That is ridiculous. The Power_Good signal is like a power-on reset that goes true within 0.1-0.5 second after the power supply voltages have stabilized, and it turns off if any of the voltages drops below its lower limit. This is what the ATX/ATX 12V Power Supply Design Guide ver 1.1 says:
"PWR_OK" is a "power good" signal. It should be asserted high by the power supply to indicate that the +12 VDC, +5VDC, and +3.3VDC outputs are above theundervoltage thresholds listed in Section 3.2.1 and that sufficient mains energy is stored by the converter to guarantee continuous power operation within specification for at least the duration specified in Section 3.2.11, "Voltage Hold-up Time." conversely, PWER_OK should be deasserted to a low state when any of the +12 VDC, +5 VDC, or +3.3 VDC output voltages falls below its undervoltage threshold, or when mains power has been removed for a time sufficiently long such that power supply operation cannot be guranteed beyond the power-down warning time. The electrical and timing characteristics of the PWR_OK signal are given in Table 14 and in Figure 2.
All this signal does is indicate the state of the power supply; it is not a security option (it's not an optional signal) that prevents PC turn-on if the motherboard connector is inserted incorrectly (very difficult, but it is possible to leave off the Aux and 12VATX connectors, which can lead to unreliable operation that will likely not be detected by the motherboard, except by showing unreliable operation).
The only purpose served by grounding the Power_Good outputs is to disable them. I believe that this can be done safely because they're open-collector outputs, but it makes no sense to ground them because if the motherboard sees Power_Good as low all the time, it won't ever turn on but stay in a reset state constantly. It may make sense to tie both Power_Good signals together so the motherboard doesn't start until both power supplies are ready and turn off if either one fails.
Please provide a translation of the text explaining the reason for shorting the Power_Good signals.