I worked as a contractor for a while at a company that was pretty much entirely great developers, but they had one guy who had more experience than nearly everybody else put together (he was in his 50s, and had worked on all sorts of cool things in his career), who was also more productive than anyone else. The founders of the company knew he was their top producer, and so they gave him whatever he wanted (within the bounds of budget, etc.).

What he wanted most was a private office, and he got it, despite no one else in the company having one (it was kind of smallish and closet like, but he had a door that closed, and that made him happy). I always suspected that was a big part of why he was so productive compared to everyone else; though the huge amount of experience and his seeming disinterest in anything but churning out code (all of his hobbies also involved code to some degree, and he was always way out front on various trends in software development) were also big contributors. It's also worth nothing that he didn't put in long hours. When it was crunch time in the office, he wouldn't be one of the people there on the weekend. Paradoxically, I think that's another contributing factor to his high productivity.

So, yeah, I've often wondered whether Google/Microsoft/Facebook/etc. has internal metrics about this. Google is infamously an open plan office, Microsoft has offices, facebook is mixed (I think?). Surely, these companies who keep gigabytes of data about everybody on the planet has some idea about the relative productivity of offices vs. open plans.

As an almost 50 year old who has worked on many cool things I'll weigh in.

I get very tired of seeing the same mistakes in different guise. I don't get the youth culture of today and don't like it: including PC oriented culture and other media specific ideas designed to focus perception. I recognize patterns rather than specifics. I look at big picture and am careful about introducing paradigm shifts. I care about efficiency and ease of use, scalability and other ideas at inception of a project. If I am not respected for my approach and ideas I leave.

I'm older now myself (closer to 40 than 50, but well past "young"), and I don't think I meant what you think I meant.

"I don't get the youth culture of today and don't like it: including PC oriented culture and other media specific ideas designed to focus perception."

"PC oriented culture" just seems like expecting people to treat everyone with respect, and I think that's an improvement. Our industry hasn't always been good at that, but it has nothing to do with productivity. The old-timer I talked about wasn't racist, sexist, homophobic, or otherwise an asshole...he just liked to be able to close his office door and think really hard about things. He worked well with others; treating the women and brown folks in the office just like anyone else. Why would that be a source of contention?

And, I have no idea what "other media specific ideas designed to focus perception" could possibly mean. I know all the words, but they don't add up to anything, as far as I can tell.

These were my comments alone as a 50 year old in addition to your narrative; not in correlation but in comparison.

PC in this context meaning behaviors for the sake of change willy-nilly (think about the injection of feminine instead of masculine pronoun that was|is fashionable). I see the mercurial aspect of PC culture as reactive and dismissive and dislike it intensely.

Everyone these days seem to have some form of progressive agenda and most change is packaged in some media that makes it seem innocuous and correct in an engaging and entertaining way that would have been nearly impossible when I was a young man. In fact(as McLuhan might say) the media is the message. Chomsky is quite insightful in my opinion as to how the medium and vocabulary restrict available concepts.

"think about the injection of feminine instead of masculine pronoun that was|is fashionable"

This seems to have settled, after a few years of mild confusion, on everyone agreeing to just use non-gendered "they/them" when the gender of the subject is unknown. Pretty easy, it has historical precedent, and it can't possibly exclude anyone from the conversation. Seems like progress, to me. Where's the harm in making your docs about everybody rather than just men?

I guess I just don't get the hostility toward inclusiveness as a value in our communities/companies/etc. It costs me literally nothing and might make someone else feel more welcome to participate in the things I'm working on. It just seems like a win with no down side.

You didn't tackle any of the real criticism in my message.

You remark on a discrete symptom (which is a precursor) and you affirm it's essential importance to your life without any historical consciousness. And that is my basic gripe with PC culture: sensitivity to every little thing that you feel hurts you or impairs you or makes you feel included. Hell is other people, get over yourselves.

I couldn't make much sense out of the rest of your comment; it seemed vague and a little paranoid. So, I latched onto the one concrete thing in your comment, assuming it was demonstrative of what you were criticising.

"sensitivity to every little thing that you feel hurts you or impairs you or makes you feel included. Hell is other people, get over yourselves."

That seems simple when you're white and male, and the entire industry (and much of the world) treats your existence within it as a default and correct thing. I try to assume honest intentions when someone tells me something I've said/done is hurtful, and try to remember not to say/do it again.

The backlash against "PC culture" just comes off as really entitled and overly sensitive, to me (which is, ironically the same thing anti-PC people accuse everyone else of being). I don't know how to resolve that clash. I, personally, want to be part of solutions to problems rather than an impediment.

I view the imbalance in our industry as a problem worth solving; welcoming more women and under-represented people of color into our industry as equals (even when it means making changes to how we do things and how we talk about things) is a net positive for everyone. It's just increasing the pool of talent devoted to solving the world's technical problems and that can't be bad.

But, it sounds like I still may not be understanding your position. Can you provide another concrete example of what you're talking about? I noted your preference for big-picture rather than details earlier, but I'm clearly not grasping the big picture you're drawing...I need more details to understand where you're coming from

You like to characterize as vague (or worse) what can be reasonably inferred or researched and have a knack for constructing positions by pleading ignorance: which makes it simpler when creating straw men and redirecting|reducing conversation scope.

You sound naive and well meaning but I think that is artifice. Your comments reflect the essentially reactive nature of the PC movement (to me) which is to reject what you find distasteful and inject whatever you believe in (edit: with no real historical consciousness).

Do you believe that experienced white men should be displaced (by default) by persons of color or women in a profession because of skill equivalency? Is there a ratio for that displacement? Is it based on merit or is it based on something else?

> It's also worth nothing that he didn't put in long hours. When it was crunch time in the office, he wouldn't be one of the people there on the weekend. Paradoxically, I think that's another contributing factor to his high productivity.

I don't see how this is paradoxical. It makes just as much sense that better focused time, not "crunches", will increase productivity as the fact that offices will increase productivity. Same thing goes for telecommuting. The worship of overtime and 60, even 80 hour weeks is one of the big reasons I'm not in industry.

It is paradoxical in the sense that "more time working" should equal "more output", but most of us know that past a certain point the result of "more time working" is worse code that has to be fixed, leading to overall less productivity.

I'm not saying "more time working" equals more output. I'm saying there is a belief that it does, even among some people who should know better. The office I was working in did have occasional crunch times, and they'd put in extra hours; sometimes maybe it was necessary. But, old-timer I was talking about didn't really participate. His part of the job would have already been done earlier in the week.

> [...] were also bug contributors. It's also worth nothing [...]

I think you have a couple of typos here. Probably you meant to say "big contributors" and "worth noting".

Typos happen. I fixed /bug/big/ (I guess at the same time you responded), but didn't see "nothing" until after my edit link went away. You are correct in your interpretation.

Swipe keyboard perhaps?

I think you've got Facebook and Google reversed. Facebook is famously all open-plan. Google has a mix of open plan areas and mini 4-person offices with doors, along with other configurations.

I don't consider a 4-person office to be an "office"; that's "open", to me. Google does have lots of meeting rooms and the like, so as I understand it, most developers can find a private workspace most of the times when they might need it. So, it's probably not awful to work at Google.

If facebook is even more "open" than Google, I'm amazed anybody ever gets any work done, but obviously they do, so I may just be more reclusive than most tech people when I work.

I heard (in Seattle area) that MS office are on the way out, depending on the team you're on :/

Maybe someone can confirm...

Yes, our buildings at MS are transitioning to open space. We now have a large room for each team. Everyone has a height-adjustable desk that can be moved around within the room and a pair of good quality headphones.

Each room has its own small meeting room for devs to have long design discussions, and a larger conference room that is shared with the room next door. For making phone calls, there are really nice phone booths in the hallways.

That's discouraging. MS did a study to determine offices improved productivity, so what happened?

Open office = more dense work areas = profit

That is certainly the idea in general but it will depend on how badly it affects worker productivity. It's just strange to see a company do a study, prove that offices produce the best results and still go with a measure that presumably isn't cost effective (else the initial study should have concluded that the difference was well within tolerable loss).

I personally love working close with my team in an open-setting. That's the reason I go to office every day even though my employer offers WFH.

I just prefer turning around and asking a question or have a quick discussion. If everyone is working from home or in their personal offices, I hate having to get on a call or schedule a meeting to gather everyone to a meeting room for a quick discussion.

Sure, at times it can be distracting - I usually put on my headset if my tasks demand uninterrupted attention. My team also understands this and won't disturb unless necessary.

> I just prefer turning around and asking a question or have a quick discussion.

I read that as "I love disrupting the productivity of everyone else by interrupting them constantly." I personally hate that. When I'm in the middle of something I don't want other people asking me questions or having "quick discussions." Save that for when I'm between work.

Just use some judgement when asking.

If getting the answer in one minute saves you one hour, it is obviously good for the company to "waste" the other person's time.

You should definitely be careful not to waste the time of your coworkers. Just don't waste your own time either!

If everyone used their judgment this way, in a team of six, there are five others trying to save their own time and punishing the person who's worked to build up some knowledge.

This has to be fixed by ensuring that the bus factor improves and the knowledge is spread around more. I disagree that the solution is to let everyone interrupt the one who is trying to get stuff done.

> punishing the person who's worked to build up some knowledge. ... I disagree that the solution is to let everyone interrupt the one who is trying to get stuff done.

In essence we are talking about training new people and so this opinion seems selfish and unethical with respect to one's obligation to the firm.

The senior person should sacrifice some productivity until the new people are up to speed so that eventually the team has 5 senior people instead of one. If the new people are forced to figure out everything on their own, they might be only 5% as productive as they could be for perhaps their first 1-2 years. If the senior person simply gave up 20% of his time in order to answer questions for the first couple months, these 5 new guys likely would be closer to 80% productivity, representing a huge net gain in aggregate.

I only spoke against interruptions - not against spending time spreading knowledge. That, in fact, is in my good books as I already stated in my comment.

I guess I see these as 2 sides of the same coin; if the knowledge was properly spread via training and easy-to-use documentation, the junior developers would not need to interrupt much during the day.

>>> When I'm in the middle of something I don't want other people asking me questions or having "quick discussions."

Yeah, I get that, but most of us work in a team.

I think people need to state what kind of work they do when offering their preferences in these discussion.

Some complicated development tasks require 100% uninterrupted concentration for several minutes at a time. You're sitting there trying to build and maintain an extremely complicated model in your head, involving many complex parts, and their interactions along a time line, introducing permutations of that model to simulate the changes you are considering and simulating the outcome in your head. Then someone asks you something, or asks someone else something, or simply passes in your peripheral vision, and poof it's gone.

There a many tasks that don't require that kind of concentration, and many jobs, even in software development, that never involve those kinds of tasks. But with a setup that encourages 'communication', in the form of forced non-stop interaction, you have effectively made sure that the developers will never reach those heights, and most likely not by conscious decision, but out of ignorance.

Reminds me very much of this comic: http://i.imgur.com/3uyRWGJ.jpg

I used to think that way.

But then I started working with people who write simple and readable code. Now my programming isn't like that at all anymore.

Many complicated development tasks also require a high level of collaboration -- because there's simply too much for one person to be an expert in all of it.

I'd imagine that most highly-paid engineers are effective at both.

Of course it's an enjoyable setting for you.

Your team "won't disturb unless necessary", but you have no problem disturbing them whenever you want to have "a quick discussion".

Not all tasks require uninterrupted attention all the time. Sure, if everyone is working on complex stuff everyone can put on their headsets; no one will disturb others unless required.

I have seen open-setting that are pretty quiet and where you can get more or less work done during regular work hours. The open space where I work today, though, is an absolute nightmare. It's not that I get interrupted for technical questions, I am interrupted for all sort of things: people around me telling jokes, playing basket-ball, laughing, singing, gossiping, reading their news feed out loud...

The thing that baffles me the most is: the person that generates much of the noise is actually quite productive, but destroys all the productivity around him.

Someone from management obviously needs to step in and do something about it. But they are probably afraid of loosing half the workforce in that area of the office, if they did anything serious to prevent this to happen.

Some people prefer them. We used to have small open offices available. They'd seat six to ten people comfortably. If the employee wasn't okay with that, there were smaller offices with room for one or two people. Depending on which facility you were at, there were larger workspaces available that also worked for meetings and our HQ also had a larger conference room space.

It worked for us. Not everyone enjoys working in close proximity to other people and technologically creative people are sometimes introverts or have behaviors that aren't social normatives.

I'm not sure why companies are forcing people into open office plans. Unless things have drastically changed, real estate isn't a very high percentage of overhead. This might be different in SV, but it's not just being done in SV.

I'd also expect the increased productivity of happy workers to mitigate the increased real estate costs.

Six to ten is not an open office, those are team rooms which are fine - when you can have multiple teams in the same room, that's when you cross the threshold to an open office.

Ah, so it has to cross teams in order to be considered an open office? I was not aware of that requirement, thanks.

There's also this image (https://assets.bwbx.io/images/users/iqjWHBFdfxIU/i2FW30Wmwe2...) of some of their programmers doing work on their laptops which they hold, well, in their laps. After 10+ years of programming for a living I can't imagine myself typing code with elbows hanging in the air, I need to rest them on a desk while pressing the keys.

Some casual chat / browsing like that is fine, working? I can't do it.

Optimal size of team in one room is about 2 - 3 members. Any outside disturbance is likely to affect all team members.

Trying to explain this to management can be very frustrating because they don't understand that coding is a focus intense task. And it does not matter if you cite research. Coincidentally, management usually have their own rooms.

Some links in Swedish with Google Translate

Worse working environment in open office landscape (2014) https://translate.google.com/translate?hl=en&sl=auto&tl=en&u...

Opening office landscapes causes stress (2015) https://translate.google.com/translate?hl=en&sl=auto&tl=en&u...

Open office gives higher sick leave (2008) https://translate.google.com/translate?hl=en&sl=auto&tl=en&u...

Opening office landscapes leads to overtime (2014) https://translate.google.com/translate?hl=en&sl=auto&tl=en&u...

[M]any sparrows died from exhaustion; citizens would bang pots and pans so that sparrows would not have the chance to rest on tree branches and would fall dead from the sky - https://en.wikipedia.org/wiki/Four_Pests_Campaign

Playing devil's advocate a bit here: maybe the negative impact of open plan offices is much lower than typical if every single employee is feeling exceptionally motivated through far beyond market rate compensation? Of course there would be more boxes to check than just that, e.g. how often are people on the phone.

It is possible to demotivate people by not paying enough but it is unlikely that such strong motivation is achieved purely from greater pay. A frustrating work environment will eventually cause th motivation to fizzle. One might continue working to keep the paycheque coming but it is a tall ask to keep the motivation levels high.

Just giving a 20% raise across the board won't do much, and piecemeal rewards for visible successes will be even worse as they will cause more optimization for visibility than for success. I fully agree so far. But an engaging work environment with high pay will keep people a lot more motivated than the same environment with low pay. You can't fix a bad environment with higher wages, but higher wages make a good environment even better.

As for open plan offices, it's not the open plan that is frustrating, it's the coworkers that the office is shared with. Nobody is a perfect office-mate all the time. But open plan multiplies not only negative but also positive morale, so I would not generally rule out the possibility that it can be net positive in places where all other factors are close enough to perfection and that the problems only really start when they are copied by the less exceptional. High pay would be an important contributing factor (maybe even one of many necessities), not a silver bullet solution.

In other words, the only thing paycheck motivates people to do is to stay in the environment of low motivation for too long.

If any company thinks that it can fix the motivation with paycheck, they are very wrong and it will actually be more costly in the end.

Their system is already built, a slight tweak might increase the revenue by $100 million a year, paying for all the salaries. The top engineers don't need to sit down and code 9-5. Plus, this is also a Hail Mary strategy, every "Google" in China is aiming for their business.

I would be interested in knowing if better developers are more improved by having their own office above average developers or if they can handle the distraction more.

I think a lot of people value having other people around and in their sight, even freelancers leave their home office and head to coworking spaces and claim productivity gains. Personally I think the office per person idea is overrated, however I've worked in many places where the draw backs of open office hit sometimes. I think the best combo is open office, with hot private offices, and with the extra space, sleep rooms, relax rooms, big working table rooms, even put hipster stuff like a standing desk area in maybe etc.. Oh and if an employees ask for NC headphones just buy them! Also if someone really in the zone on something that distractions could be problematic be okay if Fred say's do you mind if I work from home today on this problem.

I second this.

I worked at a university for a few years, and lots of people were jealous of my private office in a distant building, but I just found it socially isolating. That said, I did get a lot done.

But at least you had the option to temporarily relocate yourself to a more social area. Then you could move back when you needed focus.

I actually didn't, there was no team.

Not with that attitude.

IBM did some tests and they found about a 15% improvement and this was across the board so no change between average and high performers

I'd love to read that study if you can find a reference.

steve mcconnell rapid development mentions the study where I found the 15% figure quoted.

Thanks!