Bad designs around us that persist

Driver blind spot

I used to ride a motorcycle and was trained to glance over my shoulder when overtaking, it’s called the lifesaver for obvious reasons. Car drivers have a similar issue called the blind spot. Blind spot, really? There is a simple cure for car drivers with split side-mirrors that mostly cover the long distance but have a patch that covers the near side, eliminating the blind spot. It’s what one would expect the automotive standards bodies to regulate as a must have, why they haven’t done so is a mystery. The extra cost is negligible. In my lifetime as a driver, I’ve had about two or three near misses when I neglected to glance to the side, mercifully swerving back at the last second. Multiply that across all drivers. A bad design that persists and is easily cured.

Computer keyboards

To avoid repetitive strain injuries, the correct angle for the wrist at the keyboard is horizontal with the arm. Bending the wrist upwards creates a strain. So, what do most keyboard manufacturers do? They design a wedge shape that is raised at the back, causing the wrist to bend upwards. Worse, many keyboards have legs that raise the back even more but thankfully can be collapsed back. A bad design that persists, but if you search you can find flat keyboards.

Software design

The software industry has its share of bad designs but it’s not due to developers being bad at their job, it’s that technology keeps evolving at such a rapid pace that a new and better way of achieving the same task is sure to appear. Often the better alternative would require learning a completely new skill set, and who has the time for that in the middle of a delivery deadline. Good enough to achieve the task is typically the norm.

And then came AI.? AI is disrupting the field of software development by providing valuable assistance, improving productivity and software quality to most developers who have tried these tools. It will be interesting to see whether next generation AI assisted software development tools can assist developers with the design phase, proving valuable guidance on optimum designs. Omdia has published a report on the topic, and there is more research in the pipeline.

Further Reading:

Omdia Market Radar on AI-assisted software development, OM029805 (Nov 2023)

PS I asked ChatGPT to create an image for this article, this was its response:

>create an image of AI assisted software development

ChatGPT

Creating an image of AI-assisted software development involves illustrating various aspects of how artificial intelligence can aid in the process of developing software. Key elements to include might be:

1. AI Algorithms and Tools: Visual representations of AI algorithms, perhaps symbolized by neural networks, code analysis tools, or automated testing tools.
2. Developers Collaborating with AI: Illustrate developers working alongside AI systems, showing a synergy between human creativity and AI efficiency.
3. Automated Code Generation: Icons or illustrations representing code being automatically generated or optimized by AI.
4. Debugging and Testing: AI assisting in identifying and fixing bugs, symbolized by a magnifying glass over code or gears representing automated testing.
5. Data Analytics: Show AI analyzing data, possibly with charts or graphs, to optimize performance or make data-driven decisions.
6. Cloud and DevOps: Represent cloud environments and DevOps pipelines where AI helps in deployment, monitoring, and scaling of applications.

I'll create an illustrative concept based on these ideas. Please hold on for a moment.

It seems I missed importing the required library for creating the gradient background. Let me correct that and generate the image again.

You can see the result at the top of the article.