Hacker News new | past | comments | ask | show | jobs | submit login
A Decade at Google (sigmod.org)
202 points by dwenzek on Sept 23, 2015 | hide | past | favorite | 74 comments



Just wanted to highlight a very important advice on the article:

You do not reach [work life] balance by reducing work. You reach balance by finding a passion that draws you out of work.


My employer technically owns anything I create. Even outside working hours. They have even won cases where an employee patented an idea after leaving the company. The court granted the company the patent. Talk about screwed up. Obviously I'm not in CA but am in the US.


Leave.


Watson is cool and all. Don't get me wrong.


We did have several people leave before joining because they refused to sign the agreement. I know that it's my fault I signed the agreement. I'm also working towards your suggestion (38 applications so far this year) but the thing that scared me was the fact that they got ownership of something after the employee left. I can see if it was something they worked on while with the company and then left and tried to patent it. If it was something completely unrelated then I just lost faith in the legal system. I'm also probably risking being fired just by posting something like this, shrugs.


I think its technically only things "related to your day job" that are covered.


In California that's (essentially) true. Other states permit far more draconian limits on employee freedom.

He says he's from outside CA.


It's kind of a grey area. The court docs show that they have won. I don't know if they were directly related to the persons daily responsibilities. I think there was some wording about any thing that competes with the company but the company is in basically every market. Honestly, I just weigh the risk of weather or not I think they would actually even care about what I'm working on. I know we have people who make mobile games on the side and no one really cares.

If I'm not directly competing and stealing clients I think that chances are low that they will try to claim ownership.


Not sure I understand this advice. How to explain to my employer that I'm going to spend less time working because I have better things to do?


If you feel that your employer knowing about outside passions is going to be a problem then you need to find another employer.


The same way you explain to your employer that you have defined working hours, and will be working within them.


Thank you. I was reprimanded once for essentially "only" being at the office for the defined time we agreed I would be there. I spent some time racking my brain until I decided I wasn't being unreasonable.


There is this hyaundai commercial ( https://www.youtube.com/watch?v=UZ8Klm8vPlI ) that hits home for me. "When did leaving work on time become an act of courage". It's so true.


I've always found it humorous that the clock hits 6:00 rather than 5:00. "Leaving work on time is an act of courage, but let's not get crazy!" Presumably he courageously started at 10:00 or took an hour away from work for lunch.


It's more about the appearance of working vs actually having work to do. I used to come in at 7:30 and leave at 4:30 because my commute was terrible other wise (2 hours each way). The few nights I stayed until 6-7pm I noticed the engineers that came in at 10-11am were all gone by 5:30pm but to most managers it looked like they were "working harder" than me when I left at 4:30.


Where I worked, official hours are 8.30 to 6. In Asia of course.


Is the official work week 47.5 hours?

In the U.S., while I think it's true that people tend to work closer to 50 hours, we pay lip service to the "40 hour work week".


Unreasonable employers happen. You have a variety of options including try to convince them to be reasonable, accede to their demands, try to be sneakier, or find a new job.

In our industry where it tends to be an option for us, "find a new job" might make sense.


I think his advice was for people who are so excited about their work that they cannot stop thinking about it on even Saturday evenings.


Don't let people decide if your other things are actually "better things to do", certainly not a boss, they're hardly objective. Simply state that like work, they are obligations you can't miss. I frequently have artistic obligations in the evening, and I make sure everyone knows they are unbreakable.


The work life balance is more of a necessity then a balance. You work to live while remembering to make room for the later. True balance is when the need to work no longer exists.


Or you just give up and code with a baby on your lap. Bigger problem for me is a work/sleep balance.


Babies stay on laps? What? Oh I remember now.


How does that play out when you're a parent with a spouse and 4 children?


The spouse and 4 children are part of the passion outside of work? :)


I have heard once you have children, you only have time for one hobby. True?


The best advice I was given when my daughter was born was to not only put away hobbies, but to remove them from view. Suggestions I was given where things like: packing up the game console and games and put them in the closet, donated the pile of books that I was never going to read, and clear desks/shelves, etc that contained unfinished projects.

The first few months there wasn't time for any of that anyway and then as I began to have time I took out the few items that I really wanted to use, but all the while I didn't have a constant pile of fun things around the house making me feel bad because I didn't have time to play with them.

Depending on the circumstances you will have different amounts of time, but most importantly is realizing that you will have less time and taking steps to adjust to your new role.


No kids yet, but this rings true. The burden of hobbies I'm not working on sucks more time than actually sitting down with a single hobby.


Very small children are mostly loud meat slugs and easy to care for while you're playing games. Then there's a phase where they're very active but not potty-trained or good at talking.

The next phase, when they are a little older, you can share your hobbies with them. Kids love to do fun things like hiking and camping and building stuff and playing games. That is awesome.


Not for me. I had 3 children; also did Scouts (involving the children), wrote (1 hr every morning before everyone else woke), moderated a game club (every Thursday 4:30), visit a shooting range. Once the children are older, it accelerated of course.


I assume you had to have good time management habits to pull it off?


Scouts was always a do-everything-at-the-last-minute thing. You just have to be present for the kids, that's the most important thing.

Regularly scheduled hobbies are the easiest to pull off. If you just try to find a spare hour, you never will.


How did you balance that with your job? Do you have a normal "9-5" or "10-6" type job? I always wonder how people carve out time for stuff with family and such in the early evening on weeknights when normal leaving times + commute mean I likely wouldn't get home till 6:30 at the earliest.


I told each boss "I work with the Scouts. I camp one weekend a month, and will leave work early on that Friday. The day will come when there's a rush at work, and I really need to stay late and get something done instead. On that day, I will go camping with the Scouts."

Being upfront about it like that got me a lot of credit somehow. It was never a problem.

Also, when they were around 12, I was a partner in a consulting firm. So I could make my own hours. That definitely helped out during those years.


Thanks for the details. I'm fortunate to work in a place where this is largely not an issue. I work hard and often put in time at night and on the weekend because I'm Type A, so I don't worry when I need to take time off because I know I've put the time in and deliver results.

That said, I've definitely worked at companies that want to have their cake (in by 8:30 or 9) and eat it to (leave no earlier than 5:30-6 with expectations of frequent late nights and weekends with 2 weeks vacation). I've simply lost all tolerance for that, and always wonder how others manage it in stricter environments.


At my "generic" workplace, core hours are 9-4. Everyone is responsible for getting 8 hours in a day, so a lot of people take advantage of the spread by coming in at 7 and leaving at 4 (w/ hour lunch break).


It depends on their age. When they're babies, you won't have time for sleep. It tapers off as they get older and go to daycare/school/etc.


Being 14 months into the experience, I'd say that you don't have any time for hobbies at first, but you're surprised at how much you don't really care. My family is my passion outside of work. I don't think of taking care of my daughter as work the way I would have thought about babysitting someone else's kid before I became a father.


As my kids got older and started playing sports, I eventually got to the point where I am now coaching one of their soccer teams. It has become a real passion, and I also joined a team myself (in my younger days, pickup basketball was more my thing). My wife helps coach, and she just started a soccer clinic/pickup group for Moms.

So I guess from my perspective, I would say your passions and hobbies and interests can be molded by having children, and become something you share and develop together as a family. From an outside perspective, we probably devote too much time to this sport (I would have thought so a few years ago), but it's something we all enjoy and have found ways to participate in together.


In that case you decide which is more important. Work or taking care of your family. Now I know having a job is part of the taking care of your family thing but you have to honestly weigh "if I leave work on time am I actually going to get fired". The answer is usually no.


Time machine?

Or find something you like to do with your kids.


From the OP: Of course, family comes first on this ladder


That's the same damn thing.


> That's the same damn thing.

Not really; a side effect of finding a passion that draws you out of work is that you reduce work but simply reducing work is not the same thing.


I understand that he was trying to be clever and thought provoking, but it was neither. Reducing work just by itself is a good thing. If someone doesn't have any passions, does it make sense for them to work 12 hour days in the interim? I don't think so.


It really isn't the same. If you have a habit you're trying to break, abstention is the most obvious tool and everybody goes for it (and everybody faults themselves or each other when it doesn't work). If you've ever heard the term "white knuckling", that concept can often describe this sort of situation. "I'll just suck it Up and stop"

But many self improvement advisors across a host of problems from overeating to addiction agree that the way to kick a habit is to crowd it out of your life. You don't have time to eat a cheesecake because you're going for a run or to a volunteer event or to the park with your kids.


This guy says that he's in 'Google Research' and talks about job titles like 'junior researcher'. Is there or is there not an actual Google Research organisation, similar to Oracle Labs or Microsoft Research? With people doing full-time research and publishing papers?

I was under the impression that their policy was 'you can do research anywhere in Google' and that there was no separate staffed labs doing more academic research. You may get research in the V8 team, or the systems team or whatever, but they were doing research as part of product teams. At least that's what I was told when I applied.


So, the answer to your question is really two fold:

Google Research used to be pretty separate. That is, it was a separate product area from other product areas. It is no longer. This is an organizational issue however, and while it would seem like it mattered, it did not. It only changed who was the SVP overseeing it at some super high level. The VP was the same for many many years, and the person really in charge.

But this is irrelevant to the second part of your question, where you mention: "and that there was no separate staffed labs doing more academic research."

They do not do "more academic research", so in that sense, your recruiter was right.

Research at Google is not about academic vs industrial. The practical difference between the research scientist and SWE ladders is that in research scientist, publishing is valued as well. Outside of research, it's certainly nice, but not really part of the job description.

However, unlike most other places, publishing alone is not the goal. If all you ever did was research and publish, you'd be fired :)

Google expects research scientists to do real coding, real work, and be as good of SWE's as their SWE's.

It is a very hybrid approach, and different from most other places.


I'm interested in learning more about the research ladder (and how it's different to the SWE ladder) at Google.

So I found this: https://static.googleusercontent.com/media/research.google.c...

If anyone has other interesting resources about the general employee structure at Google/other big companies, please reply :)


and as for sourcing (someone asked privately), my group was under research and reported to the research VP for many years. I have been on promotion committees for research scientists and SWEs in google research, etc, so at least theoretically, i either understand what they are supposed to be doing, or have been screwing it up really badly ;-)


Yes, there is a google research department. Research occurs there and also in other parts of the company. It's not as well defined as MSR, though- the research department is fully embedded within the company's products. This is not mutually exclusive with "full-time research" although I don't think that sort of distinction really matters much.


What a strange comment. It almost sounds like you're accusing him of lying.

Anyway, it's pretty easy to verify: http://research.google.com/

And the guy's profile: http://research.google.com/pubs/author1112.html

Edit: Don't know why I'm getting voted down. The OP was questioning whether there's a research group at Google, in the comment section for an article by somebody who's worked in that department for ten years. If that's not a definitive answer to the question, I don't really know what the OP is expecting.


> Edit: Don't know why I'm getting voted down. The OP was questioning whether there's a research group at Google, in the comment section for an article by somebody who's worked in that department for ten years. If that's not a definitive answer to the question, I don't really know what the OP is expecting.

It sounded like an innocent question to me and it was one I also had. While there is a research site it still wasn't entirely obvious to me how it worked. You're probably getting downvoted for the lying bit and being condescending.


Your question was slightly aggressive and confrontational. Only slightly and I don't know that you intended it to be so.


I think you have me confused for the other person though I don't see how it could even sound like that ¯\_(ツ)_/¯


yes, i should have replied to "chriseaton".


Google has previously made it look like there is no Google Research "even in areas where there is a much higher proportion of research to engineering, the “Research Team” we have established is not as formally separate from engineering activities as those in other organizations" so I thought it wasn't a separate group you could be on the staff of.


Alon Halevy is pretty big in the field of data integration. Some of my undergrad project was about improving certain aspects on his work on LAV integration. You might also find his work by Alon Levy instead of Halevy. If you want a theoretical view of the field, checkout Maurizio Lenzerini's work.

For a perspective/survey check this out: http://www.dis.uniroma1.it/~degiacom/didattica/semingsoft/ma...


I found the part about coding interesting..

"My coding activities were always productive. They either launched a new project or a major change in direction of a project. They also enabled me to have discussions with my team members at a completely different level of detail (to everyone’s enjoyment)."

This can be great. It sounds like it went well. However, I have personally seen a downside, some potential risks, when the boss "keeps coding".

Here's why - if the boss is coding but staying off the critical path, this means that the often detailed, difficult implementation work will be delegated to developers who report to the boss. The old adage that genius is 1% inspiration, 99% perspiration? Well, the critical path is the perspiration part. Deadlines, estimates, glaring errors that shouldn't have happened and need to be fixed late at night or over the weekend… these are all part of the critical path. While coding can keep a boss closer to a technical team, it can also create a dangerous illusion, that the boss is still "technical" but insulated from the truly difficult aspects of a technical role. There is a big, big difference between coding on your own research project and coding under deadline pressure for a system that needs to work. This is only a risk, not a certainty, as a good manager will recognize this.

Perhaps a greater risk, though, is that if the boss gets to play around with new technology and launch projects for the rest of the team to finish off, there's a risk that the boss is pretty much eating the frosting of the team's dessert but making everyone else eat their vegetables, so to speak. If you're a manager who codes, keep in mind that autonomy and innovation may be key to the motivation and job satisfaction of the people working for you. This can lead to a very damaging management anti-pattern, where a manager who "keeps coding" essentially uses managerial authority to intercept and filter projects, make the key technical decisions, do the interesting bits that would lead to interesting conference talks, and then delegate the work of actually finishing the product to the team. This may deprive senior level developers the autonomy that they really were supposed to have on projects.

Once again, I want to emphasize that this isn't necessarily what happened here, in fact, it sounds like it wasn't. But the passage made me a bit nervous, since I consider this to be something of a management anti-pattern.


A manager who doesn't continue to code or work on the day-to-day actions may fail to keep up with necessary technical knowledge, effectively reducing their ability to manage.


The problem with a boss coding is that they don't dedicate enough time to management. Connections, management theory and functional knowledge are important.

Instead of a boss who is part-time manager, have we tried managing 3/4 of the year then doing a "developer rotation"? How does that work?



Down for me, but is it just me or is Google's cache kinda broken? When I click the cache link I get a blank page, and then "waiting for wp.sigmod.org..." at the bottom, which hangs forever. I guess it hits the live site if it doesn't have it in the cache?


You can get the text-only cached version here:

http://webcache.googleusercontent.com/search?q=cache:WpRJJJ-...

The normal cached version still tries to download images and whatnot from the original server.


Cache page doesn't hang forever for me, had to give it about 15 seconds; seems like it's trying to contact the page for images and eventually gives up and shows you the cache. Clicking "Text-only version" made it load instantly for me (there's not anything of substance but the text anyway).


Good tip for balancing work and life - just find other interests that pull you away from work.


Some sound advice. Having worked with academics before, the ones who were most successful were able to switch gears between individual and team goals as he describes.


> We then could inspect the query stream to estimate how many queries might be well served by HTML tables.

Hmm.. humans are inspecting the query stream?


Yes. Click here and you'll be one of them: https://www.google.com/trends/hottrends/visualize?nrow=5&nco...

There's lots more data over here: http://www.google.com/trends

(I don't believe the author was referring to this, I think he meant something far less interesting.)


Obviously not, except for debugging. Scoring logic is most likely where the "secret sauce" of the thing they did is hidden.


Congrats for completing a decade and for your achievements !!


> (or if you cannot code, get involved in code reviews).

This is the worst advice ever, esp. if you don't know how to code and don't have code readability for the language being used in the code review. Learn from code reviews, see what other people are saying, but don't comment on a fix unless you found something glaringly obvious.


I'm a C++ developer at Google; I've been writing software "professionally"[0] for about 16 years, including a fair chunk of C, but no C++ until I reached Google. So, arguably I started in a position where commenting on code reviews was a bad idea.

Nonetheless, I commented on LOTS of code reviews when I joined, but almost always with questions asking (usually) 'why is this done this way?' or (rarely) 'could this by done this other way?'. It helps that our code review tool lets you mark comments as 'no action required'. This allows the first form to be a genuine question without judgement. I would usually learn something, and sometimes the original author would revisit a decision and improve the code. I reserve the second form for when I'm pretty sure I think what I'm suggesting is an improvement.

In my experience this allowed me to provide feedback without perceived ego or offense[1], but also accelerated the rate at which I came up to speed on both the language and the details of the project I'd joined. Despite being a few years in now and fairly familiar with our project, I try (although sometimes fail) to maintain this style today.

[0]: My father got me my first development job as a subcontractor building terminal emulation software for the Pennsylvania Turnpike Commission at 16. I'd learned C a few years beforehand on QNX. A number of other jobs from the same folks followed over the years while I finished up high school and undergrad.

[1]: Usually, anyway. There are some people who take ANY comment, question, or critique as an accusation. Thankfully I've encountered only one of these people at Google.


I interpreted this section as saying something more along the lines of "if your job prevents you from engaging in coding as part of your day-to-day activities, but you otherwise know how to code, then maintain involvement with coding by participating in code reviews."

It seems that his intended audience is expected to know how to code.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: