Burkhard Stubert

Ports-and-Adapters Architecture: The Pattern

by Burkhard Stubert
2023/08/242023/08/24

The ports-and-adapters architecture should be the standard architecture for HMI applications. Its parts are loosely coupled, cohesive, easy to test and easy to extend. We can apply the reverse Conway manoeuvre to create self-dependent teams with minimal dependencies on other teams. I’ll motivate the ports-and-adapters or hexagonal architecture with USB ports and adapters and look at the architecture pattern from the production, testing and team perspective. I’ll apply the architecture to the HMI terminal of a harvester.

by Burkhard Stubert
2023/03/202023/03/20
3 Comments

We keep the source code of our applications and libraries in private repositories. Quite a few SoM and terminal makers do the same with their Yocto layers and their additional software. We typically use a Docker container to build the whole embedded Linux system with Yocto. Hence, we must be able to clone and update private repositories from the container.

by Burkhard Stubert
2023/02/042023/02/04
3 Comments

We can build the Qt versions from 6.2 to 6.5 with the meta-qt6 layer for the Yocto versions 3.1 to 4.1. This is a big time saver.

Using Qt 6 under LGPLv3

by Burkhard Stubert
2023/01/062023/01/06
18 Comments

The Qt Company changed Qt licensing in February 2022. All the separate commercial packages – including Qt for MCU, Qt Safe Renderer, Qt Automotive Suite and Qt Automation – were folded into two Qt for Device Creation licenses: Professional and Enterprise. The Qt Marketplace license for modules like CoAP, MQTT, Charts and for the design tool bridges was discontinued. My post helps you answer the crucial question: Shall you use Qt Commercial or Qt LGPL-3.0?

The Key Principles of Continuous Delivery

by Burkhard Stubert
2022/11/282022/11/28
2 Comments

In the seminal book Accelerate, Forsgren and her co-authors provide empirical evidence that Continuous Delivery has a positive impact on the performance of software development organisations. If organisations neglect some of the principles and practices of Continuous Delivery, their performance will suffer. They will reach the point where simple changes will take ages to implement. Not so with Continuous Delivery.

Creating A Custom Yocto Layer

by Burkhard Stubert
2022/11/072022/11/07

After having built the reference Linux image from a SoM, SoC or terminal maker and having run it on the board, we must inevitably custom-tailor this image to our needs. We must create our own Yocto layer. We must remove all the unnecessary packages and make our core application start automatically on power-up. Here is a step-by-step guide how to turn the application layer for a Toradex Verdin iMX8M Plus board into our own custom layer. The guide should also work for other boards.

Installing Linux Images on Toradex Verdin Boards

by Burkhard Stubert
2022/10/312022/10/31

In the post Setting Up Yocto Projects with kas, we built the Linux image for the Toradex Verdin iMX8M Plus. It’s time to flash the image on the board using the Toradex Easy Installer (TEZI). It’s a three-step procedure: wire up the board in a special way, install and run TEZI on the board, and flash our custom-built Linux image from a USB drive on the board.

Parking at Munich Airport: An Awful Experience

by Burkhard Stubert
2022/10/24

The parking experience at Munich Airport is awful. It is too easy to do something wrong, which can only be remedied by calling support. The bad user experience is caused by a bad system architecture. The pieces for a better architecture are already in place. Improving the interaction between these pieces improves the architecture and a fortiori the user experience. I can at least dream of a better parking experience in the future, although I can’t change the current one.

by Burkhard Stubert
2022/10/03

Not right away! Trunk-Based Development requires that the software builds and passes enough tests, before we integrate our changes into the main branch (a.k.a., trunk). We have enough tests, if breaking the software is highly unlikely. By definition, legacy code has no or not enough tests. Hence, we cannot apply trunk-based development right way, but should evolve our development process towards it.