pitchfork VS platform_system_core

Just use cmake-init which sets you up with a variation of the pitchfork layout.

Link broken, should be https://github.com/vector-of-bool/pitchfork

I use a variation of thispitchfork github layout. I have found it a very useful structure for my projects. As for file names, I always have only one class per file with the file name matching the class name exactly.

Take a look at the pitchfork standard - a lot of projects follow this convention today: https://github.com/vector-of-bool/pitchfork

Consider using the pitchfork layout convention

This is what I was directed to in regards to this question and what I use to organise structure. I was told it's a good starting point so not sure if there is something better out there. https://github.com/vector-of-bool/pitchfork

A while ago I stumbled upon this pitchfork project layout, But for simple projects I have just one CMakefile with the few sources in one project inside. If this grows, I use this and put it in a subfolder of the new project and add_subdirectory it...

As far as scientific studies go, I'm not sure if there are any that support the arguments of the linked proposal, but if you take a look at some of the most complex, large scale projects out there, you'll notice that they follow to a large extent the same structure as the one outlined in the above proposal. For instance, checkout the Google Chromium project, the AOSP projects, abseil, folly and many others. Of course, there will be exceptions so this is not a foolproof argument but I think it's interesting that projects of this scale have been designed this way.

after taking a look at the source code it's because the display field is an integer and it rounds down. that means any decimal value just gets floored to the next integer down.