You're not viewing the past with rose colored glasses. You're just viewing the past. We had simpler codecs with simpler computational complexities. Holding Shift and selecting a chunk of a video to copy was simple because videos were mostly a succession of independently compressed frames. Nowadays, we have forward- and backward- dependant frames, scene detection, and lots of other very advanced compression techniques.
There are whole projects striving to provide a reliable way to just cut videos without having to recode [1] and after years the results are mixed and only working for very specific codecs; no wonder Apple decided that doing the same, to their quality standards of the time, was not worth the effort or a secondary feature that was not in scope.
[1]: https://github.com/mifi/lossless-cut
