I suppose our fundamental difference is over whether these apps are best compared as tools or as works of art. When I evaluate and compare tools, in addition to cost, I'm looking at things like speed to perform key tasks, ease/speed to navigate the UI, reliability, error rate, user mistake-rate, latency, accuracy, feature discoverability. Does it effectively do what it says on the tin? And are these attributes greater than or less than those of competing tools? These are all things the developer can take as feedback, file as bugs, go back and improve, and re-measure.

If it were a video game or movie, then sure, subjective comparisons of the media content would make more sense. Does the plot make sense? Are the characters likable? But then, it's just one reviewer's opinion vs. another's rather than actionable measures of quality.

I'm looking at the article from the point of view of the developer accepting criticism and wanting to take corrective action. If I released a utility application and a reviewer told me [not your words, but for example] it felt cold and without pizazz, and it didn't zing and pop and it wasn't sleek enough, I wouldn't know what to do to fix this.