playground of business opportunities where a business person’s dreams are built in small chucks to test out and vetted for true business opportunity.
In today’s world, most application architects would laugh you out of the room citing the impossibility of building such a platform. But everything’s impossible until it happens.
Topic 3 – Applying the concept of SOA to Data
The sort of thinking that led to flexibly integrated SOA solutions should now be applied to data. Get rid of that single schema, concentrate on having data served up in a way that matches the requirements of the business domains and concentrate governance on where its required to give global consistency and drive business collaboration. (Jones, 2013)
I really like thinking about using big data to drive business decisions and strategies automatically. Who would ever thought that the stock exchange trading floor would be completely replaced by algorithms? Imagine what a “typically business” is going to look like in 20 years. Will we recognize it? It think many spend a lot of time trying to refine today’s processes. The real steps forward are made in paradigm shifts. Thinking of how to use SOA to improve your current process is a step backward. Using SOA to enhance a fully automated big data driven solution is what will drive far future strategic advantages.
Steve Jones, 2013. http://www.zdnet.com/article/time-for-service-oriented-thinking-in-the-big-data-space/