There is no substance in that quote at all. None of the users (of superfast lenses) said that the rendition is "soft".
This seems to be a rejoinder to my: "What's the point in mounting an ultrafast lens on a camera if extra glass (or lack of extra glass) between the lens and the sensor renders the lens very unsharp wide open?"
If no one here has noticed that the wide open imaging is unsharp with either the RX-type ultrafast lenses or the standard C-type ultrafast lenses then one possible explanation is that the pile of glass over the G1 sensor is around half way between the 0 mm thickness appropriate to C-type lenses and the 9.5 mm thickness appropriate to the RX-type lenses. "Splitting the difference," half of a very noticeable error might not be very noticeable.
Another possible explanation is that the users of ultrafast lenses here have lax standards of sharpness. Post #91 (from Photomorgana) in the "A superfast normal on the G1" strand shows two pictures made with the 26mm f/1.1 Macro Switar RX lens. The one shot at f/1.1 is, in my opinion, grossly unsharp. Look at the lower eyelid hairs in that picture! The one shot at f/1.8 looks pretty sharp (in the 1 megapixel JPEG posted).
You can object that I'm applying unreasonable standards of sharpness. I think not. It is a long tradition in photographic lens design that the sharpness at full aperture be judged against the sharpness at the next few apertures. If the f/1.1 image is that much weaker than the f/1.8 image the lens design is not regarded as succesful. But I've had experience with the 26mm f/1.1 Macro Switar RX lens used on a Bolex H16 RX camera where the f/1.1 to f/1.8 sharpness difference is nowhere near as great as in Photomorgana's pictures. So it's not the lens's fault. Perhaps the G1 does not have anywhere near 9.5 mm of glass over the sensor.
Another long tradition is to expect more image from more sensor while still expecting a less absolutely sharp image upon the larger sensor. For a nice example consider how Zeiss rates its cine lenses. Lenses for 16 mm cinema (diagonal 12.7 mm for Zeiss) are rated at 20, 40, 60 cycles/mm. Lenses for 35mm cinema (diagonal 30 mm for Zeiss) are rated at 15, 30, 45 cycles/mm. So the expected absolute sharpness of the lenses for 35 cinema is just 75% of that for 16 cinema, but since the 35 mm cinema frame is (linearly) 235% the of the 16 mm cinema frame the net comparison is (linearly) 177%. For digital photography it is the pixel count, not the sensor size, which should determine the expected sharpness of the lenses. The 12 megapixel G1 sensor is comparable to the sensors on the RED electronic cinema cameras which are using Zeiss lenses made for 35 mm cinema. Surely more image should be expected from the G1 sensor than from a 16 mm film frame.
Traditions can be trashed if you wish. Photomorgana's f/1.1 image can be called "sharp" if you wish. Have fun on eBay, because no one will be designing and manufacturing lenses for such tastes.