Hacker News Reader

Comment by e_y_ | Hacker News Reader

On paper, Google's throw-everything-at-the-wall-and-see-what-sticks strategy (that has lead to a substantial Google Graveyard) seems like it was intended to allow for some parts of the company to innovate while keeping the core products stable and boring. In practice, many of those innovations (Google Inbox, anyone?) were not deemed profitable enough to keep around. Others were never given the resources to grow beyond an experiment. And even with a long leash, a big company project is never going to innovate as quickly as a startup.

This year, however, with the extremely deep cuts to Google's internal incubator (Area 120), it seems pretty clear that they've given up on this strategy, at least for anything that isn't somehow AI-related.

View on HN

It's worth noting that was a turning point: the "more wood behind fewer arrows" policy adopted by Larry [1] initiated the die-off cycle of Google products. Prior to that, as far as I am aware, they were much more tolerant of products staying around in a mature-but-not-wildly-successful state. Afterwards, it seemed as if they would only keep things that maintained a trajectory to become as successful as their core products.

Again, this was not entirely unpredictable. While I don't remember the details of that lecture, I remember the professor calling out these sorts of big shifts in cultural values as being typical of startups transforming into large companies. And Larry himself was part of the transformation, turning into (presumably, what he believed to be) what was needed to lead Google into its next stage as a large company.

[1]: https://googleblog.blogspot.com/2011/07/more-wood-behind-few...

What’s remarkable is that that phrase was already, at that time, notorious for having been a portent of doom at Sun Microsystems.

..and now all that's left of Sun is the wood on the back of the Meta sign.

And the Java/MaximeVM legacy, alongside some entertaining podcast stories.

Every time a socket is instantiated, Sun Microsystems comes alive; so too it is with NFS, and many other core technologies; there is lots that's left of Sun which still makes it live, even though it doesn't officially exist any more.

And any time 0xide Computer ships a cloud in a rack with the Helios operating system powering it, Sun Microsystems shines bright as a beacon of indestructibility due to quality.

All the wood behind one arrow…

It seems like the energy and creativity of Google Labs may have been lost rather than infused into all of Google.

They made - and still make - one crucial error: you need to spin those projects that are simply viable out immediately after they take root. Otherwise you will end up with the brand advantage but there will always be the pressure to use the resources (people, mostly) more efficiently in terms of ROI. And so nothing ever lasts and slowly but surely your reputation as a reliable partner for new products is eroded.

You can use your main brand for the launch, but then you have to be willing to support the child.

In addition, I don't understand why they stopped them entirely, instead of spin them off as subsidiaries under e.g. Alphabet; make them financially self-reliant, have Google/Alphabet as the main shareholder, and give the people that worked on it (and whoever else wants to) the opportunity to continue working on the product.

Some wouldn't have been viable, sure. Others were probably too ingrained in Google's hardware/software ecosystem to be separated out (although I wonder if nowadays everything Google runs on its cloud offering, which would make it simpler, just change the billing).

Most of them were probably built on Google's core infrastructure in ways that make them difficult to externalize at sub scale. There's also compensation disparity - it's virtually impossible for a new startup / spinout to pay FAANG comp and remain profitable & nimble. This results in braindrain.

The reason is quite simple: why spend engineering headcount on a less successful product?

> Some wouldn't have been viable, sure. Others were probably too ingrained in Google's hardware/software ecosystem to be separated out (although I wonder if nowadays everything Google runs on its cloud offering, which would make it simpler, just change the billing).

Google Cloud is built on top of Google's tech ecosystem, not the other way around.

Because things need time and alignment of incentives between creators and consumers and if you interfere in that relationship all the time things will never ever succeed. The last thing any project needs is an investor with a majority interest that fucks up your plans all the time, can take your employees away at will and can axe the project at any time because it doesn't perform according to their metrics.

That's why VCs take a minority stake in start-ups. The trouble usually begins when the founders dilute to the point that they no longer have a majority.

Assignment: Consider Agilent, itself a spinoff from HP, also spun off four (I think) companies, because they were good businesses but distracted the management.

Is this the right strategy? I'm tempted to say Yes.

>Google Inbox

Still so damn bitter about that death.

It’s weird that Gmail never reached that point. Even years later it’s still the same it was 10 years ago.

I think after inbox died I just gave up on it and moved to fastmail.

I understand what you're saying (I miss Inbox, too). But end users like sameness and "it just works." Normies prefer stability over innovation when they are trying to get stuff done.

Inbox was not deterministic. It's like a social media feed for your e-mail. I'd rather have my mails left as-is, and allow me to work with them the way I want.

You don't have to be a "normie" to appreciate simple or old fashioned things. Most of the e-mails I receive is not for quick-consumption, and I prefer the standard way over Inbox's way.

Maybe it was a solution trying to find a problem, IDK. I used it for 30 minutes tops.

Normies like sameness, this is why God gave us configs. You can default to normal but let the crazies customize.

Thing is, they quit on it too quickly; sure, a lot of people would stick with gmail, but others - and following generations - would adopt and grow up with inbox.

I think / suspect that many people develop habits of this type in their 20's and never move away from it for the next 60 years because it works. Example, people who still use vi(m) / emacs. Nothing personal, but as an example, they use it because they're used to it and have been for decades. And no editor that claims to be better will ever replace it for them.

Gmail settings still look like they were made in 2001. Any UX designer will want to pour Clorox into their eyes by just looking at it. Remarkable how Google can just totally ignore it, but then again - email is a very specific product. Once you grab your market share, we are locked in into our emails.

I miss Inbox features too. Shortwave is nice if you want something similar today.

It is the best email client I have ever used.

I mean that incubator was a total waste of money. No one did anything, everyone was a bser from the top, and 95% of the projects were total failures. I think there were maybe 3 "successful" projects.

I ran one of the successful projects in Area 120.

I joined Area 120 with huge skepticism. It was hamstrung and inefficient in its own ways. And I agree it didn’t reach its potential - largely because it was encased in Google 2020 instead of Google 2007.

But to my surprise almost all of the projects were impressive, well-conceived, promising bets. And the people in Area 120 were among the top 10% of Googlers I worked with in my decade at the company.

Google killed Area 120 because of bureaucracy and politics, full stop. Google is worse off because of it.

Somewhat spicy take - if the people in Area 120 were among the top 10% of Googlers you worked with, they probably weren't the right builders to start a new vertical.

Most of what makes people effective at large companies is neutral or negative value when applied to very early-stage companies.

You’re not wrong. They were among the top 10% of people I worked with in terms of passion, commitment, and creativity. They weren’t among the top 10% in terms of their skill in navigating Dilbert-land corporatism.

A significant number of the people in Area 120 projects were folks who were stifled and/or wasted in their previous Google jobs. One explicit purpose of Area 120 was to prevent the loss of these entrepreneurs to outside startups. Not incidentally, this was a form of cultural reinforcement - Area 120 burnished Google’s reputation as a good home for entrepreneurial mindsets.

"One explicit purpose of Area 120 was to prevent the loss of these entrepreneurs to outside startups"

So basically google had a shed where they hoarded talented people, to prevent competition? :)

I Don't think hoarding is necessarily the right word. They were using them to research potential new products or tools. The theory being that if only a few of the projects prove high value then it's worth it. That's not hoarding that's letting them flourish.

> So basically google had a shed where they hoarded talented people, to prevent competition?

That's a succinct description of why Microsoft Research was created.

Bill Gates explicitly said in an interview that rather keep people busy than losing them to the competition.

1) in the case of Area 120, this is one of the ways it was pitched to management. “Passionate entrepreneurs are leaving to work on new ideas; if you give them a place inside Google to pursue new ideas, it keeps them and their entrepreneurial energy at the company.”

2) in general, early Google used to hoard talent all the time. The founders would keep great people (or their friends) on payroll for ~ever just to have them stick around. That was most prevalent in the first decade of Google’s life, to my knowledge, and mostly applied to very senior people.

By the time Area 120 was pitched and approved (circa 2014), those days were largely gone. Area 120 was primarily filled with junior people (L4-L6) and constantly had to sing for its supper - it was not at all a sinecure.

I know you're not wrong, but it stings a little to see L6 referred to as junior.

That assertion applies to the middle 80%, IME. The top 10% are the people you can drop on to any project of any size and any org structure and they adapt quickly and deliver. They adapt themselves accordingly.

> That assertion applies to the middle 80%, IME. The top 10% are the people you can drop on to any project of any size and any org structure and they adapt quickly and deliver. They adapt themselves accordingly.

These are rather the top 10 % sycophants, not the top 10 % researchers or top 10 % programmers.

I didn't see that mentioned, perhaps I missed it. I read it as top 10% of performers.

3 successful projects can totally justify what you call waste of money.

I sometimes wonder what people expect innovation is. You try and try and try. One thing is good and you must know how to use it - it can make history.

If I understood right, chatgpt comes from one of such ideas.... so the question is also: who evaluates the ideas? How come that Google was not able to capitalize on that idea?

So yeah, instead of treating the cause they treat the symptoms, like usual.

Agreed, we are on ycombinator.com, after all. The patron saint of failed ideas.

I think this is why these teams are really hard to have in a mature org. In reality maybe 5% of projects in one of these innovation orgs is actually great! But it’s impossible to evaluate and everyone else is thinking some variant of “this team is able to bs and show no value, while I have to hit real goals or risk being fired?”

I think the incentives would have to be much different for it to work (e.g. much lower base pay + higher rewards for success)…..but at that point just join a startup

Which 5% of projects are really great? In my experience, presuming you have tight filters such that all of your projects are plausibly potentially great, you really don’t know until you try. That’s the point of an incubator.

It’s not that hard to evaluate when something is working (ie the hard part in evaluation is false negatives, not false positives).

In Area 120’s case there was no coasting - if anything there was a hair-trigger standard to shut down underperforming projects.

I think these type of teams are a good way to give talented devs a break from the grind at bigger companies, even if the chances of a new product is low.

Not every company can afford these "paid vacations", but they do have some use at times.

Pretty standard rate of failure for early stage startups.

> at least for anything that isn't somehow AI-related.

If you can't innovate at the base level of app design .... how do you have any hope of innovating for AI apps that require research/engineering/product/marketing collaboration?

That's true. What they need is what they had started doing, i.e. breaking down Google into Alphabet and letting some companies within the conglomerate act like startups.

Why was this effort unsuccessful? Perhaps they were unable to get rid of middle management? I have had lengthy discussions with employees from several of their companies, e.g. Calico, and that seemed to be the case. This article only reinforces my view.

> Why was this effort unsuccessful?

i suspect that they are unsuccessful for two reasons: failure is not death, and success is not riches (for those who did the work).

If I recall past discussions on this topic correctly, it wasn’t just about profits. I believe the incentive structures are setup around launches and not maintenance. If that’s correct, then that would lead to people launching, collecting rewards (bonuses, promotions, etc) and then abandoning.

I think you are talking about “innovators dilemma” great book by the way

Microsoft today is more innovative than google IMO. They keep executing bold and controversial strategies, even though being older than google.

At roughly similar points in their lives, Microsoft and Google initiated video gaming platforms. Microsoft stuck with Xbox through seven years (iirc) of still-unprofitable growth to what it is today. Google abandoned Stadia after three years.

Yeah, I think Google is looking more like an IBM in the long run, while Microsoft manages to innovate despite its age and size.

If we ignore the GUI civil wars for a bit, though.

Reply on news.ycombinator.com