My technology radar

Here is where I track the technologies and techniques I work with. It is inspired by Embedded Artistry which is based on the original Thoughtworks radar. It's a way of organising tools in categories and noting the maturity or adoption level for each.

Though I use different categories: in addition to observing technology trends, I wanted to highlight some aspects of embedded software engineering that I find challenging in the focus areas section.

1. Languages & Libraries
2. Platforms
3. Tools
4. Techniques & Focus areas

Instead of including the flashy Javascript radar, I decided to keep it simple and use a text based approach. To further simplify it, I reduced the assessment scale to the following:

• Adopt: This technology is proven, mature and can be used in most cases
• Evaluate: These tools are ready to use and have been tested to various extent, but not as mature as the adopted ones
• Hold: These are the industry standard tools which I rather not use or only in special circumstances

Languages & Libraries

💡 This section contains programming languages, supporting libraries and occasionally complete software frameworks.

Adopt

• C
• C++
• Python
• Qt
• Catch
• Trompeloeil

Evaluate

• Rust

Hold

• WPE Webkit

Platforms

💡 The platforms sections includes hardware platforms, bootloaders, operating systems and cloud providers.

Adopt

• NXP i.MX6
• STM32F3
• U-boot
• Linux
• FreeRTOS
• Azure IoT Hub
• Azure IoT Edge
• Azure Sphere
• AWS IoT Core

Evaluate

• NXP i.MX8
• 🔺 GD32 RISCV
• Nordic nRF5x
• Zephyr
• ELBE
• AWS IoT Greengrass

Hold

• 🔻 NXP i.MX8X
• NXP i.MX23

Tools

💡 The tools section list supporting engineering tools: software/firmware build systems, sanitisers, formatters.

Adopt

• Yocto
• Buildroot
• FreeRTOS
• Make
• Clangd: clang-format, clang-tidy

Evaluate

• 🔺 ELBE
• Zephyr

Techniques & Focus areas

💡 This section contains engineering techniques and some focus areas which I think requires special care in embedded software engineering.

Adopt

• Lifecycle management
  • Manufacturing: Factory floor configuration
  • Manufacturing: Cloud configuration
  • Provisioning: Server assignment(multi-tenancy, geo-sharding)
  • Maintenance: Firmware/Configuration/Application update
  • Deprovisioning
• Security
  • Hardware root of trust: TPM, Fuse, PUF
  • Defense: Read-only flash, isolated cores
  • Trusted computing: ARM Trustzone
  • Dynamic compartments: Processes, containers
  • Authentication: Certificates, tokens
  • Monitoring: Agent-based, agentless
  • Renew: Firmware update
• Network communication
• Distributed computing
  • Business drivers: Security, Availability, Cost, Bandwidth, Latency
  • Computing layers: Sensor, Edge, Fog, Cloud
  • Network communication:
    • Models: Device-to-cloud, cloud-to-device
    • Protocols: Application layer, Physical layer
    • Payload: Compression, serialisation