Nuanced communication usually doesn't work at scale • Hacker News Board

Re: Nuanced communication usually doesn't work at scale

#111

Post by tharkun__ » Sat, Jan 29, 2022, 9:43 PM UTC

88913527 wrote: post #103

mcguire wrote: post #52

Earlier quoted context omitted.

How much do you want to bet most people will read your first bullet point, ignore the rest, and drop all the nuance? Hell, I've learned not to ask more than one question in an email. The first one is the only one to get answered.

It's easier to consider this question with empathy: imagine times where you replied to emails partially, answering some (one), but not all questions. Ask yourself why you did a partial reply. Then, when you ask questions of others, apply those learnings. For me, I tend to 'jump' to the first answer that comes to mind, without reading the full nuance, likely because I'm optimizing at replying sooner, so I can move ont…

    imagine times where you replied to emails partially

It will be hard for someone that always replies to the first thing only to empathize with this but: This has literally never happened to me. As in, I have never replied partially to something in an email. You will get an answer to each of your items. Granted, you may not get the answer you were looking for but I will answer each and every one, even if it's just a "I will have to look into this one and get back to you" so that the other 6 items can get answered right away.

Why do the thorough people always have to empathize and not the other way around?

Re: Nuanced communication usually doesn't work at scale

#112

Post by wallacoloo » Sat, Jan 29, 2022, 9:45 PM UTC

regarding the apparently competing values of reliability and velocity, try propagating these goals through different mediums. code quality can absolutely be ingrained into the culture. things like TDD, when/what to integration test, and encouraging PR reviewers to enforce these requirements. you can literally just run an in-house course/training for everyone and the aggressively follow through.

once you’ve got reliability deep-seated in the culture, then you can talk about velocity all you want without worrying that your teams will make confused tradeoffs. in effect, you’ve communicated the nuanced concept that “we move fast, but always within the constraints that provide for a reliable product”. most people aren’t consciously thinking about it that way (not a bad thing), but their behavior matches what you were originally wanting to convey with a nuanced message.

Re: Nuanced communication usually doesn't work at scale

#113

Post by dirtside » Sat, Jan 29, 2022, 9:48 PM UTC

mcguire wrote: post #52

logicalmonster wrote: post #18

Nuance is hard to convey in groups, but I believe that *a small part of the problem is a lack of design*. Many peoples' eyes glaze over when they see a wall of text in an email and they just skim rather than read. Some simple things to enhance communications can be the following. * Use a few bullet points to put attention on the main points you want to convey. * Without going overboard, use a tasteful amount of graph…

How much do you want to bet most people will read your first bullet point, ignore the rest, and drop all the nuance? Hell, I've learned not to ask more than one question in an email. The first one is the only one to get answered.

A major factor in this is a lack of willingness to take the time to understand something, possibly rooted in a meta-failure: not understanding that it takes time to understand things! There's various motivating forces that impel us to race along to the next thing instead of taking a little bit to absorb something, think about it, or discuss it.

Re: Nuanced communication usually doesn't work at scale

#114

Post by loteck » Sat, Jan 29, 2022, 9:54 PM UTC

emptybottle wrote: post #95

“Nuanced communication usually doesn’t work at scale” proceeds to write a lengthy nuanced post about it

Ironically, this criticism is addressed as a nuance in his post.

Re: Nuanced communication usually doesn't work at scale

#115

Post by Groxx » Sat, Jan 29, 2022, 9:57 PM UTC

logicalmonster wrote: post #18

Nuance is hard to convey in groups, but I believe that *a small part of the problem is a lack of design*. Many peoples' eyes glaze over when they see a wall of text in an email and they just skim rather than read. Some simple things to enhance communications can be the following. * Use a few bullet points to put attention on the main points you want to convey. * Without going overboard, use a tasteful amount of graph…

And your very first line of text should say who it is for, and why they should care.

Help your readers triage. They already get too many emails.

Re: Nuanced communication usually doesn't work at scale

#116

Post by andersonmvd » Sat, Jan 29, 2022, 9:57 PM UTC

Look at how many details this guy (Carl Sagan) conveyed in his 15' speech: https://www.youtube.com/watch?v=Wp-WiNXH6hI in a way that potentially everyone understood. I wonder if every explanation was like his, if nuance wouldn't be well communicated, even to large groups.

Re: Nuanced communication usually doesn't work at scale

#117

Post by loteck » Sat, Jan 29, 2022, 10:00 PM UTC

If Dan is right (and I think he is), have a stiff drink nearby and then think about how his point applies to something complicated that is life and death, such as conveying information to the public about a rampant virus.

Re: Nuanced communication usually doesn't work at scale

#118

Post by cjsplat » Sat, Jan 29, 2022, 10:00 PM UTC

funstuff007 wrote: post #9

IIRC Jonn Kerry's campaign for President was plagued by too much nuance in both his and the campaign's communications. And the Romney critique as the "flipping Mormon" can also be seen as a rejection of nuance.

I don't think nuance was the issue for Romney and the "flipping" issue.

Romney was a pretty successful Governor of a liberal state. He needed to appeal to nationwide Republicans in the primary so he slid a lot further to the right. Then he was up against a reasonably popular Democratic president, so he slid back to the middle.

The only nuance was that he tried to muddy the sloshing to make it look like he wasn't changing his positions.

Re: Nuanced communication usually doesn't work at scale

#119

Post by qznc » Sat, Jan 29, 2022, 10:00 PM UTC

smarx007 wrote: post #76

Au contraire, academia has worked out how to reliably communicate at scale: publishing peer-reviewed articles in journals.

Are you talking about the academia where everybody complains that people only read the abstracts?

Re: Nuanced communication usually doesn't work at scale

#120

Post by theptip » Sat, Jan 29, 2022, 10:01 PM UTC

tarr11 wrote: post #28

wrs wrote: post #11

Earlier quoted context omitted.

One reason is that non-scaled comms suffer from “telephone game”. If you do it like you’re saying, then go down 3-4 levels in the hierarchy and check what people are hearing, it will have mutated away from anything you originally said. Sometimes to get everyone aligned (as best you can) you have to give everyone the same message at the same time — but it has to be a simple message.

I wonder if you could checksum the telephone game by having a 2-level comms. That way the VP could validate that the Director didn't make a mistake when communicating to the EMs. Eg, n-level comms are where a CEO communicates to the entire org 1-level comms are where CEO->CTO->VP->Directors->EMs->ICs 2-level comms: CEO->CTO+VPs CTO->VPs+Directors VP->Directors+EMs Director->EM+ICs

The term of art for this practice is “skip-level 1:1s”. It’s not ubiquitous but many think it’s a good idea, particularly when stepping into a new role/org. Obviously it can end up being a lot of meetings so it’s typically on a quarterly or less cadence.

Because of the branching factor it’s not feasible to do this for every bit of comms.