Low-Code programming, or even No-Code programming: buzzwords of the hour that you encounter at every turn. Instagram ads are wooing programmers with no coding skills, cloud providers like Amazon and Microsoft are making entire low-code building systems available, and industry giants like Siemens are investing huge sums in platforms like Mendix. Looking at the automotive industry, it can be said without a doubt that the sector is on the move, beyond the obvious and less elegant pun, accelerating even: autonomous driving, increasingly intelligent and fully networked systems, Cyber security regulations throughout the life of the vehicle: They ensure that new solutions have to be found. For ever more modern customers with ever higher requirements within the framework of holistic concepts, in a competitive environment that is increasingly characterized by fast, agile competitors who are pushing out of the pure IT sector into established business areas such as the automotive industry - and who sometimes put the legacy systems of the OEM top dogs in a tight spot for explanation.
Coupled with an increasing shortage of skilled workers in almost every industry, the question arises as to what means future-oriented manufacturers and suppliers will take to avoid being overrun by the - if I may quote - turn of the millennium. Low-code and no-code are emerging alternatives whose advantages and disadvantages this article addresses.
Ca. 12 min
Why do we need low-code in industry?
There is no question that the automotive market is accelerating. Former Silicon Valley players are entering the industry and developments are picking up speed, especially looking at the autonomous driving market. How do both small and larger companies keep up with the status quo of development, how do they work more efficiently and faster, closer to the product and the need? Classic project-based development can rarely keep pace here. Low-code development offers a solution approach to develop required applications quickly and to be able to present first references. Often, this is not even just a question of being able to act quickly or having staying power: legacy systems from established OEMs simply become outdated, fall out of time and need to be replaced – and promptly.
The marketing of many products, not only in the automotive sector, also speaks a language that promises simplicity: everything goes plug-and-play, no setup necessary, everything works from the scratch and without effort. This is also reflected in development: code practically writes itself, building blocks are simply dragged and dropped into the right place in graphical development interfaces – and é voila, as if by magic, a new app is created, practically by the housewife next door while she was cooking potatoes. All right, we’ll anticipate what comparable articles like to save for last: We’re not quite there yet. Low-code and no-code development (whereby low-code dominates the still young growth market to date and complete no-code solutions are still the exception, at least in the professional sector) naturally simplify things. Of course, industrial customers still require specific applications and solutions tailored specifically to the automotive industry, for example, which cannot be fully implemented with low-code application development. The popular and often quoted Pareto principle is a good indication of what is to come: Up to 80 percent of the basic development is conceivable to be implemented via low-code development even by less specialized developers. However, the remaining twenty percent of development will continue to account for eighty percent of the effort, as these must be programmed specifically for the customer. Not to speak of the still existing and often very individual enterprise systems of large system customers
Low-Code / No-Code – Mendix and the Lock-In Effect
While the providers in the LinkedIn feed that promise to train developers completely without code knowledge are likely to suffer from a lack of seriousness, there are very much serious platforms like Mendix, in which industry players like Siemens and Cognizant have already invested early on.
Low-code platforms like Mendix have gained popularity in recent years because they allow companies to quickly and effectively develop applications without requiring deep technical knowledge.
Mendix offers a wide range of features and tools that enable developers to quickly build and deploy applications using pre-built low-code building blocks and built-in functionality, based on a visual interface that can be operated via drag and drop. The platform is particularly useful for companies that need to respond quickly to changes in their business environment, as it allows them to quickly make adjustments and add new features.
Mendix is also capable of developing applications for multiple platforms, including web, mobile and cloud. This makes it a versatile platform suitable for both internal business application development and customer solution delivery. Thanks to the option of data integration, applications can also be developed for existing systems – a major advantage for classic automotive OEMs, for example, which often have legacy enterprise systems that are difficult to replace.
Another advantage of Mendix is that it has an active and engaged community that helps developers use the platform and connects them with other users. This can help increase the success of projects and solve problems faster.
Overall, Mendix provides a powerful and easy-to-use platform for rapid application development that can benefit companies of all sizes and industries.
Especially in the field of automation, it is possible to profit from such building blocks as those offered by Mendix, and where standards prevail, low-code development celebrates triumphs. Whether hundreds of processes, huge ticket systems or dm general cloud sysadmin area: Here – in contrast to the often still somewhat undifferentiated automotive industry – real use cases exist that can be addressed in a solution-oriented manner.
To be sure: Mendix and Co. do not yet offer solutions for all scenarios, and not all platforms have only the customer’s best interests in mind. Where AWS and Azure offer their own low-code platforms such as Logic Apps or Lambda Codes, these exacerbate the already latent danger of a lock-in: Anyone who relies entirely on the easy-to-use low-code building blocks from Amazon, Microsoft and others (or the frameworks from Qualcomm and others) will quickly find themselves in trouble. Simplicity also has its dangers, which users of low-code and no-code development should be aware of.
Nevertheless, profit can be made from the current low-code situation. Vendors like Cognizant are already using many applications within the Mendix platform, which can serve as a basic setup for many customers, and developing entire frameworks and systematics. Use cases are conceivable in the aftersales area, in sustainability development, in enabling quality, and for mobile, web and other platforms.
How to integrate applications programmed via low-code development into large enterprise systems is a case-by-case challenge that can sometimes be simple and sometimes nearly impossible. It is not yet state-of-the-art to work with low-code solutions. However, eighty percent of the development effort can be greatly simplified, which of course also means lower costs for clients.
However, certain risks also need to be taken into account with regard to the use of low-code applications: In the long term, low-code and no-code approaches and the associated heavy use of prefabricated solution modules could well result in the loss of programming skills and basic IT knowledge. If, for example, along IT training programs and in the university environment, instead of the previous, open-source programming languages, greater reliance is placed on non-free modules of the various hyperscalers (AWS, Microsoft, etc.), the lock-in effect described above can, in extreme cases, extend to entire cohorts of developer graduates. Admittedly, this is an extreme case, but it makes clear to what extent local value creation would be affected by these effects and how difficult it would be to break such a lock-in again without available, basic programming knowledge in the free programming languages. This brings us seamlessly to the next question:
Will developers be unemployed in 10 years thanks to low-code development?
Another cliffhanger that would have made an excellent end to the article, but here’s the answer: No, developers will still have work in 10 years, and thanks to ongoing hyper-individualization everywhere, even more than ever.
Let’s look at it like the comparison of fast food versus haute cuisine: A system caterer prepares a burger and already has all the ingredients for it ready. Just assemble and prepare, done. Compared to upscale cuisine, where these ingredients are also available but are first prepared individually, this is simpler – but not a final solution for all tastes, and: Someone had to make the building blocks first. So, unlike the famous “last mile” concept of many service providers, it is the “first mile” that can neither be automated nor implemented with insufficiently trained specialists. This continues to be the area where the highest development effort is incurred, which generates the most costs – but also provides an opportunity for core competence that will remain a unique selling point.
Of course, it must also be said: What the future holds is uncertain. Ten years ago, even five years ago, what is possible today in terms of low-code and no-code solutions was hardly conceivable. Visual code apps, drag and drop, no-code developer academies. And time passes faster, progress becomes more rapid. So who can say with certainty what will be the standard in development in 2030?
One thing is certain, however: new technologies are now more in focus than ever, not only for providers, but also for young developers. Those who start working on the topic today want to work with new technology stacks that will still be around tomorrow. The “Evolution of Programming” is now at the stage of developing languages, the next (fifth) generation will be modeling: so experts will always be needed, and vendors that are adaptive and flexible will always be needed. However, the fact that low-code development could take over up to 80 percent of development capacities seems conceivable according to current findings. Platforms like Mendix and providers like Cognizant (Mobility) are already setting themselves up for this: Frontend and fullstack development, data logic, integration, augmented reality applications, and frameworks for the latest technology stacks: Already possible and under development. The evolution of development, who may forgive the clumsy pun to conclude.