With everyone blogging about tech trends out there, we noticed that there weren't many talking about trends specific to business and business applications. We have worked with enterprises for over 16 years to solve the real-world challenges that real businesses face. After a very busy 2018 working with all sorts of business across industries, here are the top four business technology trends we noticed. Here's to a happy, healthy, successful 2019 for all!
By many accounts, 2018 was the year Kubernetes cemented its place as the container orchestration stack of choice, and we saw the same trend here at Fairway. Enabling microservices architecture was the primary driver for adoption, and managed Kubernetes solutions from all the major IaaS cloud providers was the enabler. Enterprises with more mature IT infrastructure and significant investment in on-premises hardware and software have started moving to a hybrid cloud and Kubernetes is generating a lot of interest for that configuration.
- David H. Principal Developer
It seems that not a day goes by now without some mention of machine learning or artificial intelligence in the mainstream media. Disembodied but seemingly smart voices emanate from our mobile phones and devices in our homes, responding to our spoken queries about the weather, traffic, or just about any other topic. These now-ubiquitous technologies are built on machine learning, and the underlying math and science that makes these seeming miraculous capabilities possible is now being commoditized by companies like Amazon, Microsoft, Google, and others. These companies are providing generic machine learning platforms that anyone can use to build new applications imbued with the ability to make sense of large quantities of messy, “real-world” data.
As awareness grows of the wide range of problems machine learning can be applied to, more and more companies are asking their technology partners how it can be used to solve their particular business problems, to help them run their businesses more efficiently, or to provide new and never before seen services to their customers. Fairway Technologies is already experiencing this trend with our clients, across a range of industries and businesses. An industrial irrigation system manufacturer wants to make sense of huge quantities of historic data to detect leaks and irrigation hardware that is on the verge of failing. This will allow them to offer a revolutionary service to proactively alert their customers about preventative maintenance and ultimately to save water and money. A government client wants to analyze historical inventory and supply data to better forecast demand and improve combat readiness. Their current process doesn’t take into account the rich and varied data that they know really drives supply requests, and they want to use a more sophisticated approach that brings that data into the decision cycle.
Both of these client scenarios have something in common: the client has access to a lot of historical data, and their ongoing operations, and those of their customers, continue to generate data. This is a key enabler to machine learning approaches to solving problems: lots of good data from which to learn, and more data coming in to keep learning. Not all businesses have this sort of data-rich operations, and machine learning is neither a panacea nor magic dust that can just be sprinkled onto a business problem to generate a valuable solution. At Fairway, we perform in-depth business analyses with our clients to assess if their particular scenario is ripe for a machine learning solution, and if the data available to them can support it. When a business problem and its underlying data can be brought together in a machine learning solution, amazing insights and capabilities can be created.
- Chris M. Business Solutions Architect
Cryptocurrency and Blockchain
One of the hottest topics in 2018, where real millionaires were made and unmade from what seemed to be play money. The tech that allows cryptos to exist is called blockchain. Blockchain sounds like some fancy New Age dance that millennials came up with, but it’s simply a chain of transactions, stored in chronological history, forever. Or a public, append-only, ledger of financial transactions. This might seem like insane amounts of data to store, which it is. That is why it’s so hard to encrypt or hash the blockchain, which is what has to happen after every transaction.
Many companies have looked into blockchain as a solution for different problems. While this can be a viable solution for some issues, it can also be heavy-handed. It takes an insane amount of power to hash even a small blockchain. Blockchains are generally a public ledger, meaning sensitive data needs to be obscured in some way. If the data doesn’t need to be publicly managed, or seen, then a solution like event sourcing might be a better option and also doesn’t have the security risks a blockchain might hold. However, banks are currently seeing great success with using blockchain for bank-to-bank equity transactions.
Sadly, in 2018 we saw startup after startup fail from trying to use blockchain in applications it’s not really designed for. We will continue to look for good use cases; just be careful to make sure it is the right tool for the problem you are facing before adopting it.
- John G. Software Developer
GraphQL is a client-server communication specification for querying and mutating data with implementations in many popular programming languages. It offers many technical advantages over RESTful APIs, helping developers create more flexible interfaces for clients to fetch and update data from servers. It can also help tame microservice architectures, as well as provide a cleaner interface to the RESTful architectures it will eventually make obsolete. GraphQL may not be a technology everyone is talking about just yet, but it saw a big jump in adoption last year. 2019 may be the year where the wider web development world becomes aware of those benefits, and asks whether they should implement a GraphQL solution instead of more REST services.
- Craig M. Principal Developer