It wasn't the grammar :-) it was that the author considered the 10K resistor to ground as "not strictly necessary". It is critical if the microprocessor pin can float (tri-state). Without it, and a floating gate, not only will the FET eventually turn on, it can be damaged by excess charge build up.
Not just excessive charge up it could be possible for it to be stuck in its linear region and dissipate a lot of power as it's spuriously turning on because there's a charge on the gate. I've managed to do that one on accident with a bad motor driving circuit. Though it was also largely caused by coupling on the breadboard.
nit: It's not wrong to assert that the microcontroller software will prevent the output pin from ever floating. And the gate's voltage will be constrained between Vdd and Vss due to the output pin's likely body diodes. Is it worth it to save a penny part, especially in a hobby/prototype design? No.
Also if you throw away all your BJTs, what do you use to drive the gate capacitance of your fancy FETs?
