Hacker News Reader

Comment by sz4kerto | Hacker News Reader

I don’t think that what you’re doing is very common, but it’s not uncommon either. What you need to do this is to work in an environment where not being productive is not harmful. There are plenty of environments like this. For example, not doing any work but still getting paid doesn’t do any harm at Google. The company is making money anyway. At a small startup that’s under immense pressure people would realise what you’re doing because you’d harm the business.

I’m a big believer in Price’s Law - the square root people produce half the value. If there are 10000 people at the company, you can very well survive in the group of 9900 that does the other half.

View on HN

I’ve worked at a small startup and find this to be untrue. If you’re working with people 10-100x worse than you, you can be sandbagging as hard as you can and still be miles ahead and look like a paragon of productivity.

Now, will that particular startup succeed? Probably not in the way their investors would like.

As someone who worked at both Microsoft and Google... you can get await with doing little work at those companies for maybe a year, two years at most, but after that you're gone. The expectation at those places skyrockets after the first year.

hypothetically, do you think if people weren't allowed to drink caffeine or use performance enhancing drugs (adderall etc..), what % would not make that threshold of performance?

No idea; most people I worked with drank coffee but never once was I exposed to performance enhancing drugs. At Microsoft I worked on IronPython and the people I worked with were productive not in the sense that they churned out 1000s and 1000s of lines of code a day, but they were productive in the sense that they were constantly outputting fairly novel solutions.

A similar thing was true at Google, I worked in the platforms division there (storage systems, BigTable, Linux kernel tweaking), and it was mostly about identifying bottlenecks and coming up with interesting solutions to squeeze out more performance. Most of this work wasn't like massive lines of code, but you had to really understand the problem at a deep level and have the kind of rigor and discipline to make changes to large systems without breaking anything in the process. That last bit was very hard to do, since coming up with optimizations that don't break anything is surprisingly difficult.

I’ve never mentioned my ADHD diagnosis and medication to my coworkers and none have ever mentioned it to me.

At Microsoft at least that wasn't my experience at all. I knew several people there who hadn't done any real productive work for the last 7-10 years, probably longer.

I've been in Azure and GCP. IDK how anyone would get away with it. You constantly have customer issues, people from other teams pinging you, needing help with something. I don't see how it would be possible just to do nothing and survive.

Maybe in some team that is further removed from customers?

Aren’t there still tons of developer jobs at Google, Microsoft, Intel, Oracle, etc. where you aren’t on call and just get to write code?

Not to disagree with your experience at Azure and GCP, but the grandparent specifically mentioned their parent entities: Microsoft and Google, each of which have several divisions apart from their respective cloud divisions.

Keep in mind that Azure and GCP are playing catch up to AWS, so they necessarily have to run a tight ship in order to close AWS’ lead.

So its possible for laxity to exist in other divisions in a large company, as long as the money keeps rolling in, e.g. the Windows division at Microsoft (which still enjoys a good share of the desktop market) and the Search division at Google (which still enjoys a good share of the search market).

Yeah, I mentioned Azure and GCP specifically, to imply that other divisions within those companies may be different. Anything where you're constantly deploying to enterprise customers, you're going to need to be responsive to issues that come up, and that alone will take a good chunk of the day.

At Google anyway, it's fairly common knowledge that cloud is a far more difficult place to work, both in culture and expectations, than most anywhere else in the company as a SWE.

That's surprising because I came from Azure, and Google Cloud is way slower than Azure was. Frankly I think that's why I failed at the job (separate thread). It feels like a "me too" offering, and the initiative and creativity just isn't there, and I just couldn't be bothered to care. I felt so much more energy at Azure (even though they're the ones most often criticized as being "me too").

That said, it may have just been timing or team fit or seratonin levels or whatever. I'm sure plenty of others have had the opposite experience.

It's easy to be productive in that context. People pinging you to get help with something, that's a "positive distraction." You'll write a block of code super fast and send it back to them.

As someone who put more lines of code into <50KLOC hobby project, that is not guaranteed, especially when your "little work" affects the bottom line on megabucks scale.

Love the async-ian slip

It also depends on the company. I've worked at companies where the #1 project can change weekly and by that I mean the previous project is completely abandoned for the new shiny. If you've worked at a place like this (there are lots of them), you have to not burn yourself out. It is very demoralizing to give your all to a UI just to have it literally thrown away. Sometimes slow rolling can let you move things forward in a protective way. If it is still #1 after 5 weeks, maybe it is time to dive in more fully. Digging holes and filling them back in is tough.

With my current manager this has become a skill I'm honing. He's senile and asks for all sorts of things he will have forgotten about in a few days. So I slow roll any new idea he gets excited about until it has percolated for at least a week or two. The trick is figuring out how to wait long enough to avoid doing throw-away work without waiting long enough that he feels ignored. But, like I said, he's senile, so that helps with the latter.

For my previous boss, I have the rule of 3. Until he asked for something 3 times, I just said "yep" and didn't start it. If he asked three times, he wasn't going to forget so time to start moving it forward.

> He's senile

Figuratively speaking, right? (Or did you mean literally? Dementia? If so, is he old)

Different formula but I've always liked:

productivity = (time * effort)^talent

Just as a general model.

let talent=0 then productivity = (time * effort)^0 = 1 = 100% Sounds about right

Yup! Time and effort absolutely count.

But the "star line" folk out of Heinlein's Beyond this Horizon are the ones that catapult the entire species forward.

The Wright brothers, the Norman Bourlags, the Nikola Teslas, the Alan Turings.

What about charismatic poor decision makers, pulling everyone in the wrong direction? Or evil geniuses

Maybe multiply with sign(talent)? And the exponent could be abs(talent):

    sing(talent) *
       (effort * time) ^
                 abs(talent)

I’ve had coworkers like op almost in every team that I worked for and google was one of the only places where some folks like that got walked out so may not be the best advice ;)

Reply on news.ycombinator.com