Lately, I've somehow managed to turn the concept of using an external keyboard with one's handset into a quasi-political, emotionally-charged issue. Corresponding with this doctrine was a specific narrative regarding the iPad Pro's introduction, symbolic of iPadOS' split from iOS, taking virtually all public-facing documentation of Bluetooth keyboard support with it. I even went so far as to describe it as a "class issue." In my defense, I was bitching under the assumption that greater attention from Apple, Inc. devoted to hardware keyboard support on the iPad would be to its benefit. Compared to the current state of said support on iPhone, the truth is... much more complicated. We'll get there, eventually.
My boss and mother has run her private therapy practice entirely on her iPhone and MacBook Pro for years, now, but the latter - a 15-inch, mid-2015 model - saw its last supported named macOS update in Monterey. Though her example has stood up incredibly well considering it's been used daily for 7 years, it was/is approaching definite in-need-of-replacement status. In what I believed to be my wisdom, I suggested that she take a chance and buy a top-of-the-line iPad Pro instead of another MacBook, mainly because I felt the design priorities of the latter - centering around thinness and sheer computing grunt - had long since departed relevance in her day-to-day computing life, and that the money would be better spent on something "different." I went so far as to suggest she buy the latest, largest, second-highest-capacity iPad Pro model1 with basically everything aside from cell connectivity.2
In further, probably unnecessary defense of this suggestion, I'll note that it was arrived upon with great care. A very substantial consideration behind the insistence involved my readership/membership of MacStories and their Club MacStories program, specifically. Federico Viticci, founder of MacStories, famously, has been using an iPad as his primary computer for going on a whole decade, by now.
 Known as "iPad13,9."  Which would prove far more limiting than I originally suspected, certainly.