Comment by morphentropic
7 months ago
Man they used to drop some awesome stuff: Google Maps, GMail (remember the hype over Gmail invites?), Google Earth... then they just stopped improving stuff and started releasing multiple versions of things and abandoned them all, over and over again. Very strange.
I was there around this time and remember the first time someone said out loud that they were doing project Z because "that's what will get me promoted". I argued until I was blue in the face that it was a bad idea, but they didn't care: they had their objective and knew how to get it. Unhappily, everyone was right: he got promoted and the project was an expensive failure.
My two cents is that Google has been consumed by its performance review process; the amount of money made by advancing dwarfed the amount of money made by making advancements, and as always the metric was the outcome.
I agree with you somewhat, having spent ~4 years at Google... though I think "perf-driven development" is IMO a symptom and exacerbating factor, but not a root cause.
Advancement and fulfilling of personal ambitions is a common thing in basically every sufficiently large company. Google isn't unique in having that problem - nor is their promotion process markedly different than everyone else's!
What is different is that Google is extremely metrics and OKR driven, combined with a near-total absence of product leadership. There is often no broader product strategy besides "grow X by Y".
This results in a critical weakness where you can get promoted for shipping Obviously Dumb Shit, because it hits some ill-defined OKR. It's practically an annual tradition within Google's management: creatively interpret pointless and vague OKRs so that you can make a (contorted) argument that Projects X and Y contribute to it, so you can ship it and get everyone involved their promos.
People in other companies are ambitious and want to get promoted too! The difference is that in many other companies there are other sanity checks in place that you don't get promo'ed for shipping Obviously Dumb Shit.
Google's root problem IMO is that there is an extreme lack of product leadership and product vision at the very top levels of the company. This results in a near-total inability to mitigate meta-hacks of internal promotion systems.
At companies with more product strategy at some point someone at a high level goes "Projects X makes no damned sense!" and puts the kibosh on it. At Google Project X will ship, and then after its badness becomes inescapable, get shut down.
The dominant culture in the company began to mimic the history of Raghavan: failing upwards, short-term gains with long-term detriment. When you get back far enough you begin to see a recurring pattern of it with these MBA/exec types who basically only have a bean-counter, extractive, understanding of running a business or making things.
Do you have a link to anything about performance review process . I am curious how msft or nintendo which known for innovation handles it .
They could at least open source all the stuff on google graveyard which will give us so many awesome softwares . Sadly they are now now sitting on some random hard drives.
Google Wave was open sourced as Apache Wave, I think. Not sure whether anyone actually utilized it...
Not really. The relevant parts of those programs are basically buildable based on a list of their features.
The technology that one would get in an open source situation isn't very usable outside of Google's ecosystem because Google builds software on top of Google's stack. Like, without the monitoring infrastructure they've built or the Borg scaling infrastructure, their software is actually kind of fragile because the ethos is " If it starts to malfunction break it quickly so it can cause a monitoring event and to get replaced by auto restart."
The Google way of doing things is actually not a great way to architect most software that isn't running on a giant data center structure.
That's not something Google invented, it comes from erlang. Systems in erlang (and other beam langs) are designed to fail and die, and get restarted by the supervisory tree.
2 replies →
Not that strange if you think about the nature of transformation Google went through. With time they grew, hired more administrative staff and executives with fetish for growth and shareholder value which caused a fundamental shift in incentives and they reduced themselves from an innovative tech shop to an ad selling business. Sad but common and as old as Jack Welch style capitalism where engineering excellence gives way to corporate greed.