Unsurprisingly, private offices score the best, by a huge margin. Right now, I've got a huge office to myself (separate from a two-person office next to me by an open passageway). It's super quiet because there are bookshelves full of books to absorb any noise, as well as thick 1970-era carpeting. It's the most productive place I've ever worked. I have no idea why tech companies would pay engineers gobs of money,[1] spend gobs of money on huge office campuses,[2] then inflict open plan or cubicles on said engineers.
[1] Office space is expensive, but not that expensive relative to an engineer's salary. The going rate for class A space in Manhattan is about $70/square foot per year. That puts a 10x10 office (assuming 85% usable space) at about $8,000 per year + build-out + maintenance.
[2] Apple's new spaceship building (in the middle of a faceless suburb) will cost $5 billion and have 2.8 million square feet of space. One World Trade Center (in downtown Manhattan) will cost $3.9 billion and have over 3 million square feet of class A office space. A lower-profile project, the New York Times building in Midtown West, was constructed for $850 million and has 1.5 million square feet.
Apparently Google spent almost $2 billion on a squat 18-story structure in Manhattan with an over-abundance of depressing window-less interior spaces... I'm going to assume it was the difficulty of getting 2.9 million square feet in one place rather than just bad taste.
EDIT: I'm not saying Apple's offices are open plan. I have no idea what Apple's setup is. I'm pointing out that companies are certainly willing to spend a lot of money on these big suburban office parks, multiples of what it would cost to just buy a couple of skyscrapers in SF. Thus, cost does not seem to be the motivating factor (except maybe in a penny-wise pound-foolish sort of way).
Keeping programmers in cubicles or in open office space prevents them from drawing parallels between themselves an other sorts of professionals that command similar or greater salaries.
As long as corporate programmers think of themselves as socially below doctors, lawyers, professors (who tend to get paid less!), accountants, etc, then management can continue to treat them poorly ("Everyone should be working late into the evening this month, but don't worry, free pizza!" ...I doubt those sort of tricks would work well in other professions.)
> I have no idea why tech companies would pay engineers gobs of money, spend gobs of money on huge office campuses, then inflict open plan or cubicles on said engineers.
It's largely due to the conventional wisdom created in the 90s by Microsoft's Jim McCarthy known as "21 Rules of Thumb for Shipping Great Software on Time".[1] Rule 6 (or 8 in his later talks[2]) is "beware of a guy in a room":
> Specialist developers who lock themselves away in a room, going dark for long stretches, are anathema to shipping great software on time. Without regard to their individual brilliance, before investing a developer with a significant assignment, it is essential that they understand and agree with the type of development program you intend to run. They must be capable of performing on a team, making their work visible in modest increments and subjecting it to scrutiny as it matures. Some people find this intolerable, and though there is a role for people of this disposition in the software world, it is not as part of a team devoted to shipping great software on time.
I remember that one. It doesn't literally mean "don't give engineers rooms." It means there should be regular public (to the team) milestones that force engineers to show their work.
The room thing just comes from the visual of a rockstar coder coming into work and locking himself in his office and saying don't worry, everything's fine, trust me I'll have it done in time. Beware of that.
> It means there should be regular public (to the team) milestones that force engineers to show their work.
While that is part of the underlying point, "beware of a guy in a room" is very much "don't stick a guy in a room by himself and let him do his thing" (he emphasizes it more in the video I linked to). The larger, more figurative point about having regular, public milestones is rule 4, "don't go dark", of which he claims rule 6 is a special case.
Not to be pedantic but he's making a point about "preeminent developer types" not office space. Specifically don't bet the schedule on one rockstar developer's word alone.
It really has little to do with the physical office space configuration. This problem could happen in open plans too, assuming a sufficiently complex project that can't be assessed by merely glancing over their shoulder.
I feel strongly that 99% of the people who are responsible for making this decision have no idea who "Microsoft's Jim McCarthy" is, other than maybe Jenny McCarthy's husband.
I agree that it is now conventional wisdom, passed down from manager to manager to entrepreneur to entrepreneur without regard from whence it came. But Jim McCarthy is somewhat of a legend, particularly in Microsoft circles, having led the team that created Visual C++ and helping to form the basis for the Microsoft Solutions Framework. Jenny McCarthy doesn't have a husband, and Jim McCarthy's spouse is Michele McCarthy.
I have no idea why tech companies would pay engineers gobs of money, spend gobs of money on huge office campuses, then inflict open plan or cubicles on said engineers.
A firm is a hierarchy, and this is a form of social control.
Yes, there "are" companies where the CxOs are out on the floor like everybody else, but I've only seen it once and that was when we were ~5ppl doing an officeshare.
> Apple's new spaceship building (in the middle of a faceless suburb)
> will cost $5 billion and have 2.8 million square feet of space
… that no employee below VP level can afford to live near. It's OK, though; Apple doesn't have to pay them for the tenth of their lives they waste commuting.
And above the entrance these words appear: “My name is Jobs, CEO of CEOs: Look on my works, ye Mighty, and despair.”
They can absolutely afford to live near it. Based on my own anecdotal experience at least, the people who work for the big tech companies in the South Bay with the long commutes are generally the ones who choose to have long commutes so they can work in the South Bay but live in SF. If affordability were their concern, they could save a ton of money by just living nearby in Sunnyvale or Campbell or wherever.
I also know people who work for those same companies who have long commutes for affordability reasons, but they seem to be the exceptions rather than the rule. Even in their case, it's usually for reasons other than there's just no way they can afford to live in the South Bay. More often, it's because they are determined to have a really big house with a really big yard and are unwilling to concede on that (such a thing is quite a luxury in the South Bay). Even then, they're accepting the commute for because they want to, not because they have to. It really sucks that housing is that ridiculous in the Bay Area, but it's not like any of these huge companies are going to pick up and leave and build their giant new campus out in the middle of the desert or something.
I don't think it's the affordability that's the problem - or rather, I know plenty of people at Google / Facebook / Apple / any other company in the valley who make the choice to commute not because of any cost saving, but because they don't want to live somewhere with very little going on socially and pretty much constant need for a car.
> … that no employee below VP level can afford to live near.
San Jose, Santa Clara, Campbell, and even parts of Cupertino and Sunnyvale are far more affordable than SF. This is especially so for couples and families.
(Suburbia is a far cry from SF in terms of culture, obviously, but that's a different topic)
Do you actually know if the new Apple building is open plan? I believe they currently have offices.
Pixar, the previous Jobs-designed office building, gives employees a lot of control over their personal spaces. Designers actually get their own mini house. (Kind of like an Ikea showroom.) They have lounge-like common spaces you can work/meet in too.
...your point still stands, but with regards to Apple - many engineers actually have offices, rather than cubicles or open-plan. You can be pretty sure they'll be staying at the new campus, or there would be an outright revolt.
> I have no idea why tech companies would pay engineers gobs of money,[1] spend gobs of money on huge office campuses,[2] then inflict open plan or cubicles on said engineers
It's because cubicles can be depreciated as furniture, over 10 (or 7, I've never been clear on the difference) years, whereas internal walls and doors get depreciated as real estate over a 27.5 year schedule. The CFO has to do something.
Anyone who has read Christopher Alexander's "A Pattern Language"[1] can easily extrapolate his ideas to cubicles. Alexander pointed out that if you drop any human in an open field with a tree, within a few minutes they'll be sitting with their back to the tree, facing the sun.
We are animals. We get physically and psychologically uncomfortable when our vision is restricted and our backs are exposed. It means we are vulnerable to predators. As such, cubicles are pretty much a nightmare.
[1] Christopher Alexander wrote about architecture, but his terminology was adopted by the design patterns movement in computer science, so it's familiar to most programmers. It's one of two building architecture books I recommend to programmers - the other is Stewart Brand's "How Buildings Learn".
>We are animals. ... As such, cubicles are pretty much a nightmare.
and open space plan even more so if you don't get your desk placed so that your back is to the wall and you're facing "the space". In low density office this results in people's desks placed along the walls with emptiness in the center, ie. waste of space compare to cubicles, in high density - people in the middle are miserable (speaking from the fact, we have 2 very similar buildings on the campus - both open space, one is low density and another is heavily populated).
This fascinates me. I find open-plan offices really hard to work in. Meanwhile—perhaps surprisingly—my favorite places to work are coffee houses or college libraries. I'm surrounded by strangers, but their anonymous presence starves off the feelings of loneliness that I can get working from home.
I think the difference between a public place and an office is that it's really unlikely for any of those strangers outside of an office to interrupt your work. In an open office, every conversation and set of footsteps feels like a potential disruption.
> I think the difference between a public place and an office is that it's really unlikely for any of those strangers outside of an office to interrupt your work.
Yes, I think this is the difference. The noise around you becomes background/white noise.
But honestly, I don't know how you can work in these areas. Doesn't it prevent you from making phone and conference calls? Assuming yes, maybe that's another reason you can become more productive.
I'd agree. Back when I was working on campus (now, mercifully, purely remotely), I'd sometimes head off to the main coffee joint instead of the completely open-plan office. Whilst it was hardly any quieter, it somehow worked much better for me, probably just as you describe.
For my part, I avoid phones anyway - which seems more commonplace than I'd imagine, amongst other technically-minded sorts.
IMO it really depends on how open they are, and what your company/team size is.
I love open cube "bullpens" for a team of 4-5 people working on a shared goal, but a completely open office where anyone can disturb you has given me issues in the past.
For me I think that is very close to the sweet-spot. Both for number of people I can put up with working in the same room with* and conveniently the size of a team which I find work best together.
* Exaggerated a little, I'd probably drive myself nuts in an office alone.
I'll put a finer point on it: in a coffee shop, you're guaranteed that zero of the conversations will require your attention. In an open-plan office, almost all of them will concern you even if you don't have anything direct to contribute. By this relationship, all voices in an open-plan will steal your attention.
>> Meanwhile—perhaps surprisingly—my favorite places to work are coffee houses or college libraries.
Totally agree, on days when I WFH, I drive my wife a bit crazy in that I like the TV on, yet get a bit perturbed if she interrupts with something not too important. :) The TV eases the pain of silence and does not require me to answer it. At work, there are at least a dozen direct potential possibilities for someone to interrupt me at any moment.
I worked in a company that changed their old, legacy office block for a brand new open plan one. Everyone was out in the open, and though there was an obvious hierarchy of seating, most of the desks were unassigned and changed ownership each day. The floors themselves opened out into a huge open space, and all the meeting rooms had glass sides.
I have to say I wasn't a fan. There was a lot of background noise, and it was very difficult to have a private conversation. I constantly felt like someone might look over my shoulder (a problem if dealing with sensitive documents), or interrupt work I'd struggled to get focussed into. It was also hard to have private conversations about sensitive topics without having to search for a space away from everyone's desk. Previously we'd just talk about stuff without moving.
My biggest frustration, though, was the fact it was very hard to be anything other than corporately bland in our attitudes. When we were split into floors and rooms, it was much easier to laugh and joke, and I often saw people bring something in to announce someone's birthday and make a fuss of them.
After going open-plan, it seemed the only way to celebrate anything was to interrupt the entire office, which put a large dampener on it.
There's also the issue that if you tend to be introverted, open-plan offices can be exhausting.
Remember that what we're asking folks is to self-report on what they like or not.
This may sound either pedantic or highly-controversial, but what people like and what might be best for the work they are doing are two different things. We do a great disservice to both to confuse the two.
I would like to have a great big office. Better still, I'd like a cabana at the beach, a laptop, and a hammock. What's best for my team or the project I'm on? Different things entirely.
What disturbs me is this idea we see repeated over and over again that if I like it, it must be good. No, that's not true at all. Going to medical school and having a brutal internship are not things you like, but they are things that take you somewhere you want to go. Life is full of things that are unpleasant yet work out for some greater good. We cannot blithely assume that we can follow our instincts on what feels good with every project we engage in. It doesn't work that way.
This may turn out to be one of those things. Or maybe we get our cake and eat it too: what feels good is what's best. But that determination has not been made, and this study doesn't advance the conversation along these lines at all.
We could take an anecdote from Richard Hamming's talk "You and Your Research"(+) - speculation on why some scientists do great work and go on to win Nobel prizes, and others don't, even if they are also clever and hard working. It includes this paragraph:
Another trait, it took me a while to notice. I noticed the following facts about people who work with the door open or the door closed. I notice that if you have the door to your office closed, you get more work done today and tomorrow, and you are more productive than most. But 10 years later somehow you don't know quite know what problems are worth working on; all the hard work you do is sort of tangential in importance. He who works with the door open gets all kinds of interruptions, but he also occasionally gets clues as to what the world is and what might be important. Now I cannot prove the cause and effect sequence because you might say, ``The closed door is symbolic of a closed mind.'' I don't know. But I can say there is a pretty good correlation between those who work with the doors open and those who ultimately do important things, although people who work with doors closed often work harder. Somehow they seem to work on slightly the wrong thing - not much, but enough that they miss fame.
It might be similar with open plan offices - less pleasant and more difficult to concentrate in today and tomorrow, but resulting in better, more competitive, more coherent software in twelve months.
I coach teams to perform better, so this topic is a bit of a sore point for me. I think Hamnming was on to something.
I've worked in about every possible configuration, and I've seen teams work in about every possible configuration. In addition, I've coded in all of these environments.
After working from a nice, quiet home office for many years, I'm trying out coworking. When I'm not with clients I drive an hour to an office to work with random people.
Why would I do this? Don't I value my time? Sure thing. But I find that the hassle and pain of being interrupted actually helps me to a reset to think about whether what I'm doing or not is important. Without that reset, I'll just grab on to some problem and keep tweaking it. I'm like that -- and I suspect many others are like that too. We desire uninterrupted time because we desire a deep problem to dive down in and forget about everything else. And sure thing, there are problems like that.
But 99% of the time problems are not like that, and we hurt ourselves more long-term than we help.
All of this is just conjecture, mind you. But this is an area where I see the natural inclinations of technology people diverge for what looks better for the effort as a whole. It's non-intuitive and uncomfortable. But that's the way life is, right? Dang humans.
I like cubicles and would prefer to be in my own box than in an open space concept room. Cubicles reduce noise. Respects my privacy, and puts up barrier against people distracting me.
Methinks "open offices" score better than cubicles on many points precisely because there is no expectation of privacy/quiet/etc. Lacking the expectation of better, there is less dissatisfaction with reality being worse. Cubicles at least give the occupant a semblance, and thus hope for, privacy/quiet/etc. so one is more keenly annoyed that it's not up to par.
I think it could also be a potential for distraction that prevents you from even starting to begin with a productivity task.
If you were in an open plan coffee shop things would be different as you don't expect someone to tap you on the shoulder with some arbitrary "urgent" bug to fix or fix the printer because you're "the I.T. guy".
So you actually begin real work rather than only do tiny chunks of busy work.
Agreed. Many people in cubicle farms won't think twice about taking a phone call at their desk. But this is frowned upon in an open office plan, where it is usually expected that you step into the hallway or a phone room.
In the past 6 months I've worked in a cubicle, open office space, and in an office of my own, with a proper door. I would say that productivity was lowest in the cubicle, and is highest in the office. Open office space was the most unenjoyable however.
The best combination I have seen is a two-floor suite connected by stairs. One floor has project rooms, which are small-ish open plan rooms, holding up to about 10 people, with reconfigurable walls tables and seating. The upstairs has some traditional offices (seemed like for business development and other phone-bound users), conference rooms, and a very quiet, library-like open plan area for individual on-site work.
I keep wondering if there's a market for "co-working" spaces that don't follow the open-plan style. What would it look like? I know Joel tried to floor-plan the ultimate developer office, has anyone followed up or built on that since?
Rent-a-carrel seems like it could at least make a go of it in a high-density coworking market like SF. Heck, it would be cool if a coworking space had various areas with varying degrees of openness.
Some problems get solved really quickly because of them. The number of times the person who knows the solution overhears others having a problem is astounding.
But then there are problems that just cannot be solved due to the level of noise. Especially some collaborative problems, where a problem requires two people to focus on it really carefully. Distractions just from hearing others talking can interrupt lines of thought and beyond a certain point limit mental ability.
This is never a positive thing to me. It helps when everyone has the same level of problems but it really sucks when you never need help from others. This is a personal problem with the open office I deal with now. I had a coworker that would make disruptions interesting and light hearted. Now its just me and a junior and his disruptions are everything from eating to talking to himself to having conversations with others at volume level 11. The last is not really his fault when someone else leads that charge but a fidgety person is not a great fit for an open floor office shared among 4 people. A much bigger open plan where that can be dissipated by the distance between others. Right now it just feels like someone is constantly in your ear eating potato chips loudly. I purposefully eat potato chips with my mouth closed and always think about my "audible footprint" but when someone else doesn't it's just so fucking annoying.
I currently work in an open office space and I enjoy it. I love being able to ask the analyst across from me how s/he goes about doing something without having to get up and peer over a cubicle wall or walk into an office. Bouncing ideas off people and getting an immediate response is great. I could see it being more distracting than a cubicle or a real office. I've never worked in a place with those so I can't compare.
> I could see it being more distracting than a cubicle or a real office.
Exactly. The problem is when the guy beside you is trying to focus. Your ideas bounce off the analyst, but they also bounce off your neighbor whether he likes it or not.
There are technologies that allow you to get immediate responses like instant messengers and email. If you absolutely want an audio or video conversation, a better place for that would be in an isolated area (like an enclosed room) where you won't disturb anyone else.
>I love being able to ask the analyst across from me how s/he goes about doing something without having to get up and peer over a cubicle wall or walk into an office.
Your cartoon doesn't really pertain to this because telling someone they have an e-mail is zero value added. Me asking my neighbor how to forecast something and saving what could be minutes to hours, despite interrupting a train of thought and setting them back a few seconds, definitely does add value.
> Me asking my neighbor how to forecast something and saving what could be minutes to hours, despite interrupting a train of thought and setting them back a few seconds, definitely does add value.
Only if either A, your interruption costs them less time then it saves you, or B you believe your time is more valuable than theirs.
I think its a good idea to establish within a workplace, esp. an open seated workplace, a way of broadcasting the price of interrupting you. Sometimes this is informal with headphones as a sign, other times people make DND signs or lights, etc. Its just much easier when everyone has an office- you can shut your door.
I don't think you can meaningfully compare open workspaces to cubicles without talking about how densely people are packed. I really like the open plan office at my last job, but that was because people were spread far enough apart. And we had side rooms for taking phone calls, having discussions, etc.
I've been in an open workspace that had three or four people to a desk and it was a nightmare. Now I work at an open workspace where everyone is spread out and it is much better. The closest person to me is ~2 meters away.
The best setup I think I've ever seen was a large room with tiny personal cubicles lining the walls, and islands of open work areas in the center. You could go to your cubelet for phone calls, email, storing your coat, etc, but coding work was done at the islands. Each island consisted of five workstations two on each side and one at one end. Each workstation had two 24" monitors and a wide berth, with two chairs and more available. Pair programming was comfy, and getting four or five people around a single piece of work was easy.
If you were at the islands, you were working and available to be called over. If you were in your private cubicle, you were to be left alone unless it was really important. From total openness to decent privacy in a few steps. That's as good a balance as I've seen.
I would like to see something that talks about the efficiency of both approaches - ideally from a controlled test.
The reality is offices are much more enjoyable then cubes. Many years ago I was part of a consulting project that cube-farmed a large organization previously used to offices. Everyone but the senior-most execs were pushed out of offices. It was done under the guise of improved teamwork. I saw lot of theoretical support for this, but never did see empirical evidence. Maybe it exists - I'd like to see it.
I was going to suggest that the headline needed editing since the article doesn't seem to support that conclusion. But then saw that HBR actually used that mis-leading headline.
The research seems to only deal with employee preference and not much with performance.
I'm surprised there is still so much disagreement on what works best and even more surprised how popular open plans are. Open plans only work for the very few people who either never need to talk to anyone and/or are comfortable with do-not-disturb headphones on.
Weird. I think it was on the Discovery Channel recently ( Past month? So also "recent research") where a study on open-plan offices was mentioned and how workers felt they had no privacy saying cubicles were better.
I think it comes down to the person, why not offer both choices?
There's also a self selection bias. I'd never subject myself to open plan voluntarily due to the performance hit. I suspect having personal standards probably has a major influence on general attitude.
"They pretend to house us, we pretend to work" type thing vs being a bit more driven resulting in, on average, more grouchiness.
>The bottom line: workers in enclosed offices were by far the happiest, reporting the least amount of frustration on all 15 of the factors surveyed. Workers in cubicles with high partitions were the most miserable, reporting the lowest rates of satisfaction in 13 out of those 15 factors.
Really? you don't think that he people with personal offices are not in better, higher paying, less menial jobs to begin with? I think that would be obvious and clearly a better reason that they are happier. If there are some distractions and lack of total privacy, but you love what you are doing, do you really think that you are going to be miserable?
Please stop with the correlation posts where there is more obvious potential conclusions. So annoying.
While I share your (healthy) skepticism about scientific reporting, a good researcher would've attempted to structure their experiment to control for this very effect.
I don't have ScienceDirect access so I can't check the research or methodology, but from reading the abstract & the tables I can see it appears the reporter's conclusion came directly from the experiment, so it's far less likely that this is a reporter mistaking correlation for causation.
>A good researcher would have the cash to hire hundreds of people and intentionally randomly assign them to different work environments, I guess.
Wrong, you lack the same basic, problem solving logic that I am referring to.
A good researcher would have asked additional questions such as "On a scale of 1-10" - "how enjoyable is the actual work that you do here?", "how satisfied are you with the amount of money that you are making?", "where do you see yourself in five years". These aren't the best quesions that I could come up with, I am just giving examples off the top of my head.
It would be easy to see how unhappy these workers are with other things in their lives rather than if they were in a cubicle vs a private office.
[1] Office space is expensive, but not that expensive relative to an engineer's salary. The going rate for class A space in Manhattan is about $70/square foot per year. That puts a 10x10 office (assuming 85% usable space) at about $8,000 per year + build-out + maintenance.
[2] Apple's new spaceship building (in the middle of a faceless suburb) will cost $5 billion and have 2.8 million square feet of space. One World Trade Center (in downtown Manhattan) will cost $3.9 billion and have over 3 million square feet of class A office space. A lower-profile project, the New York Times building in Midtown West, was constructed for $850 million and has 1.5 million square feet.
Apparently Google spent almost $2 billion on a squat 18-story structure in Manhattan with an over-abundance of depressing window-less interior spaces... I'm going to assume it was the difficulty of getting 2.9 million square feet in one place rather than just bad taste.
EDIT: I'm not saying Apple's offices are open plan. I have no idea what Apple's setup is. I'm pointing out that companies are certainly willing to spend a lot of money on these big suburban office parks, multiples of what it would cost to just buy a couple of skyscrapers in SF. Thus, cost does not seem to be the motivating factor (except maybe in a penny-wise pound-foolish sort of way).