> The channel width of face-to-face communication is higher than that of any other method.
Research to that effect? I know some people have that, I know I do not. So this absolute statement is already 1 datapoint (me) off. And I know others (who I mentor) who absorb mentoring far better via remote methods. When someone mentors me, I like to chew the information over, research it etc. This goes for most communication for work related matters; I absorb far less if it is told to me face to face than when I read it and have time to thoroughly understand it all.
I agree somewhat with the rest but that again can be done remotely: we develop almost everything while in chat; this acidental transfer is even far better than face to face imho. In chat someone says something like ‘I refactor this to that by hitting ctrl-something’ ; all people see it, not only the accidental junior next to you but all of them. And later this can be searched by others while the verbal incidental comms is gone forever.
I definitely see the need for face to face meetings, but far less than some people here do. And it would be good to have more research done into this matter: I simply do not see many programmers (and I managed 100s over my professional life) being productive in the setting you suggest. Looking and feeling busy yes (so for the bums on seats type of manager), but productive as in just getting stuff done, in my experience no. But that is anecdotal, like your view, so real research and numbers would be good.
"I simply do not see many programmers (and I managed 100s over my professional life) being productive in the setting you suggest. "
To be precise, I did not suggest anything else beyond people being in the same location. I agree the type of bullshit you describe is plausible in a typical corporate scenario.
Given the reproducibility crisis in modern psychology, I would be hard-pressed to claim most thing dealing with humans as better than "anecdotal". I pass along those anecdotes that I find useful. In this context literary references are not "academic proof" but rather probabilistic proof that someone else shares my view and that it might be applicable to third parties as well.
The face-to-face plus whiteboard offering the widest communication channel -anecdote was taken from Alistair Cockburn's "Agile software development" that can be referred from example here (just look at the graph):
I have no idea how true that is in the general sense, but personally I've found this to be true. I've had and continue having tremendously valuable whiteboard impromptu chats that have clarified a complex topic to all stakeholders, helped me to understand some issue and thus saving worktime, or just helped to create a shared symbolic context for a complex technical discussion that otherwise would have been imprecise and hand-wavy. The fact that you can draw boxes, point at them, and perhaps draw arrows between them in a shared physical space with people just somehow seems to make everything easier.
Research to that effect? I know some people have that, I know I do not. So this absolute statement is already 1 datapoint (me) off. And I know others (who I mentor) who absorb mentoring far better via remote methods. When someone mentors me, I like to chew the information over, research it etc. This goes for most communication for work related matters; I absorb far less if it is told to me face to face than when I read it and have time to thoroughly understand it all.
I agree somewhat with the rest but that again can be done remotely: we develop almost everything while in chat; this acidental transfer is even far better than face to face imho. In chat someone says something like ‘I refactor this to that by hitting ctrl-something’ ; all people see it, not only the accidental junior next to you but all of them. And later this can be searched by others while the verbal incidental comms is gone forever.
I definitely see the need for face to face meetings, but far less than some people here do. And it would be good to have more research done into this matter: I simply do not see many programmers (and I managed 100s over my professional life) being productive in the setting you suggest. Looking and feeling busy yes (so for the bums on seats type of manager), but productive as in just getting stuff done, in my experience no. But that is anecdotal, like your view, so real research and numbers would be good.