5 ms·

but then there aren't many places where you can basically work whatever you want to on. Take Allard example, he has worked on TCP/IP protocol in Windows, Intern

by awa 16y ago

but then there aren't many places where you can basically work whatever you want to on. Take Allard example, he has worked on TCP/IP protocol in Windows, Internet explorer, Xbox, Zune, and now courier.

j053003 16y ago
This is why I prefer startups.
- hello_moto 16y ago
  Can you do anything you want at someone else's startup when startups are supposed to be frugal and focus?
  - MartinCron 16y ago
    At someone else's startup you're often required to do a variety of different things, which provides the sort of intellectual stimulation that extremely differentiated larger companies lack.
    - hello_moto 16y ago
      Yes, variety of different things that most of the time didn't yield the best solution. Pardon me for being a bit sarcastic. It is what it is.
- j053003 16y ago
  To clarify: This is why I'd prefer to start a startup. You might not succeed, but at least you're working on the project you want to work on.
  - hello_moto 16y ago
    Makes absolutely perfect sense. Working at someone else's startup sucks unless you're given full freedom; which is rarely the case.
mwerty 16y ago
You can basically be assigned to any team you want but the leadership generally does not let you do work except in a few teams (CLR, Research, etc.). There are a few reasons that struck me during my time there - all related to perverse incentives: 1. There are too many PMs. You could fire all the engineers and the PMs would keep holding meetings for a couple of years reshuffling priorities, co-ordinating interactions across teams and "making hard feature cuts". The last thing is especially comical: The dev team would be behind schedule and the PMs would be very proud that they are macho enough to drop certain features in order to ship. In spite of this, I saw some very good people become PMs instead of devs since that's the only way you get any say in the direction of the product. They all end up spending an insane amount of time with powerpoint decks for shallow reviews by upper management. 2. Secondly, the split between test and dev orgs leads to some perverse incentives. As a dev, you are looked upon slightly favorably (though not goaled on) for fixing bugs that you created. 3. The mid-level leadership is not hungry/driven enough. Virtually all the mid-level people running teams have little to prove after they make partner. Its the standard principal agent problem where the agents are too risk averse. These factors make it hard for people (especially engineers) to feel like they are making a difference.
- awa 16y ago
  The dev team would be behind schedule and the PMs would be very proud that they are macho enough to drop certain features in order to ship Whats wrong with the above?
  - mwerty 16y ago
    A large part of your performance as a PM is measured on how much BS you are capable of generating in order to manage perceptions - which is natural since you do not have authority over engineers. This is one instance of that. Obviously, teams differ so you might have a different experience.

by awa 16y ago

j053003 16y ago

This is why I prefer startups.

hello_moto 16y ago
Can you do anything you want at someone else's startup when startups are supposed to be frugal and focus?
- MartinCron 16y ago
  At someone else's startup you're often required to do a variety of different things, which provides the sort of intellectual stimulation that extremely differentiated larger companies lack.
  - hello_moto 16y ago
    Yes, variety of different things that most of the time didn't yield the best solution. Pardon me for being a bit sarcastic. It is what it is.
j053003 16y ago
To clarify: This is why I'd prefer to start a startup. You might not succeed, but at least you're working on the project you want to work on.
- hello_moto 16y ago
  Makes absolutely perfect sense. Working at someone else's startup sucks unless you're given full freedom; which is rarely the case.

mwerty 16y ago

You can basically be assigned to any team you want but the leadership generally does not let you do work except in a few teams (CLR, Research, etc.). There are a few reasons that struck me during my time there - all related to perverse incentives: 1. There are too many PMs. You could fire all the engineers and the PMs would keep holding meetings for a couple of years reshuffling priorities, co-ordinating interactions across teams and "making hard feature cuts". The last thing is especially comical: The dev team would be behind schedule and the PMs would be very proud that they are macho enough to drop certain features in order to ship. In spite of this, I saw some very good people become PMs instead of devs since that's the only way you get any say in the direction of the product. They all end up spending an insane amount of time with powerpoint decks for shallow reviews by upper management. 2. Secondly, the split between test and dev orgs leads to some perverse incentives. As a dev, you are looked upon slightly favorably (though not goaled on) for fixing bugs that you created. 3. The mid-level leadership is not hungry/driven enough. Virtually all the mid-level people running teams have little to prove after they make partner. Its the standard principal agent problem where the agents are too risk averse. These factors make it hard for people (especially engineers) to feel like they are making a difference.

awa 16y ago
The dev team would be behind schedule and the PMs would be very proud that they are macho enough to drop certain features in order to ship Whats wrong with the above?
- mwerty 16y ago
  A large part of your performance as a PM is measured on how much BS you are capable of generating in order to manage perceptions - which is natural since you do not have authority over engineers. This is one instance of that. Obviously, teams differ so you might have a different experience.