What Google Learned from Its Quest to Build the Perfect Team | Hacker News Reader

What Google Learned from Its Quest to Build the Perfect Team

nytimes.com

94 pointsskrish51 comments

previous best (of 5 submits), 4 months ago ~ https://news.ycombinator.com/item?id=11174399 worth re-reading this because it is a timeless problem.

And also as mentioned in previous threads where this link was posted, all the other stories in this "The Work Issue" are equally high quality.

Got through the first 3 paragraphs where the title hadn't even been acknowledge, let alone answered, and stopped reading.

Increasing I find storytelling in journalism to be less and less tolerable. Don't get me wrong, there is absolutely a place for long form non-fiction. If this were titled in such a way that it were to be a story and not a fact-piece I would probably have less of a problem with it; I could have filed it away in a to-read-later bookmark.

But when the title elicits that there is information to be had; I want that information, and nothing else.

The whole article is about how empathy and a safe place to speak up make a better team.

But as usual for nytimes journalists, the reading is barely tolerable.

That's my understanding as well. Perhaps the author wanted the reader to empathize with the people in the article?

Seems reasonable, but perhaps it might be premature, given the premise is that we have some problems empathizing in certain contexts. Maybe this is one of them.

I skimmed the entire article and it just kept teasing and teasing but seemingly never revealed the promised information. If it's in there at all, it's well hidden.

Here's the TL;DR

After studying over 180 google teams they didn't find any evidence that the composition of the teams mattered (i.e. friends outside of work, male/female, introverts, extroverts, intelligence). Instead what they focused on were the “group norms” or unwritten rules of how the groups interacted with each other. There were two norms that were most important to a successful group.

1) Members spoke in roughly the same proportion, what the researchers call, ‘‘equality in distribution of conversational turn-taking.’’

2) Members all had “average social sensitivity”. They could understand how other team members felt based on non-verbal queues. The underperforming teams had low sensitivity amongst the team members.

Thanks for the TLDR.. that's almost Feynmanesque..

That's really unfortunate and naive. Not everything, particularly the subtle dynamics of human interaction, can be succinctly distilled into an abstract-style paragraph. Oftentimes it takes great writing skill to set right perspective, context, and background for the information to make sense or "click" - perspective which I find to be more valuable than the raw "information" or study results.

Storytelling is sometimes the only way to properly convey information.

- slightly edited for clarity -

No, I'm with Touche here. Too many articles hide the actual story under a "human angle" aspect where we have to learn about someone's early childhood growing up in Soviet Russia, or their first failed business, or that time their father said Something Memorable to them. I've read so many of those now that I start to fall asleep before the article begins talking about whatever it is talking about. The naivety on display here is the journalists sticking hard-and-fast to the rules we're taught in middle school English class: you have to have a "hook" so that the audience get's caught! But those hooks are artificial and repetitive - let the content speak for itself. If it can't, then the article isn't worth my reading.

Anecdotes are still relevant. But using them religiously as the opening turns me away.

That's especially annoying since journalists should know better! It's them who have "golden standard" of the "inverted pyramid". I.e. a headline should be the "tl;dr" of the article. Then the first paragraph should be a little longer "tl;dr" of the article. The reader should be able to recursively learn more details, but also should be able to walk away with all the core facts after the very first paragraph.

I enjoyed listening to the book "Smarter Faster Better" in which this article is one of the chapters.

http://www.audible.com/pd/Business/Smarter-Faster-Better-Aud...

It also deals with how the script of movie "Frozen" evolved and I found that very interesting as well. It showcases how your idea gets better & better for execution because of the team.

You remember the concept of a thesis statement? This author didn't.

Storytelling is NEVER the ONLY way to convey information, though it can be the MOST EFFICIENT. Specifically, the narrative form tends to be more memorable/impactful than disconnected propositions, so if the majority of the context in the story is necessary to understand the information, it ends up being a net win.

Read my other comment; the title didn't convey this subtlety so I didn't expect the article to.

I think this is like learning styles, different people need different communication methods.

That first anecdote in the article is relevant, and the article does eventually get to talking about Google. This is the nature of long-form, it doesn't have to dive straight in.

Titles are important. They not only tell you what an article is about, but also the form the article takes.

Using "What" in the title was a mistake here. When you see "What" you expect mostly facts, mostly findings. If they had titled it with "How" I would have been more prepared for a lack of immediate answer. "How" denotes that an explanation is needed, and backstory is often necessary in explaining.

Titles are especially important on the web because links only have titles. There's no other context to which you can get an idea of what you're getting into. It's not like in newspapers where you have the context of what section an article is in or visually see how long it is from a glance.

If you're looking for ways to prepare, at the top of the page, it also says, "The New York Times Magazine", which has been known for long-form journalism for more than 100 years.

> This is the nature of long-form

Let's not confuse long-form "there's a lot to be said here" with long-form "we really like padding".

"The story of computing begins roughly 4000 years ago..."

There is so much content online, and readers are so fickle that it makes too much sense to me for writers to be as concise and straight to the point as possible. I read the article on the subway train so I didn't have much else to do but if this was a tab I was glancing over at work I wouldn't have made it past the first paragraph.

There are different reading modes. The New York Times Magazine in particular caters to people who are sitting down on a quiet day for long reads. It was created specifically to hold pieces longer than appropriate for the normal newspaper.

So here, it definitely doesn't make sense for writers to be as straight-to-the-point as possible. It's like asking for some long, cinematic movie (e.g., Lawrence of Arabia or The Revenant) to be edited down to a 60-second YouTube video. Sure, you could squeeze in the highlights, but it would miss the point.

Sometimes you want a grab-and-go sandwich from a drive-through. Sometimes you want a relaxed 2-hour meal with old friends.

Fair enough. Didn't know about NYT's target demographic.

Yeah, this has been driving me crazy lately. Reminds me of local news. "Important information for parents. Find out the dangerous thing that may be hurting your kids.". But we're not going to tell you. We'll keep teasing it over and over. It's really important, but we really don't care if you get the info.

I'm glad someone else shares my frustration with long form journalism these days. Your thesis should always be clear from your first paragraph. I read some of these articles and I'm half way through and I still don't know what the point is.

Yes, but they aren't nearly as bad as the new yorker. A few weeks ago, I started to read this article on the hiroshima bombs and it started out awesome. However, after the few first paragraphs, they start going off on the life stories for each of the characters while adding tons of extraneous details. I gave up in frustration after 10 minutes of that crap. That's happened to me almost every time i go there so now I usually just autofilter new yorker articles out.

I've been trying to think of a new term for this type of writing as it's not quite clickbait but I haven't come up with anything quite as catchy.

That's actually an incredible article that started the whole long-form news trend (1946). The New Yorker dedicated the entire issue to the story, and it's been republished as a book:

https://en.wikipedia.org/wiki/Hiroshima_(book)

I'd recommend you try and read it again.

Sounds like you're referring to "Hiroshima", published as the entire editorial content of an issue of the New Yorker shortly after the war, later published as a book of the same title, and IMHO one of the most important magazine articles ever written. (For instance, here's an entire web site about that article: http://www.herseyhiroshima.com/index.php)

This is why I (often) check the comments first

Here is a high-level summary...

1) Members spoke in roughly the same proportion, what the researchers call, ‘‘equality in distribution of conversational turn-taking.’’

My guess is that teams which have members with high sensitivity didn't do well, either. If you're afraid of speaking up because someone will get upset at you, you're only a superficial member of the team.

Both of these criteria amount to different facets of the same thing: teams have to work together as a team. That sounds stupid, but practical criteria as given above help to define that simple statement.

> high sensitivity

I agree. That is probably the wrong term for the phenomenon. There are elements of forthrightness and trust that aren't captured by the word 'sensitivity'. On top of that 'sensitivity' carries negative connotations to many.

I think this is what's called Power Distance Index (PDI) in Hofstede's cultural dimensions theory. It's a fairly well established set of sociability measurements that's been around since the 1970s and is quite well established. Interestingly it also emerged from another (perhaps the original) large multicultural global technology organisation: IBM

  https://en.wikipedia.org/wiki/Power_distance

This continuous Google worshiping is really tiring. When you talk to Googlers nowadays, they seem to be pretty depressed as interesting projects dried up, and demoralized by seeing people spending 90% of their time playing office politics and backstabbing getting promotions. Compare Google in 2006 with the one in 2016 - would you really like to work there these days? If you are smart, you can build a core of your business and then let yourself be acquihired by the big players if you wish to, why bother with employment at Google unless it's your first job outside university?

Any data to back up your assertions? It's often rated #1 best company in the USA to work for [0] and has a pretty high 4.4 rating on Glassdoor (from 4300+ reviews) [1].

[0] https://en.wikipedia.org/wiki/Google#Corporate_affairs_and_c...

[1] https://www.glassdoor.com/Reviews/Google-Reviews-E9079.htm

Disclosure: I do NOT work at Google.

Honestly, despite years of thinking otherwise, I have absolutely no interest in building a business. Wanted to do the startup thing, tried bootstrapping a mobile app consultancy around 2008, etc. I went entirely broke (after having been a grad student, so no great feat, really) trying to convince myself to do the things I knew I needed to do to build a business.

After a few years elsewhere, I now work on Google Compute Engine. In particular I'm tech lead for the team that builds our virtual NIC and hypervisor-level networking features. I get to spend most of my days focussing almost entirely on how to make a piece of technology better and faster for our customers. Moreover, I get to do this with a set of team of people who are all outstandingly bright, polite, and dedicated to our common cause. Sometimes a business case has to be made when particular decisions will move the needle on costs, efficiency, revenue, etc., but that's a far cry from having to actually build/run the business itself.

I very much like working at Google in 2016. I get to work on technology and witness clear customer benefits without having to worry about building/running the business, meaning I also get to have a solid work/life balance.

This article is not worshiping Google and talks about timeless issues in every org/team. It's a long read but worth the time.

> If you are smart, you can build a core of your business and then let yourself be acquihired by the big players if you wish to, why bother with employment at Google unless it's your first job outside university?

Because if you care more about tech than about building products, startups suck. It's a whole culture based on wasting money with inefficient technology (cough cloud) at a ridiculously small scale, both in usage and in complexity.

I've had 20% projects at Google that I spent a few weeks working on and likely handle more traffic than a whole batch of YC backed startups.

Depends on the startup

Not everywhere is a joyful money incinerator.

>> Compare Google in 2006 with the one in 2016 - would you really like to work there these days?

That would apply for any company that's been a while.

tl;dr - teams are most effective when people feel psychological safe to contribute equally. Creating psychological safety can be accomplished through conversational turn-taking and empathy. In the best teams, members listen to one another and show sensitivity to each other’s ideas feelings and needs.

See Asimov's ideas on creativity - it is essential for the production of new ideas that team members feel they are in an environment where "stupid" ideas are not mocked. For every 1 good idea it's fair to imagine there might be 10 daft ones. The more ideas in circulation the better - and it also gives people with daft ideas an opportunity to have them challenged in a constructive fashion - which may lead to them developing the kernel of their idea that might not be so daft.

    http://www.openculture.com/2014/10/isaac-asimov-explains-the-origins-of-good-ideas-creativity.html

This is the best high-level summary so far. The thesis doesn't really crystallize towards the end of the article. I'm not sure if people aren't reading the entire piece or if the author needs to work on clarity a bit.

As other are saying, so many words, so little content. tl;dr: you can't rely on data alone to build optimal teams (sometimes it's about experiences and such) and Google pretty much did nothing to optimize teams (in spite of spending years on it)

What I get from this article is that working collaboratively is a lot more efficient than working in a spirit of competition. Specially, if you have members with different backgrounds among your team. Working collaboratively requires a trust bond between each members and usually it takes time. As we build and dismantle team sometimes at a fast pace it may be difficult to attain this state in today'S work environment.

I am really tired of long articles. how hard is it to provide a few bullet points or a two lines summury for people that do no afford to read the entire page?

I was playing around with this idea at Abbreviated Press. http://abbr.press - Pet project because I was sick of reading long articles, and just wanted bulleted format as well.

The day my manager takes me to an offsite and requires personal self-disclosure to my coworkers is the day I quit.

I like my coworkers--some of them quite a bit, but I have boundaries. And my personal life is not material for my company to improve its efficiency.

I don't want Sakaguchi to die!

While reading that article the page changed its font-size by at least 150% five times (thus far). No I did not press any buttons. There is some detection Javascript running that gets confused. Maybe my touch screen (laptop)? Page zoom does not change, CTRL-0 does not reset the size, so it's not me.

As for the article... I'm amazed this is so popular (given the attention previous submissions here already got). Well, I guess it's nice to have a link to point to for all the things that do not matter.

This sentence scares me:

> Rozovsky and her colleagues had figured out which norms were most critical. Now they had to find a way to make communication and empathy — the building blocks of forging real connections — into an algorithm they could easily scale.

And this is a surprise:

> ‘‘By putting things like empathy and sensitivity into charts and data reports, it makes them easier to talk about,’’ Sakaguchi told me. ‘‘It’s easier to talk about our feelings when we can point to a number.’’

and

> And thanks to Project Aristotle, she now had a vocabulary for explaining to herself what she was feeling and why it was important. She had graphs and charts telling her that she shouldn’t just let it go.

Really? I have to crunch some numbers how I feel about this.

PS: figured it out: When I just touch the screen the font-size changes. There are three steps. It takes a double-click with the mouse - but only a single touch with the finger on the touchscreen. I can't imagine this being useful anywhere - especially since when you have a touchscreen you can also already do a two-finger zoom if you want to.

Reply on news.ycombinator.com