I am fairly confident that the best days of the linux desktop are ahead of us. 3% might become 30% in a few years. Technology adoption takes strange paths but

I am fairly confident that the best days of the linux desktop are ahead of us. 3% might become 30% in a few years. Technology adoption takes strange paths but the history of linux is one of remarkable resilience and capability which augurs well for the future. My guess is that a key driver of growth will come from power users that will make best use of local AI and data science tools, preserving privacy and commercial secrecy. While many many casual users will continue using thin clients, effectively outsourcing their digital lives to the "cloud", this is unlikely to continue being the 97%. The opportunity is, e.g., for all the apps of the linux desktop to start making use of the python ML tools to empower users with unprecedented functionality. There is no reason that all of that new universe must be ceded to a small number of actors. This opportunity was there for a long time now, ML is not new, but the broader FOSS community is quite reactive. Emulating proprietary / closed tools to a fault (eg the libreoffice suite) rather than exploring the unique opportunities of alternative paths has been the norm. Luckily the AI hype will work as a kick in the butt. The future is for linux to lose.