The mantra “think big, start small” has been around for some time. Nowadays, many people hasten to add “… and grow fast”, “… and fail fast” or similar well-meant extensions. And of course, there is the real challenge. Perhaps it’s the main reason why so many start-ups never take off. It is relatively easy to dream up a big design, sketch some idealistic future state, and to dominate the world you dreamt up. As long as details don’t matter, money is no issue and technology has no limitations, you’re good to go.
It almost certainly starts to get challenging when you have to think about a small implementation of your grand design. Perhaps a proof of concept, a pilot, or a trial. How are you going to scale down, and more importantly, scope down your idea while still creating a meaningful first step? After all, you want an outcome that can answer your most compelling question, namely: is my idea going to work in practice, do I need to modify it, or would it be better to look for another venture altogether? At the same time, you do not want to share your idea with too many people, at least while it is still in its infancy. You never know with whom your precious ideas are safe – or not.
Let’s suppose for a moment you’ve cracked it. Kudos for now. You have an idea with a reasonably small scope on a reasonably small scale and with a reasonably clear expected outcome. And you’re ready to disrupt. You’re gearing into start-up mode. Now you’ve three challenges on to tackle. Firstly, you need the money to feed your initiative. For a small project, this should be a small challenge. If not, you’d have to seriously rethink your small start. Secondly, you have to find the right people. People who you can trust, people who have the right skills, people who have the right spirit to build a great team from. Not seldom, those people are hard to find and in high demand. So, you will need a compelling story to convince them. This may very well be the first of the big hurdles you need to take along your route. The more successful you are in creating a great team, the more promising your initiative becomes.
Now why is all this start-up prose on a blog by Yenlo? Surprise! Your third challenge is to select the right digital transformation technology for your initiative. You want to move fast, so you want to use tried and proven solutions wherever you can. But you don’t want to pay big, and certainly not upfront. That’s why, at least traditionally, open-source technology has been an important driver behind many successful initiatives. No upfront investments needed whatsoever, relatively easy to find skilled people, and the inevitable changes you will face are less painful to embrace. After all, these open source products tend to implement open standards – doing a much better job than many established parties. A no-brainer. At least, so it used to be.
Nowadays, cloud platforms with cloud services may also be tempting. Here you also have little or no upfront costs, you start paying as you start growing, and skilled resources are aplenty. There is also reason for caution. If you fail to robustly secure your cloud resources yourself, chances are that uninvited guests may come in to hunt you. You can manage that, and you should. More of strategic importance; as soon as you start using those attractive high-value platform services, you lock yourself into a vendor platform. Choices are still limited; little standards apply and migrations may become extremely painful. So be cautious and do think your options through.
Say you need a key-value store, or document database – who doesn’t these days. Amazon DynamoDB is right there to cover your needs. And so is Windows Azure Tables. And, they differ a lot. Hence, don’t expect a smooth transition should you ever want to migrate from one to the other. On the other hand, should you opt for MongoDB or Cassandra instead, you have the upfront effort of setting it up yourself. But now you have the added benefit of portability. From now on, you can basically decide to process your data anywhere you like – including AWS, Azure, on premise or even a combination of these. And should a new, attractive alternative ever come along, chances are you can jump right on the bandwagon. Freedom unfortunately doesn’t come for free. So, weigh it against the additional efforts needed.
There are many similar technology trade-offs to consider. Obviously, there’s nothing fundamentally wrong with a cloud strategy as such. Elasticity is great, cheap storage and computing power is also great, cloud VMs are solid – as long as you’re aware that clustering is pretty much a prerequisite. And if you plan to use a relational database, a carefree cloud service is hard to beat. In fact, we use cloud services all the time ourselves, to create a solid, well-managed foundation for our integration platforms. However, on top of it, we always recommend investing in freedom to manoeuvre. Freedom to use the technologies and standards that most fit your needs today. Most likely, you will have changing needs over time, in ways you nor we can predict. That’s when your investment in freedom really will pay its dividend.
At Yenlo, we love open-source technology. Just like the cloud, it allows you to move fast, with little or no upfront investment. Moreover, there is no lock-in, so the freedom to manoeuvre is there. We’re especially fond of WSO2. They distribute their digital innovation platform under a very business-friendly Apache 2 licence, they truly embrace industry standards, and they offer a rich set of products, which are integrated by design.
WSO2 provides a couple of flagship products. Take for instance WSO2 Identity Server. This is a best in class product, providing the flexibility to authenticate according to all current standards, and at the same time working with a host of identity providers – be it public or private – while also providing its own user store manager. Have a look at the WSO2 store to get an idea on how rich your choice is. Now if you’re working with customer accounts in here, it is kind of important that nobody should take these ransom, right? That you remain in full control of those precious assets. After all, this is your first point of contact. Having the freedom to plan and do whatever interests you most in your account management practices can hardly be overstated.
Or take application integration. Sooner or later you will select a financial system, a CRM-system, a content management system an such. They may or may not run as a SaaS-service in the cloud. Now you need to bring them together in an integrated flow. You certainly wouldn’t be the first. Again, it is a strategic decision what provider you trust to run these flows. WSO2 offers a tried and proven Enterprise Service Bus, as part of the WSO2 Enterprise Integrator suite, that comes with an unrivalled price-performance ratio. And it includes the freedom to choose.
WSO2 Data Analytics Server is yet another great candidate. After all, you want to be in control of your platform. You will need information to be in control. Be it information derived from continuous real-time event stream analysis, perhaps coming to you through alerts, or continual reporting on trends and statistics, or an ad-hoc predictive analysis, Data Analytics is your friend. Unlike some alternative products, WSO2 Data Analytics Server does not shy away from your big-data use-case. In fact, it complements the Hadoop file system, or Cassandra and such.
These are just three of the flagship products WSO2 ships. They offer many more. And they are innovating fast. Not only to keep up, e.g. with new and evolving standards, but also to expand. Expand into new technologies. Internet of Things, for instance. Enterprise Mobility. Microservices. And there’s more to come. They’re committed to your future needs in a way you hardly find with any vendor.
You might be wondering how this relatively small company is able to work these wonders. That’s actually a really good question. If you’re think they’re probably really smart and hardworking people, you’re not that far from the truth. Actually, much of their smartness is in their core architecture. Other than most suppliers, they’ve succeeded in designing a holistic architecture for their integration platform, and creating a rich set of more than 300 components, called “features”, which can be deployed as OSGi bundles, and composed into products. These components are not simply designed for reuse, they are really designed for joint use.
There is a common set of features that together constitute the Carbon framework – implementing generic functions such as clustering, logging and user management. All products are composed on top of this generic layer, thereby providing a lot of consistency across the products. And because many products share more functionality, creating a new product is generally just adding a few components to the family, and harvesting everything else from the repository.
Figure 1: Generalized WSO2 product architecture
That’s why we’re convinced that WSO2 is a great strategic choice for a start-up. It’s architecturally sound, it’s easy to set up, and it will grow as you go, wherever you go. So, you’re ready for any change, no matter how transformative. Please get in touch, if you want to know more.
By the way, WSO2 once was a start-up too. They have developed their award-winning platform through innovation, not through mergers and acquisitions. And they’ve built on top of (and are actively contributing to) the wealth of Apache frameworks. That’s still their winning strategy. Perhaps an idea for you too?