The symbol and the names both make more sense for point contact transistors than BJTs. Like the article says, I had never heard of point contact transistors' different operating principle.
FETs have a schematic symbol that describes their operating principle pretty well - they have never had a shift in meaning.
The names continued to make sense for the alloyed transistors (most of the germanium transistors were alloyed) and for some of the earlier diffused silicon power transistors, the so-called single-diffused transistors, of which the most well known was the RCA 2N3055, which was used by many ancient audio amplifiers.
In those alloyed or single-diffused transistors, the fabrication also started with a semiconductor crystal that was the base of the transistor, in which emitter and collector electrodes were created, but by alloying or by diffusion, instead of point contacts.
Only when the double diffused transistors, especially in their planar version invented at Fairchild, became the dominant kind of bipolar transistors (after 1960), the name "base" for the control terminal became no longer appropriate.
While "gate" was inherited by the FETs from the vacuum tubes and gas tubes, the names "source" and "drain" mean exactly the same thing as "emitter" and "collector", so there was no reason for introducing these 2 alternative terms.
FETs have a schematic symbol that describes their operating principle pretty well - they have never had a shift in meaning.