Low-code and no-code development platforms have gained significant traction recently, and more so with the rise of artificial intelligence in 2023. This technology promises to democratize application development and empower “citizen developers.” 

If it sounds familiar, we tried this back in the ‘70s with Cobol and many times after that. Executives won’t write code no matter how easy it is.

Low-code and no-code platforms provide visual interfaces and prebuilt components to simplify the coding process so people with minimal coding experience can create applications quickly. Although these platforms offer advantages, they also introduce trade-offs that must be carefully considered within an exemplary cloud computing architecture, including design, development, and deployment.

Let’s go over these drawbacks and what will likely evolve during the next few years.

Flexibility versus customization

Low-code and no-code platforms excel at streamlining the development process by offering prepackaged components and templates. This is the same concept as using a template in your word processor, such as a generic thank-you note or resume. Today, we use our favorite generative AI platform to write them for us.

These platforms may have limitations when it comes to customization. As application complexity grows, developers may need help to achieve the customization and fine-grained control they desire. This can be a barrier for organizations with unique or highly specialized requirements. This is the same issue we had with enterprise resource planning (ERP) platforms in the ‘90s. We had to rewrite them using whatever customization tech the ERP provider offered to make them usable. Many companies found they could have just written the application themselves and saved 90% of the money.

Speed versus scalability

Low-code and no-code platforms enable rapid application development by abstracting away the complexities of coding. This is nothing new, but today we can do it much better with layers of AI to assist us.

This can be advantageous for organizations that need to prototype and launch applications quickly. However, scaling these applications may reveal the limitations of the low-code platform as demands increase. Suppose the platform is not built to handle large user bases or high data volumes, as most need to do. You’ll end up hitting a wall, and since you did not create the system in the first place, I’m not sure how easily you can fix things. 

Security and control

Low-code and no-code platforms are built to make development accessible to a broader audience. They often incorporate security features, but the level of control and granularity may be limited compared to traditional approaches where security should be part of overall development.

Organizations must carefully evaluate the security measures provided by the platform and ensure they align with their specific security requirements and industry regulations. I’ve not found a low-code or no-code system yet that’s able to address this. Many unwisely go without sufficient security for the convenience of leveraging this technology.

Integration with existing systems

Low-code and no-code platforms can simplify the development of stand-alone applications. However, integrating these applications with legacy systems or other cloud services may be a challenge. This largely depends on the platform’s capabilities and API integrations and may require additional development efforts to achieve seamless integration with existing systems. 

Much like the security trade-off we just mentioned, this lowers the value that low-code and no-code technology bring. We have to layer complex code into systems we really don’t understand because we did not develop them. A robot did.

Once again we have a technology that seems to be a game-changer for many enterprises. My concern is that low-code and no-code will cause more work and add more risk if you’re not very careful in how it’s used and applied. Sorry if I burst some bubbles.

Copyright © 2023 IDG Communications, Inc.



Source link

By i53gf