4K resolution allows to watch content as close as 0,75x the screen diagonal without seeing the pixels. This distance allows a great experience with image detail. But there are some important points to respect to get good results.

When uploading a 4k video to youtube and vimeo, the heavy compression will hurt the shadow areas with macroblock and mud, so an image with more light is welcome to preserve the image quality. So when shooting a person face it is good idea to use a fill light or to use a reflector sheet if the face is in the shadow when shooting outdoor, or even a strong 1000 watts bounce halogen light pointed to the ceiling or to the wall when shooting indoor. This will help to avoid macroblock and mud in the skin after the web compression. A balance between quality and message needs to be found.

4k video is more pleasant with steady camera. The camera movements disturbs the perceiving of image resolution and detail. So using lens and sensor stabilization is welcome. Also using a tripod and camera stabilizers are good idea. A higher shutter speed is welcome when doing camera movements like pan and tilt to avoid blur in resolution. So a balance between action and resoluiton needs to be found.

Sitting very close to the screen to enjoy the great 4k video resolution is much better when the image is more steady. If the image have movements it is more pleasant to watch the content in a distance equivalent to 1.5x the screen diagonal and at this distance there is no significant difference between 4k and 1080p video. Camera movements watched very close can make the audience to feel distracted, dizzy or sick, so a 1080p video is enough if you use camera movements in your shooting aesthetics because audience will feel more confortable sitting away from the screen. So 4k video is more pleasant with steady images.

The M43 sensor and S35 sensor are small and magnify all lenses imperfections in 4k video. So a good solution is to use lens aperture in f4, f5.6 and f8 to get the best resolution and to avoid lenses imperfections (softness and diffraction) and get a good 4k image. These fstops increase MTF results and allows to use a full frame lens in small sensors for 4k video. Also the focal reducers get a good result with these fstops. This will result in less shallow dof, but for 4k I think less shallow dof is good because you get more perfection in the image resolution and more things will be in focus. So a balance between shallow dof and quality needs to be found.

Maybe the 4k video is asking us to develop a new video/cinema aesthetics with steady images, more light and less shallow dof to allow people to sit close to the screen and enjoy the image detail. But we cannot change the video/cinema language just due to image detail experience, so we need to find a balance between screen distance and detail experience versus lens, light and movements.

See this test to perceive the results with different fstops with and without focal reducer: