John Rymer has been doing research into middleware and application development for more than 25 years. I myself have also been active in the software industry since that time and I can identify with his view of low-code.
Rymer says that initially he was quite skeptical about low-code, partly because of the high marketing content, for which I cannot blame him. However, his interest grew about five years ago when he saw that low-code development produced concrete results. Companies that were not able to solve their problems quickly enough with traditional development tools, managed to deliver the software they needed so much faster with the use of low-code. And that is in line with what I have always said: low-code is a much more efficient tool to produce complex, tailor-made business software and achieve quick results.
More than just web applications
However, John Rymer also confirms that low-code is still associated with small web applications or with old code generators from the ’90s. Enterprise low-code platforms have now largely surpassed this level, even though there are still many low-code platforms that are particularly suitable for creating web applications, mobile apps, workflows and filling the gaps between other applications. According to John Rymer, however, much progress has been made in recent years to enable business critical core applications to achieve the benefits of low-code. This is a strong and growing trend.
From coding to modeling
I can confirm from my own experience, that even before the term low-code was first used in 2014, there were already companies involved in software modeling. The companies interviewed at the time by John Rymer, considered the following characteristics to be the most important:
- They liked the fact that they could experiment with applications at relatively low costs, as opposed to traditional development methods. Subsequently, they could expand their financial investment as the new applications delivered more value.
- The emergence of cloud computing accelerated the growth of low-code. This was an opportunity for companies to easily experiment with low-code development via web portals, sometimes even for free. It was a far more flexible model than the traditional development tools, which required companies to organize servers, install and configure software, only to actually start reaping the benefits six months later.
- However, they found that the main feature of low-code was the possibility to use declarative techniques. The revolutionary idea that complex programming code gave way to a visual environment to model functionality and workflows, with all kinds of automated technologies, running in the background, to produce functioning software.
However, looking beyond those general features of low-code, the vendors soon opted for different approaches with regard to the use of programming code. That was the start of no-code platforms, where the use of code has been completely eliminated and which are particularly helpful for end users to develop their own productivity tools. New platforms emerged that offered a mix of visual development and programming. And, finally, the ‘code behind’ approach, in which low-code is used for the basic development and which then allows you to delve deeper into the code to make modifications.
The breakthrough for low-code came in 2010, according to John Rymer, when more and more companies began to see the benefits of low-code. This was partly driven by the rapidly growing demand for new mobile applications and the possibility of delivering them via the cloud. However, in the general perception low-code did not seem to be suitable yet for core enterprise applications.
Platform or point solution
Fast forward to 2020 and the low-code landscape has practically exploded in response to a huge surge in demand. There are now over 250 players, from no-code to enterprise low-code, and from products with a specific industry focus to highly versatile platforms.
John Rymer rightly says that the selection of a low-code platform is a very strategic choice today. Who are going to use it for software development? The ‘citizen developers’ with limited technological knowledge or fully qualified programmers? And what is your own preference as an organization? If you want to keep control, it may be wise not to allow too many code modifications. And to what extent do you intend to create critical business applications with low-code?
The low-code landscape is currently divided into two categories:
- Low-code platforms for Application Development & Delivery professionals.
- Low-code platforms for business developers.
According to John Rymer, both categories are rapidly growing in popularity and in that sense there are no winners or losers. He also believes that the role of the ‘citizen developer’ should definitely not be underestimated. He even regards this role as crucial for the digital transformation of companies. It is important to facilitate this new way of developing software in a structured manner and promote a healthy interaction between business and IT.
I am not sure if we should allow people in the workplace to produce software on a large scale, but I totally agree to involve the business in a structured manner. This means that IT has to come down from its ivory tower.
From the very beginning, I was convinced that low-code should not be a point solution in the long term. An organization that wants to use low-code strategically, should choose an enterprise low-code platform. In this way, you ensure that your low-code applications are more than just the glue between other applications or serve as temporary disposable solutions. With an enterprise low-code platform you add structural value to your business and you avoid the gradual accumulation of an enormous amount of interconnected low-code applications within the organization. This is not what you want, because with such an application landscape you are slowly but surely creating the legacy of the future. On the other hand, with an enterprise low-code platform, legacy software will become a thing of the past.