The "but everyone does it" discourse is so tired, and so much so for cybersec employees.

The main value of ‘together mode’ was that everyone had a consistent view of the layout of the ‘room’. Doing intros by going around the table was easier, for example.

The one technique wish they’d implemented on top of this was based on some papers from the Rainbow Group in Cambridge that let you adjust the eye position of videos in real time. I wanted them to add something that would track which person you were looking at on your screen and then make you look directly at them on their screen and in the direction of the speaker in a consistent-layout render of the other videos. The technology for doing this existed five or so years ago.