[ible]

It must be uncomfortable being a google engineer on these products.

Having your work cancelled and thrown away is one of the most disheartening things as a developer.

It has to be back of mind for them when they are building. When it gets released they get to come see all the comments better on how long until Google kills it, and the people who won’t try it because of that risk.

[renewiltord]

Interesting. I don't think everyone shares that perspective. Certainly I think this software is sort of like an oil exploration expedition. It's going to cost a lot of work, there's a lot of stuff on-site you can't really reuse, and at the end you sometimes fail to find oil. But you built the thing to search for the oil, not to get the oil.

Shell spent billions on trying to find oil in the Chukchi Sea and there isn't enough to be worth it. But if you don't do things like that you don't find the other big finds.

That's why you don't fall in love with the software. It's there to see if the product is viable. Honestly, I like that environment.

[rurp]

But in Google's case they often do find oil, but then just dump it in the ocean. Lots of people used and enjoyed Inbox, Hangouts, Reader, etc, etc; but Google just discarded them anyway.

Spending years building something that thousands/millions of people end up using and liking, and then seeing it killed anyway, has to be discouraging.

[renewiltord]

Honestly, that's the dream. If you have product market fit, Google paid for the search, and now you can go make it later because it doesn't fit into their vision? Amazing. Ready-made startup.

[TrainedMonkey]

That is hearsay, but I heard that career growth and compensation at Google is heavily tied to hitting milestones. Largest possible milestone is launching things, so there is bottom up pressure to launch products... regardless if they succeed.

[joecot]

Right. The point of the oil exploration expedition is not whether you find oil or not. It's that at google, running an expedition gets you a promotion, whether you found anything useful or not.

[vernie]

Having your work cancelled and thrown away without any change to your compensation is greatest relief you can experience as a developer.

[dyingkneepad]

About 6 years ago I took ownership of a very complex subsystem at work in order to fix a bunch of issues. Things started working after I fixed them, so I moved on to other things. No one else stepped to maintain it, so work done on it was minimal since then. Bugs kept appearing (the rest of the system changes a lot), and I still get emails with bug reports and/or people trying to understand the code asking me for training or even worse: patch review requests. I'm not even in the same team anymore!

[Traster]

Here's a magical phrase "Sorry I really don't have time to help you with this". As much as it's an asshole move, the people asking for your help are being paid to work on it and you're not, so unless you're trying to garner brownie points for mentoring, you need to be clear about your boundaries.

[always_left]

So... a startup?

[ilikehurdles]

Except, the engineers get paid adequately.

[renewiltord]

Sure, it's a trade-off in what sort of upside and control you want.

[summerlight]

The reality is exactly the opposite, at least for less senior (< directors) engineers in Google. You get a promotion (even double) for launching the product and don't have to worry about its technical debts. This is because those less senior employees don't have a real product ownership and why they focus much more on launching a product than its growth. What Google needs to fix is this incentive structure.

[notatoad]

if you're working on a mainline google product that gets shut down after being hyped as the next huge thing that everybody will be using then yes, that must be disheartening. Being a random developer on something like Allo would have sucked.

But my understanding is that these area 120 projects are built with more of a startup attitude, that you're taking a risk and there's a high probability of failure. It sounds really fun to me to build a "pure" product, being able to create exactly what you think the product should be without having to concede to business requirements, and just see how your vision will be accepted.

[emerongi]

Marketing & sales make products succeed though. If it's just an engineering team chipping away, the probability that it will get mainstream traction is close to nil.