Comment by neilv | Hacker News Reader

Comment by neilv | Hacker News Reader

Parent Full thread

> I've never encountered anyone else who combines being an incredible engineer, and being an incredible product person, and an incredible manager, and an incredible designer.

> As the PM on Google Maps, he was frustrated that the JavaScript was really slow. [...] So, one weekend he said, “I'm going to rewrite it”. And he rewrote the entire thing, making it 10 times as fast and a third of the size. He completely threw away the code that a team of engineers had been working on for months.

Maybe it wasn't as bad as that sounds, but the situation was something like the very ordinary... A team of engineers had been working on a lot of aspects, and optimizing was intentionally put off. Maybe there were also some suboptimal coordination or project management going on, which was hurting the team's effectiveness (which is something a product manager would speak about with the project manager or tech lead). Maybe he did it quietly, as a proof of concept, to confirm his intuition that the team was having problems, and as "data" to quietly show team leader that something is wrong. Or maybe he had a prior OK/encouragement from the team to rewrite the code they'd working on.

Alternatively, maybe it was as bad as it sounds -- a PM From Hell, pulling a move that engineers would be WTFing about for years -- but his "incredible" skillset included also being an incredible politician, so team morale wasn't destroyed.

Regarding the morale of teams who care deeply about their project, and WTFing about some management move for years later...

I was interviewing for a principal-level role at a startup doing something exciting, and one of the engineering leaders told me a story about when they'd been at one of the legendary earlier computer companies. The company had done some WTF thing involving pitting two engineering teams against each other, with awful incentives. (I had a WTF response, and articulated why, and maybe that was a factor in them making me the offer.)

There are some morale-destroying engineering/project management moves that people will still be talking about for decades, and which many experienced engineers and managers who hear about it will immediately understand (to some degree) as awful.

(As I said in previous message, I suspect that the GMaps move wasn't as bad as it sounded, and hopefully it wasn't one of those WTF stories.)

I’m not sure exactly what the PMs role would be in regards to software quality, but from the sounds of it, the engineering team was poorly lead or had unhealthy practices (which likely means it was poorly lead). Which isn’t too uncommon in software engineering mind you, but if that’s not the responsibility of the PM then I don’t really see what is wrong with what Bret Taylor did.

If you could improve my code like that, I would welcome it and learn from it. The only people I think should feel “wronged” in that situation is the people who were responsible for an entire team of engineers not being put to better use. Of course unless Bret Taylor worked on that part or the issue, then whatever lead to the production of bad code was never really solved at google maps, but again, unless that’s the PMs area of responsibility then well…

"Alternatively, maybe it was as bad as it sounds"

I'm sorry but in what world is improving the performance of the product you're working on 10 fold "bad" ???

Yes, it was probably a shock for the team. But those kinds of moment are also a huge opportunity to realize there was something wrong with the way you were working.

I've had a senior engineer erase the entirety of my code once in front of my eyes and rewrite it from scratch in a few hours, still in front of me. Well, i still joke about it with friends that experienced the same thing from the same guy, but it was definitely the kind of things that motivates you to improve the quality of your work.

Of course, the senior guy was a friend and it was all done with smiles and laughs. He didn't make it sound worst than it was.

We don't know that the team was actually dropping the ball.

But even if they had been, addressing it in a way that sounds like you're saying "you clowns, I can throw out your months of work, and do it myself 10 times better in a weekend" (whether or not that's accurate) from the PM... sounds like a good way to ruin morale of a team, and seed a toxic engineering culture throughout the department/company.

« Whether or not that’s accurate » is not a minor detail. Some people need to be told when they’re doing shite work, otherwise how can they learn ? The forever supportive culture has some limits and sometimes a bit of spanking can be a great boost, and help you make the required radical changes in your life that you need.

You do ask - if the performance was not great before with a team of 10, what was the culture like before? Sometimes, the only way to improve matters is to build up a new team and sideline the previous one where it can't do more damage.

I have no idea what the background is in this particular case; but where I work I‘m sick and tired of people „improving performance“ of systems which are not critical at all, at the cost of code, architecture and infrastructure complexity. So sometimes it can be bad.

Reply on news.ycombinator.com