One example of where this is counter-intuitively wrong is that you end up with more diverse ideas if people brainstorm in private than in a group.

I think the reality of face to face meetings is that the result is often almost predetermined based on the existing group dynamics. How many people are turning up to the meeting feeling their input will make a difference versus people with (probably) the dominant personalities who are happy their established conclusion is chosen. And seeing the later as naturally coming to a better conclusion, faster.

As a dominant personality in meetings I definitely think about the effects of that a bunch.

> If the other people in the room are also knowledgeable, you will come to an even better conclusion much faster face-to-face.

> In my experience, having "group reasoning" is faster and with better results.

Why is face-to-face required for "group reasoning"?

> Why is face-to-face required for "group reasoning"?

Because it raises the bar on starting a meeting. And it keeps half the meeting from being muted/camera-off and totally zoned out.

If your meeting attendees are zoned out, they either need to be held accountable for not being attentive, or they shouldn't have been invited in the first place.

I agree, but with WFH its so easy to invite the entire team. I found the number of meetings I attend has skyrocketed in WFH - most would be better off with 1/3 of the attendants who took time to prepare instead of the whole team just to gather info.

It’s a round-trip latency difference. Standing in the same room, we can get perhaps 100, but certainly 50, responses in an hour from a small group. Video conferencing might be 15% less and that relatively slight difference is already frustrating. In asynchronous written form, it probably takes 3 days to get 100 “turns” and maybe a week or more if there’s timezone “smear”. The written async responses can be higher quality on average, but where reaching consensus in minimal wall clock time is important, face-to-face wins.

Why is Slack “better” than physically mailed correspondence? It’s roughly the same phenomenon, just on an even shorter timescale (including that Slack isn’t better for everything).

If you're a proponent of written communication- you'll argue that 100 'turns' of written communication will have significantly more valuable discussion than 100 turns in a meeting. Most of the turns in a meeting can usually be avoided by a google search beforehand

Totally true. There are many cases where even 2 or 3 turns of written comms beats 100 of spoken (and I'm one of the biggest proponents of written work and async work at my company).

But there are situations where speed of alignment matters more than the specific direction of alignment and other situations where the nature of humans requires a live, synchronous discussion. I couldn't build a relationship with my family using only written feedback and requests.

You want high and fast bandwidth between the different brains.

Me writing something in my cocoon for 15 minutes, then someone reading that and replying what is wrong with that is not very efficient.

Just throwing some thoughts in the open air and others responding to that, crystallizes much faster.

Same with giving feedback about user interfaces. It's so much easier to just walk to someone's desk and point at what you think should be changed. You can interact instantly.

That is not to say that I don't see the value in written communication, documentation and having a log of what happened. It's very much a trade-off.

What crystallizes may have been reached quickly but there is no guarantee it is the best possible choice given all the known and unknown unknowns that the experts may have not considered.

Then there is also the hierarchy of expert advice to consider where a room of experts some who are more senior (or louder or more convinced about their rightness) might push you towards a worse outcome than a well thought through option.

I feel a decision taken “live” with the help of experts needs to be viewed with more skepticism than less.

> a decision taken “live” with the help of experts

Maybe we are talking past each other, because in that case I might agree.

I'm talking about the case with only experts present. I know about A but not so much about B, the other person knows about B but not so much about A. We sit together and try to come up with the best technical solution.

Calls are even lower latency, you skip the "walk to their desk" bit.

Had mixed results.

I've felt like "group reasoning" in person is more like getting people on the same page to execute on an idea, which doesn't equate with the merit of that idea.

Once that discussion kicks off in a large-ish group, I've found it hard to pivot the direction of that conversation - if you think there's a better, independent solution that's not aligned with the group train of thought.

I can see your point.

I think it works best when there needs to be a solution where no single expert knows enough to come up with the best idea. So a group of complementary experts where the solution is more important than individual ego's :).

But when it's kind of decided already, so more 1 way, written communication works better.

I was in some situations where there was long emailing back and forth between people. Then it was decided "to just have a call on this", and came to a solution in 15 minutes.

It's definitely about trade-offs, and I can see benefits in both. So I wouldn't just throw away the power that "a bunch of people in a room" can have.

It's very likely that meeting resolves the issue in 15 minutes precisely because of the back and forth before. The back and forth syncs ideas, so the meeting can be used for the decision process.

> I think it works best when there needs to be a solution where no single expert knows enough to come up with the best idea

...then you have an organizational problem that should be solved.

Sorry for breaking the news to you, but a single person cannot know everything.

Except for everyone on HN of course. We're always fastest and know everything.

> In my experience, having "group reasoning" is faster and with better results.

Speed is not a proxy for decision quality. It's almost always the opposite.

Very few technical topics require quick forth and back. Writing is more effective communication.