Sometimes when I am developing something, I want to do an experiment or implement a prototype which is based on an existing design. I’ve found that modifying an old design to accommodate a new idea can be difficult. Also, it tends to make it hard to compare the new version with the old version, which is a really useful thing.
For example, when refactoring a function, I think it’s better to write a new function. This allows you to test — or better, prove — that the two versions agree. If you’re refactoring a function that your code calls directly in many places, this approach allows you to introduce the new function gradually rather than everywhere all at once. If you’ve ever faced lots of compiler errors during refactoring, you know what I’m talking about.
I’ve reached a point in one of my projects where I think a real-time operating system (RTOS) would be useful. The program will have some mostly independent concurrent tasks, and it would be easier to let the scheduler … schedule … them, rather than me dictate how to break up the tasks and order them in time.
I’ll probably go with FreeRTOS! But adding it to the existing MCU application has been tough.
The parallel approach can help. I can add a separate target to my build system which will allow me to bring up a stub FreeRTOS system. After I’ve got FreeRTOS working on the microcontroller, I can migrate the application to the FreeRTOS system. Having two build targets will allow me to build the old and new executables simultaneously and easily compare them.
I recently designed some business cards for myself, to hand to people I see often but haven’t given my contact info yet.
After I got the prints, I noticed certain aspects of the design that could be improved. I could have used a smaller, sans-serif font. I could have drawn a circle with a thicker stroke and a cool color.
I changed the SVG file, and now have a design that might be better. Is it actually better? The old no longer exists. The parallel approach would make it easier to compare the two designs.
Well, I lied. I versioned my SVG file using Git, so I still have the old design. Moreover, I don’t have to manually check out the old version and copy it to a separate file. Git supports multiple worktrees, so I can do something like
git worktree add previous/ HEAD^
This will create a new directory previous/ that contains the original
design. When I’m done comparing, I can remove the worktree with
git worktree remove previous/
But this isn’t the best approach. I could have added the new design as a separate layer in the SVG file, allowing immediate comparison.
How might you apply the parallel approach?
Discuss this page by emailing my public inbox. Please note the etiquette guidelines.
© 2024 Karl Schultheisz