4 ms·

| I've found that advocates of WFH are mostly motivated by that's what they want to do more than anything else. I've heard endless people state that WFH produc

| I've found that advocates of WFH are mostly motivated by that's what they want to do more than anything else. I've heard endless people state that WFH productivity is through the roof with absolutely no evidence. And lets not pretend that we can accurately measure productivity of software devs - we've all been down this road with agile.

ARandomerDude 6y ago
Perhaps if we measured lines of code... ;-)
- nchelluri 6y ago
  > If I had more time, I would have written a shorter letter.
aswanson 6y ago
Can only speak for myself. Im far more productive without an hour commute, people constantly interrupting me in my cube with trivia, interrupts to explain you can't do X in the code because the repercussions are X^10, etc followed by a long commute home. But hey, to each his/her own.
ponker 6y ago
Here are some numbers... I work at a FAANG and 28-day commits in my division (~5K engineers) are down ~3% from February. Not an exact science but some real data I just pulled for you.
- Ankhers 6y ago
  This is also a fairly bad time to try and put a measurement on how effective WFH is. The fact that schools and daycares are currently closed means that parents are having to take care of their children while also working. It is also hard on people that are used to working in the office to be told that they now have to work from home suddenly. I'm sure a non-trivial amount of people did not have a space for themselves to be able to work and they are trying to figure things out as they go. Not to mention figuring out how to communicate properly while everyone is remote. There is also not a one size fits all solution. People will be more productive being where they are more comfortable. That may be at home, in a coffee shop, at a library, in the office or any other number of places. People are expected to work from home right now, so a lot of those other options are out of the question. TL;DR, don't judge work from home when everyone has to do it during a pandemic.
- wtetzner 6y ago
  What are "28-day commits"? Are you talking about commits to source control?
  - ponker 6y ago
    Commits over the trailing 28 days.
- gregmac 6y ago
  It's an interesting data point, but hard to judge much by it. Two things: 1. Working-from-home-during-global-pandemic-with-kids-and-spouses-and-roommates-and-not-ideal-setup isn't the same as working-from-home. Undoubtedly some people are less productive today. 2. Commits isn't necessarily a great proxy for productivity to begin with. For example, with greater focus comes less "oops I missed a unit test and broke the build" type commits, or even just more time to interactive rebase and clean up commits before pushing. "Go slower, you'll be done sooner."
  - ponker 6y ago
    My point was that in my org, WFH has been staggeringly successful. 3% drop in the middle of the biggest global crisis in living memory means that there will likely be no loss or a gain from WFH once people get their emotional states back to normal and their kids back in school.
- pmiller2 6y ago
  What are the stats per engineer over the past year or 2, including standard deviation? I suspect +-3% is within normal variation. Are you measuring literally the number of commits or number of lines changed? All of these are important metadata.
  - ponker 6y ago
    LOL I’m not going to give all that data. It would be immediately traceable to me.
    - pmiller2 6y ago
      Why bother posting, then?
bcrosby95 6y ago
I've also heard anti-WFH people state that its a bad idea. No one has any evidence. Just anecdotes. I think part of the shift is because companies are realizing it isn't as bad as they expected. A friend works for a company that announced they will allow 100% WFH in the future too. The CEO said she thought WFH would cause a 50% loss of productivity. But after 2 months of it, no significant loss in velocity has been noticed. Not just for development work, but for other office work as well.

ARandomerDude 6y ago

Perhaps if we measured lines of code... ;-)

nchelluri 6y ago
> If I had more time, I would have written a shorter letter.

aswanson 6y ago

Can only speak for myself. Im far more productive without an hour commute, people constantly interrupting me in my cube with trivia, interrupts to explain you can't do X in the code because the repercussions are X^10, etc followed by a long commute home. But hey, to each his/her own.

ponker 6y ago

Here are some numbers... I work at a FAANG and 28-day commits in my division (~5K engineers) are down ~3% from February. Not an exact science but some real data I just pulled for you.

Ankhers 6y ago
This is also a fairly bad time to try and put a measurement on how effective WFH is. The fact that schools and daycares are currently closed means that parents are having to take care of their children while also working. It is also hard on people that are used to working in the office to be told that they now have to work from home suddenly. I'm sure a non-trivial amount of people did not have a space for themselves to be able to work and they are trying to figure things out as they go. Not to mention figuring out how to communicate properly while everyone is remote. There is also not a one size fits all solution. People will be more productive being where they are more comfortable. That may be at home, in a coffee shop, at a library, in the office or any other number of places. People are expected to work from home right now, so a lot of those other options are out of the question. TL;DR, don't judge work from home when everyone has to do it during a pandemic.
wtetzner 6y ago
What are "28-day commits"? Are you talking about commits to source control?
- ponker 6y ago
  Commits over the trailing 28 days.
gregmac 6y ago
It's an interesting data point, but hard to judge much by it. Two things: 1. Working-from-home-during-global-pandemic-with-kids-and-spouses-and-roommates-and-not-ideal-setup isn't the same as working-from-home. Undoubtedly some people are less productive today. 2. Commits isn't necessarily a great proxy for productivity to begin with. For example, with greater focus comes less "oops I missed a unit test and broke the build" type commits, or even just more time to interactive rebase and clean up commits before pushing. "Go slower, you'll be done sooner."
- ponker 6y ago
  My point was that in my org, WFH has been staggeringly successful. 3% drop in the middle of the biggest global crisis in living memory means that there will likely be no loss or a gain from WFH once people get their emotional states back to normal and their kids back in school.
pmiller2 6y ago
What are the stats per engineer over the past year or 2, including standard deviation? I suspect +-3% is within normal variation. Are you measuring literally the number of commits or number of lines changed? All of these are important metadata.
- ponker 6y ago
  LOL I’m not going to give all that data. It would be immediately traceable to me.
  - pmiller2 6y ago
    Why bother posting, then?

bcrosby95 6y ago

I've also heard anti-WFH people state that its a bad idea. No one has any evidence. Just anecdotes. I think part of the shift is because companies are realizing it isn't as bad as they expected. A friend works for a company that announced they will allow 100% WFH in the future too. The CEO said she thought WFH would cause a 50% loss of productivity. But after 2 months of it, no significant loss in velocity has been noticed. Not just for development work, but for other office work as well.