5 ms·

Google has always been a fascinating company to me. A lot of my friends still work there and love it and my impression of their internal developer experience is

by mlazos 6y ago

Google has always been a fascinating company to me. A lot of my friends still work there and love it and my impression of their internal developer experience is that it’s top notch. This shows me though that despite them being an engineer-first company (or used to be) they really haven’t created another dominant business since ads and consumer gsuite. Basically when I was a new grad I thought if your company had technical prowess everything else would follow but I’m slowly learning that it’s not the case, sound strategy is really all you need and you need good leadership for that, which comes and goes.

toohotatopic 6y ago
Are they an engineering-first company though? E.g. the entire XMPP fiasco wouldn't have happened if engineers had made the decision. Which engineer would shut down an open protocol to fight a competitor instead of offering the better product? In the end, they lost the space to Facebook and now Zoom anyway.
- dub 6y ago
  Did XMPP support seamless conversation history and avoiding duplicate notifications across all of your devices?
- nemothekid 6y ago
  Which engineer would shut down an open protocol to fight a competitor instead of offering the better product Probably any engineer that wanted to implement cross device message syncing or end to end encryption without dealing with the politics of 1000s of other companies. Not saying what Google did, but I'm tired of hearing about this XMPP (and/or IRC) pancea when both had non-trivial UX issues.
- majormajor 6y ago
  Open != good. "Not Invented Here" doesn't just come about from anticompetitiveness or stubborness, sometimes there are real problems encountered - especially in edge cases or at scale - where trying to fix them through navigating open standards vs making your own thing that works better for you is a really viable option. See also Google's reinvention of things like build tools internally. Lots of open source versions, but they caused plenty of engineering pain of their own.
justicezyx 6y ago
The engineering-focused mindset is hilariously short-sighted in hindsight. When I was AWS 2012-2013, the sense that the Cloud is going to be huge got more and more evidences from customer feedback. When I joined Google in 2013, the sentiment was that Google can easily beat AWS. But, for all those talks, they have not mentioned any significant understanding on what customers expect from GCP, and why and how GCP can win. The idea was simply that Google's tech (and a self-complacent idea that Google engineers is somehow superior) is just better, therefore it will win. The vast majority of hard work in between was never discussed. Even later in 2016, I was pitching my Director to build a Cloud service that exposes Borg's experience (I worked at BorgMaster team then); the experience would be close to AWS Fargate (or Cloud Run). And the reply was that its technical complexity would be too high. Once again, there is no thought given to whether or not customers would be loving to use such services. Granted that, I did not have the PM expertise or connections to build a stronger proposal, but the focus on technical thinking is evident.
- Infinitesimus 6y ago
  > PM expertise or connections to build a stronger proposal, but the focus on technical thinking is evident Were you in a PM capacity there? It's a shame you couldn't find someone else who saw the vision of easier developer tooling being made widely available.
  - justicezyx 6y ago
    No, I am a SWE. I indeed later was reminded by a PM college that I should have been talking to a PM for supportive information. Now have been working in a small startup for a year. I definitely can see that I was lacking the expertise.
- hangonhn 6y ago
  Someone who used to be a Googler once told me: "A group of great engineers lead by a terrible manager will never succeed but a team of mediocre engineers lead by a great manager just might." I don't think anyone doubts Google has great engineers but I don't think it's enough for them to win. All you'll end up with is a lot of great products with no futures. I found myself in a similar situation in my previous job: a bunch of ex-Googlers and a weak leadership team. In the end, we ended up making a lot of components that went nowhere because no one would pay for them.

by mlazos 6y ago

toohotatopic 6y ago

Are they an engineering-first company though? E.g. the entire XMPP fiasco wouldn't have happened if engineers had made the decision. Which engineer would shut down an open protocol to fight a competitor instead of offering the better product? In the end, they lost the space to Facebook and now Zoom anyway.

dub 6y ago
Did XMPP support seamless conversation history and avoiding duplicate notifications across all of your devices?
nemothekid 6y ago
Which engineer would shut down an open protocol to fight a competitor instead of offering the better product Probably any engineer that wanted to implement cross device message syncing or end to end encryption without dealing with the politics of 1000s of other companies. Not saying what Google did, but I'm tired of hearing about this XMPP (and/or IRC) pancea when both had non-trivial UX issues.
majormajor 6y ago
Open != good. "Not Invented Here" doesn't just come about from anticompetitiveness or stubborness, sometimes there are real problems encountered - especially in edge cases or at scale - where trying to fix them through navigating open standards vs making your own thing that works better for you is a really viable option. See also Google's reinvention of things like build tools internally. Lots of open source versions, but they caused plenty of engineering pain of their own.

justicezyx 6y ago

The engineering-focused mindset is hilariously short-sighted in hindsight. When I was AWS 2012-2013, the sense that the Cloud is going to be huge got more and more evidences from customer feedback. When I joined Google in 2013, the sentiment was that Google can easily beat AWS. But, for all those talks, they have not mentioned any significant understanding on what customers expect from GCP, and why and how GCP can win. The idea was simply that Google's tech (and a self-complacent idea that Google engineers is somehow superior) is just better, therefore it will win. The vast majority of hard work in between was never discussed. Even later in 2016, I was pitching my Director to build a Cloud service that exposes Borg's experience (I worked at BorgMaster team then); the experience would be close to AWS Fargate (or Cloud Run). And the reply was that its technical complexity would be too high. Once again, there is no thought given to whether or not customers would be loving to use such services. Granted that, I did not have the PM expertise or connections to build a stronger proposal, but the focus on technical thinking is evident.

Infinitesimus 6y ago
> PM expertise or connections to build a stronger proposal, but the focus on technical thinking is evident Were you in a PM capacity there? It's a shame you couldn't find someone else who saw the vision of easier developer tooling being made widely available.
- justicezyx 6y ago
  No, I am a SWE. I indeed later was reminded by a PM college that I should have been talking to a PM for supportive information. Now have been working in a small startup for a year. I definitely can see that I was lacking the expertise.
hangonhn 6y ago
Someone who used to be a Googler once told me: "A group of great engineers lead by a terrible manager will never succeed but a team of mediocre engineers lead by a great manager just might." I don't think anyone doubts Google has great engineers but I don't think it's enough for them to win. All you'll end up with is a lot of great products with no futures. I found myself in a similar situation in my previous job: a bunch of ex-Googlers and a weak leadership team. In the end, we ended up making a lot of components that went nowhere because no one would pay for them.