The major takeaways I have from reading about this when I started in tech entrepreneurship are:
1) Knowledge workers need 30 square feet of desk to get optimal efficiency
2) They need 100 square feet of space
3) cubicles are as good as an office if the walls are over 8 feet high (!)
4) small teams tend to want to talk at the same time, and be quiet at the same time.
These together round up into recommended 400 or so square foot offices with doors that close and house teams of four people, all working on the same thing at the same time.
If teams are 2 or 3, they should have a smaller space. That's enough space for desks, a couch, a whiteboard area, etc.
I have found over the years that this tracks fairly closely with my own preferences. I really want to be able to talk and interrupt a team-mate when we're working if we're working on the same thing. We also often are hacking away and need silence, but instant access to whoever is in the know about a certain bit of code is undeniably great, and provided we aren't breaking each-other out of flow, is fine.
This system works really well when it's a small team and the team is working on the same stuff. It breaks down when there are multiple things happening or more than four or so people: interruptions abound, killing productivity.
I think you'll often see workers getting this situation when they really need to be productive by commandeering a conference room and shutting the door. But, it makes sense to provide it for teams in general, in my opinion.
1) Knowledge workers need 30 square feet of desk to get optimal efficiency
I expect this to be different now, and possibly more dependent on screen space than on physical desk space.
I think you'll often see workers getting this situation when they really need to be productive by commandeering a conference room and shutting the door. But, it makes sense to provide it for teams in general, in my opinion.
There have been a couple big projects where we flew in the remote people just to get the whole team in a room together. So it's not only the workers that know this; management does as well or they wouldn't approve the travel budget.
Maybe a little different, but having a lot of physical desk space is still very important IMO. It feels great to have lots of desk space, it helps carve out more privacy since there are less possible closer places other people's desks can occupy. I dislike when people peek around my screen just because it is convenient to.
There's definitely research that shows more resolution and screen real estate are helpful for productivity. And, of course, more of our work does happen on screen now than in the 70s.
Still, I'm personally happier with more desk space. We still use paper all the time. Tufte points out in one of his books that humans are better at spatial comparisons than time-based comparisons, hence his fetish for data density, and mine for lots of screen and desk real estate.
If I remember correctly, the 30 square feet of desk space was largely to be able to have a large number of tech manuals open to the necessary pages at the same time. Given how much documentation is now digitized, I wonder where the same recommendation would still hold, or if you could just condense it down to the space of a couple of monitors?
I have ~22 sq ft of desk space http://i.imgur.com/hKkW0T4.jpg and it's glorious, custom built my desk to my requirements (standing because I have a back condition) with enough space to thrown a laptop and another screen on the end, it also feels lovely to work at because I never feel 'hemmed in' by all the stuff on the desk.
This is entirely off topic, but I'm starting to think about making my own desk at home. How'd you do that, what did you use? Your desk looks great, and worth emulating.
Scaffolding was old stock from a local scaffolders (I wanted the distressed industrial look, just given 'em a ring).
Fittings are from ebay (there are standard sizes and joints), the legs are spinners (if you turn the handles the desk moves up and down).
The top panel is 28mm structural birch ply (structural ply may have surface blemishes and patches rather than finishing ply so if that matters get finishing ply but since I wanted the industrial look and it's way cheaper to boot), routed with a 5mm edge and covered in 4 coats of heat resistant wax (makes it water proof without ruining the color).
Original panel was 8ft x 4ft so I took a strip of the long side for the shelf (if you work out your dimensions most timber places will make the cuts for you).
Through holes are 60mm and there are 3 on the back (UK plugs are massive).
http://i.imgur.com/YzfWv5E.jpg in construction, two cross beams are for the shelf which sits on them, the middle upright is because even 28mm play will sag.
Total height range is ~85cm (I'm quite tall and normal desks are a tad too low) to ~140cm.
Top panel is held with wide screws, 4 on each post (the fittings underneath had holes already), the shelf with the PC has a slat underneath which slots between the two horizontal bars.
Not interested in design awards ;), I wanted something brutally simple that would be the last desk I ever need (worst case I replace the panel if it gets battered).
As for sturdiness, it's weight and rigidity make it a very stable desk.
I've already had two people ask me to make them one (with a nice margin on top), I think a lot of people are tired of crappy fall apart desks.
I haven't had that exact structure yet, but what I have had tracks well. I've had fully open-plan, single cubicles, and 4-6 person super-cubes. The super-cubes definitely seemed like the best option. I had easy collaborative access to my immediate teammates, with a bit of visual and audio barrier between us and other teams.
Open-plan was workable. Some sound pollution from other teams, but easy access to my team was good.
Individual cubes make no sense to me at all. All the sound pollution of open-plan, with none of the easy access to teammates. I hate it.
Wow, do I hate open plan. Have seen it implemented at several companies and all suffered trackable drops in productivity, even as they continued to tout how much better it was.
I think the real reason is 'cost', and the loss of productivity isn't quantifiable anyway; so whomever makes the decision can put an exact number on how much he 'saved' while being able to ignore the side effects.
It's like people who buy solar panels in north england; they think it gives them 'free' energy and they are being 'green', while ignoring completely the carbon footprint of making that panel in the first place; while that panel will /never/ repay itself with the energy it harvests.
Going back to open offices (and I'm typing this from one!) I think it's a disaster. There are days when all I can really do is read HN, as there's about 3 loud conversations within 5 meters of me, including one Italian guy who's yelling into his skype microphone.
This is basically it. There's a thing I've noticed, I like to call it the "inverted reason list". It's an upside-down ordered list of reasons to do something, delivered to a skeptical or hostile audience. The list is ordered such that reasons the audience will like are delivered first, as if they are the most important, and the reasons the audience will hate are delivered at the end, as a side-thought.
"and oh....it turns out open offices will save us $5m a year in facilities costs." is usually delivered at the end, when nobody is paying attention anymore, because it was preceded by "free snacks at every 25th cube!" and "will allow you to configure seating to keep your team near you"
For work that requires lots of talking (or not lots of deep thinking), they work "ok", but for knowledge work, they're basically the worst idea ever conceived outside of sending all your coders to rock concerts all work day and wondering why nothing is getting done.
Where I work, most people get or split offices, but we can work in a few other places, including an open work area. Most people hate working there because it will inevitably have somebody on speaker phone having an argument with their health insurance over an unpaid bill, and that conversation will go on for 2 or 3 weeks solid. Appeals to management have basically been ignored, and now we're getting ready to move to a new facility that's all open space.
> they think it gives them 'free' energy and they are being 'green', while ignoring completely the carbon footprint of making that panel in the first place
It's a little OT, but do you have some real data to back this up? This argument came up a lot in discussions about light bulbs and fluorescent lamps, but in reality cost of fluorescent lamp is around 1% of its lifetime energy consumption. I highly doubt that solar panels cost more energy to produce than 10% of their lifetime output, even in such poor conditions as England.
I don't have any knowledge about the efficiency etc. but one possibly relevant note is that sometimes there are government subsidies, so even if there was a situation where it wasn't environmentally better it could still be a financial win for the end user. (I know this scheme was or may still be available in Oxford, I don't know if it was UK-wide or could have been the local council, but I know it wasn't a private company offering the incentives.)
Actually, I have to correct myself - it looks like most estimates give PV panels EROEI somewhere above 6, which would mean about 15% of produced energy is needed for the production and maintenance of the panel itself. But even then it would probably be net gain.
Subsidies etc. are a different matter, because then you're talking about competing with other sources and the strawman argument "it costs more energy to produce than it saves" doesn't work.
But couldn't that be solved through office culture? Simply telling people "skype calls happen somewhere else"? Though obviously it would take a lot of work to get everyone to agree that that is the "best" culture. Especially for people who talk for a living (customer service, sales)
Surely you have an empty conference room somewhere
I do embedded work; I have a workstation, and a board about 70x50cm long, cables, and an oscilloscope.
So I cant /really/ shift that to a conference room to wait until whomever stop arguying with his mom...
It's not just that tho, even people who move they ass to talk to someone else usualy /stand/ by that someone else, so instead of having a quiet head-to-head, it always become a 'broadcast'.
I know that they could at least have signs saying 'quiet space' and all that, but that would fly in the face of the idea they have that it 'improves communication'. It's blatantly false, I have no friggin' clue what 80% of the people are doing in this office anyway...
Just having to get up and move is a pain in the ass and loses productivity.
My firm moved me to a shared office for 6 months (finally I got my solo office back last month). They had that policy. "Oh for conference calls just use this admin office we set up." But nobody ever used it.
That's a huge "FUCK YOU" to your employee. The company is cheaping out and then making your job harder by forcing you to relocate.
And it plain just doesn't work. If a guy I'm working with calls with a quick question, what am I supposed to do? Call back?
There's 8 of us in our programmer section; people take Google Hangouts and their phone calls within the section. It takes a few minutes to walk out to a private phone area and the other conference room's door doesn't close properly so you can't block off your noise. So what we do is just wear headphones and try our best to ignore other people's conversations. Which is a problem because all the other office workers are usually so quiet that it's like a library in the office.
The real reason certainly isn't cost when it involves one or two person pre-revenue businesses burning £200+ per month and an hour of daily commute in order to sit at a desk in a loft surrounded by other people rather than sitting at a desk in their nice quiet bedroom...
I don't buy the opening argument; essentially "rich people are smart". We've endlessly witnessed stupid corporate acts. To name a few that have nothing to do with open offices: Conglomerate building (if your name ain't Warren Buffett you almost certainly destroyed value), corporate raiders (destruction of value to line the raider's pockets), the 1999 bubble (let's throw billions of dollars for 'eyeballs'), off-shoring, American automakers (do I have to get into the details here), stack ranking, faulty risk assessment (nearly every NYC banking/investment firm). Waterfall (the stupid kind, not the way it was meant to be done, which is quite good for projects involving airplanes and such). Agile (as it is practiced, the manifesto is nice for certain kinds of projects). Windows 8, or MS' ignoring the internet for a long time. I could type all day.
Companies do stupid things all. the. time. They fly in the face of all available evidence. People are put in charge that shouldn't be in charge. They are driven by ego, or other measures that lead to sub-optimal decisions. Sometimes they are really good, but just make mistakes. I'm a programmer, and my compiler tells me I make 100 mistakes a day; I don't fault upper management for the same. They just don't have that immediate feedback loop that we do. Of course they will make expensive mistakes.
Heck we see it in this thread. Evidence shows that productivity is lowered in open office environment, yet we read claims that it is improved.
Just in case you're not being sarcastic—the level of resources you have access to as a child is much more tightly correlated to how much money you make as an adult than individual IQ, at least in the United States.
I am an old curmudgeon who hated working in an open office (at Google in Mountain View) for a few years.
My personal theory is that the open office trend at Google (and probably Facebook, and others) was initially driven by the fact that the founders jumped directly from school, and never had a "real job" with a normal, old-school office when they were first starting.
Thus they have never personally seen the immense value of a real office, but they HAVE seen the value to the "computer lab" sort of bullpen environment that was common in the 90s (when almost everybody had to show up to the lab to get their work done, as very few people had *nix machines elsewhere where they could do their assignments).
Yes, that's my theory as well. As noted in other comments, some studies have shown that an office of 4 people all working on the same project is the most efficient, and other studies have shown that a single office is best for companies with < 10 people.
And Zuckerberg and Brin and Page probably switched from mostly programming to mostly management by the time their companies reached substantially more than 10 people. They probably have really fond memories from their time when their companies were <10 people.
When a company hires you they are getting: your availability to work, and your capacity to work.
When problems are well defined (you have workable requirements) then you availability becomes less important, so your capacity to work becomes relatively more important.
When problems are not well defined or written down, then availability becomes more important, and capacity for work less important.
Given that most startups are working on simple CRUD apps, where their most difficult problems usually stem from their own poor organizational skills (poor build procedures, badly configured servers, technical debt in the code) rather than a difficult problem domain it seems clear to me that availability would become more important to employers than capacity for work.
Also managers prefer to be able to _see_ people working. (Never mind that the work being done in a bullpen is not productive.)
Open offices are like hives: filled with busy looking, low cost replaceable drones.
tl;dr: Maybe open offices are popular because most startups aren't working on hard problems.
> Given that most startups are working on simple CRUD apps, where their most difficult problems usually stem from their own poor organizational skills (poor build procedures, badly configured servers, technical debt in the code) rather than a difficult problem domain it seems clear to me that availability would become more important to employers than capacity for work.
Yet these startups hire people as if they had very difficult problem domains to tackle, thus filtering for people who work better in the other environment you mentioned.
I'm sick of the circle jerk hatred about open offices.
Yes, let's discuss the pros and cons and acknowledge that they don't work for everyone.
But no setup works for everyone. And we need to stop pretending that open office plans are some poorly thought out pranks invented by clueless managers that offer nothing to developers.
I am a developer. I am not an extrovert. I love working in an open office, and the one job I had almost 10 years ago that was rows and rows of offices made me MISERABLE. I almost quit but luckily was able to move to a team that had a team area with an open plan.
Fact: In today's age of agility, rapid development, minimum viable products, continuous deployment, and short iterations it is much more important that you can discuss things with your coworkers and stakeholders quickly and figure out if you are working on the right thing, rather than going away for 6 months into some basement, close the door, and come back with a finished product.
Fact: Distractions are cultural not physical. Provide the team with good quality headphones, shame people who have non-silent notifications on their devices, create a policy where people can indicate when they're deep in thought on something complex and would prefer not to be distracted (eg. Do Not Disturb status on internal chat that you check before messaging someone or tapping them on the shoulder), and provide flex space people can go to on a case by case basis for privacy. Do this and all the problems with "focus" that open-office haters complain about GO AWAY.
> Fact: In today's age of agility, rapid development, minimum viable products, continuous deployment, and short iterations it is much more important that you can discuss things with your coworkers and stakeholders quickly and figure out if you are working on the right thing
Fact: that has absolutely no relation to offices. And "offices" doesn't have to mean singles either (they usually don't in my experience).
> Provide the team with good quality headphones
I can't wear headphones. I literally, physically, can't wear headphones for more than 10~15 minutes, it starts hurting like somebody's driving nails through my eardrums. Now what?
> shame people who have non-silent notifications on their devices
And people who have bodily functions, and people who put the volume too high on their headphones, and people who type too hard, and people with tics. Just shame people on a general basis really, that makes for great esprit de corps.
> Do this and all the problems with "focus" that open-office haters complain about GO AWAY.
Yes, everybody must love what you love and you can't be wrong, because "no setup works for everyone" except for the one you want.
>I can't wear headphones. I literally, physically, can't wear headphones for more than 10~15 minutes, it starts hurting like somebody's driving nails through my eardrums. Now what?
You think? Sadly having fucked up middle ears isn't exactly fixable, it's something I can only make do with, so that didn't answer the question: how do you[0] handle employees who can't wear headphones when your[0] solution to noise issues is "just wear headphones"?
[0] this is a generic "you", I know the specific you is not the one who originally made this assertion.
Large over-ear headphones exert NO pressure on any part of the ear. They are basically inverted cups that go over and around your ear and insulate.
Are you saying those don't work for you?
Unless you're talking about the volume of music making your ears hurt in which case guess what - I didn't say anything about listening to anything on those headphones. Just having them on provides enough insulation that you don't need to pipe in any additional sound
So then what are you saying? Wearing comfortable large headphones with quiet ambient noise and active noise cancellation causes you pain in your inner ear?
You possibly just accept the 0.05% of your employees who cannot wear headphones are going to work somewhere else / be less performant.
Aside: I still don't quite get what exactly about the headphones is causing you pain? If you have them on, without playing anything, is that painful? (in which case maybe lighter ones that exert no pressure on the skull could help?) Is it the volume? Is there some weird property of sound produced by the headphones that is not present in general noise, that causes the pain?
(please don't take this as me disbelieving you, I'm just curious)
Just because you like it does not mean that a lot of people likes it. As to reflect to your "Fact"s:
"Fact" 0: In today's agility etc: this is mostly true for startups where you have a product owner who does not really know what he wants. If you use for example SCRUM you plan beforehand and you have an immutable sprint when you know what to work on thus you can work uninterrupted for 1-2 weeks then you can have a meeting to discuss the direction and have a retrospective + planning. The "work for 6 months in a basement" is an exaggeration and no one works like that today (except for the developer of Dwarf Fortress but it is another story). So the point is: you can be agile/rapid/do MVPs/etc and uninterrupted at the same time. I have done it at many companies so I know it works.
"Fact" 1: Distractions have nothing to do with culture. People in an open office distract you simply because your preferences does not match:
* switches off air conditioning
* bumps your chair because of small spaces
* turns on the lights (which you don't like)
just to name a few.
Another thing is that policies work in theory but in reality most people will think that their problem is more important than some policy and will go to your desk and ask about something (and in most cases the will know the answer when the have finished speaking). If you have a private office it acts as a barrier for such distractions.
You can go very far down the wrong path in a 2 week planned sprint. Sorry, even with a well planned sprint and an experienced product owner, you still have a lot of questions that need to be evaluated halfway through the sprint, and talking to others is how you resolve them.
As for distractions...are you such a precious princess that someone adjusting the light level or bumping into your chair completely removes your ability to be productive?
How do you get by and achieve anything in this world?
Why are we optimizing to create work environments for such snobbery? What makes you assume that by default EVERY single problem that others might have is not worthy of an interruption enough that you need your own private office to protect yourself from the proletariat?
> In today's age of agility, rapid development, minimum viable products, continuous deployment, and short iterations
You know, that line is a loud bullshit alarm all by itself.
But I'll bite anyway. Open offices make communication harder. Much harder. You can barely talk to someone with you equipment around without disturbing someone. So you must choose, either you talk or you produce something.
Which part is bullshit? Prototyping? Shipping fast? CD? MVPs? If you are working at a company in 2015 and you are not working on an OS, and you are not doing these things, you are a dinosaur.
Yes, you disturb someone. You let them consider whether they want to pause what they are doing and join your conversation, and make it more productive, or put their headphones on and tune you out. The choice exists.
No, none of those are bullshit. The idea that they are new is complete bullshit, and the idea that throwing those terms around justifies something is also so.
>Fact: Distractions are cultural not physical. Provide the team with good quality headphones, shame people who have non-silent notifications on their devices, create a policy where people can indicate when they're deep in thought on something complex and would prefer not to be distracted
Managers who deliberately create environments where headphones are mandatory should be brought up on charges.
Basic rule of thumb: If your headphones are loud enough to block out ambient conversation, they're damaging your ears. In 20 years, deafness is going to be associated with programmers in the same way we associate black lung with coal miners.
Also, I'm happy for you that your brain is only derailed by particularly loud ringtones or people tapping you on the shoulder, you lucked out on that one! But my brain is different: https://en.wikipedia.org/wiki/Misophonia
It isn't just loudness that affects hearing loss. A large amount of the damage coming from headphones comes from the fact that they increase the temperature of your ear canals. That temperature increases the amount of bacteria which cause significant hearing loss.
Oh for fuck's sakes. I wear headphones 90% of my day WITH NOTHING PLAYING IN THEM.
We need to stop treating developers like precious children. You do not need ABSOLUTE SILENCE to be able to do your work. 99% of us are not writing the next PageRank or inventing the next Djkstra's Algorithm. And even when we are 99% of the time is spent on build environments and configuration and plumbing.
This does not require absolute uninterrupted focus surrounded by silence. Making a mistake does not cause a bomb to go off or a plane to fall out of the sky.
Put on a pair of comfortable headphones and do not plug them into anything and 75% of the noise around you is gone. Turn on some white noise or some gentle ambient music of dozens of lyrics-free genres and it goes up to 95%. If the only way you can work is if you can't hear any other noise outside of your own head, you have failed as a human being because no other profession requires that.
Everything is a tradeoff. Cognitive load decreases productivity. Open communication and easy collaboration increases it. The argument is that it makes up for it. The current popular opinion on HN is it doesn't.
Rationally, it doesn't matter whether you have empathy or not. But you need to demonstrate rational understanding of the issues, and I don't think you've done that here.
There's a lot of evidence that open offices destroy productivity, and almost no evidence that they increase it.
So whatever your reasoning hiring people, paying them a lot of money, and then making it harder for them to do their jobs is an irrational position.
While its true that headphones are associated with hearing loss, there are lots of headphones out there that have noise-canceling capabilities without being excessively loud.
Noise cancelling works great with cyclic sounds, such as an airplane engine. With acyclic sounds, such as conversations? They're rubish. No headphone in the market today can cancel out[1] a conversation with a method other than playing another sound high enough to cancel it.
[1] By cancel out, I mean make it impossible to understand the conversation, while the headphones are silent or playing in low volume.
I've got a pair of [1] in currently, no music, and they're doing a pretty good job of drowning out people talking at normal volume 6ft away. I'll admit they're doing a less good job of the strident conversation 10ft away and the plague of ringing phones but without music, I'm ok with that.
(NB. This reply entirely useless if you can't tolerate in-ear buds.)
In ear monitors are often a combination of ear plugs and headphones. Mine are rated at -30Db rather evenly across the spectrum. I have to take them off to effectively engage in conversation below shouting level even when the audio input is silent. On the other hand, they also are quite expensive. At any rate, if you do have effective noise cancellation it can be quite alarming when someone tries to interact with you by tapping your shoulder and such.
I was not totally clear. My comment was about active noise cancellation. Passive cancellation, such as what you get with protective gear or, as you say, studio monitors, has a nasty side effect I can't tolerate: suddenly, sounds transmitted through bone transmission are much louder (breathing and heartbeat mainly).
Sounds like a LOT of complaining, making engineers here look like a bunch of divas who are never happy. Stop whining and do your job instead of making up excuses why you can't.
Sure, but that doesn't mean you just choose whatever because it won't be best for everyone. You pick what will work best for most people, and studies consistently show that is not an open office layout.
> In today's age of agility, rapid development, minimum viable products, continuous deployment, and short iterations
Not every developer is writing code for a seed stage startup trying to get an MVP out the door. In fact, most aren't. For small/early-stage startups I totally think open offices make the most sense.
> Provide the team with good quality headphones
Headphones help decrease audio distractions, but not nearly as much as offices or cubicles. They don't do anything for visual distractions. And not everyone is able to use them (many people find the music as, or more, distracting than the audio distractions they're trying to avoid).
Until recently I was sat next to someone who not only drummed on the desk and with his feet but spent half the day twirling around on his chair like it was a damned fairground ride. Since moving to a new desk, my office life is much less tense and I'm a lot more productive.
Headphones are a cognitive and physical distraction to me. When I think, I only want to think about what I'm thinking about. I only like listening to music when I'm not trying to accomplish something.
Headphones are merely an inferior office, with its walls shrunk down to fit your head.
> "Fact: In today's age of agility, rapid development, minimum viable products, continuous deployment, and short iterations..."
I don't think I would take a job anywhere that held these buzzwords in high regard. They have been discredited time and again as the wankery of company culture.
"age of agility"... No.
"MVP".... No!
"continuous deployment".... kill me now.
You mention stakeholders. At my last job, stakeholders were the biggest problem in the company. I had a desk in an open office, and for some reason that meant the project manager had an open invitation to walk up to my desk repeatedly throughout the day bugging me with brain-dead questions and pointless "standups". Yep, she wanted regular standups even though it was just her and myself for many of the minor projects. Jesus I'm glad I got out of there.
> "create a policy where people can indicate when they're deep in thought on something complex and would prefer not to be distracted"
>she wanted regular standups even though it was just her and myself for many of the minor projects
This was the most hilarious part of my last company; they hadn't hired any coders yet and I was expected to do a standup meeting with myself and two of my managers. Daily. At my current company we have a daily standup meeting for everyone where the CTO is in on it (can you say micromanager?) and it's just a hassle because you know he's going to come back a few hours later to each of us and ask for a status report anyway.
Just because "Agile" has been coopted by consultants, doesn't mean that the agile manifesto isn't a worthy axiom to organize by.
Minimum Viable product just means get things into customer hands early and get feedback, and refine.
Continuous deployment means build quality verification into your process and automate it.
These aren't buzzwords this is where modern software development at any serious software company is small and large. These are not fads, these are evolutions. If you don't see that, I don't see a lot of promise in the rest of your career as you will only be able to find work at old-guards that are not working on anything meaningful or significant.
> I almost quit but luckily was able to move to a team that had a team area with an open plan.
I've worked on a team that had its own large cubicle, with four people in it. We all mostly worked on the same thing, and we could roll ourselves over to screens and have whiteboard discussions. That was great.
I've had my own office, with a window. That was even better. Then I shared it with another person. That was still great. Then there were three of us. Conversation stopped. None of us worked on the same thing.
And of course I've worked in cubes. What evs.
I've never worked in an open office of long tables and no partitions. I know I'd hate not being able to raise my eyes and stare off into space without inadvertently staring at someone nearby. And the distractions would kill me. But I guess that's just me, and all I have to do is see that at the interview so I can say no; problem solved?
Every time when open offices are mentioned, there's a cure for people who prefer quiet working environment. Well, fuck headphones. I don't want to wear headphones all day just to be able to focus.
For me, life at work is really important. I want to enjoy my time at the office, and that includes socialising while working (where appropriate) - talking about the music we're playing, sharing work problems, complaining about being hungry, etc. I'd find it too stagnant and boring otherwise. My desk jobs (all web stuff) have been:
- open room of around 15 people, bit of chit-chat but mostly quiet
- small area of 3-4, disparate types so very quiet
- large divided room of 10, pretty raucous, very fun, lots of trash talking
- small area, quiet, mostly alone
- room of 4-5, very social
- room of 3-4, we barely shut up all day while working; put headphones on if you want to concentrate
I'd hate to have a room to myself all day. If I want a stretch like that, I come in early or work later than others.
You need to chill. It reads like you're shouting this to my face. "The best defense is an offense" style. Not saying you aren't allowed to be annoyed, just saying it's hard for me to read what you say and not be distracted by the boatload of hurt accompanying it.
One possibility is some people genuinely prefer them. I
keep talking myself up to that, only to read another
article about how everyone is miserable and unproductive
in them.
Given the choice between open plan and private office in San Francisco (where office space costs ~$36/square foot) people say they'd rather have a private office.
Given the choice between open plan in San Francisco and a private office in Chattanooga, Tennessee (where office space costs ~$11/square foot) my guess is you'd get more job applicants for the job in SF.
Perhaps this is a revealed preference for living in SF, and people are willing to trade off private offices to get it.
Or option three: Given the choice between open plan in San Francisco and open plan in Chattanooga, Tennessee, they choose the job in SF.
There's just as many open plan offices outside of SF. I'm in the Chicago area, and I've never worked at, with, or interviewed with any company around here that didn't have an open office plan (with one single exception, and it was a game company, and they were split up into rooms of 6-8 people, no private offices except for management).
> Moreover, engineers are really, really expensive, and making us less productive is costly. The extra space necessary for doors or cubicles could easily pay for itself.
The problem is, there is no good way to measure engineer productivity, but it's extremely easy to figure out the cost of office furniture. Modern corporations believe in optimizing what they can measure and completely ignoring what they can't.
The one thing that kept blaring out at me to be addressed was that your daily productivity doesn't matter. The incentive here is to make the company as productive as possible over the long term (relative to costs).
This is touched upon only slightly here when she considers replaceability.
I'm not saying that I think open offices are or are not great. But the issue of personal productivity simply has to go away in the discussion for me to take it seriously. Otherwise, you just sound like most software engineers I've known who love to write gobs and gobs of code at the expense of everything else.
But the issue of personal productivity simply has to go away in the discussion for me to take it seriously.
And yet, nothing can possibly get done without someone doing it.
I see lots of insistence that without open offices people woild spemd so much more time on wrong things that the lost productivity is more than made up for by better focus. I also see no evidence for these claims.
I see this as businesses pushing off even more of their part of the early employment process.
We've seen how requirements for entry level jobs have risen, colleges work with businesses to try and produce workforce ready graduates but are met with ever increasing minimums.
This is companies deciding that personal productivity isnt something they need to worry about. If someone is not personally productive, then they aren't a good fit.
Why should the business spend resources trying to facilitate a productive environment when they can just hire people who will work in a disruptive environment?
Just as, why should a business spend resources training employees when they can demand only employees with very specific experience.
With outsourcing, visas, and remote working a company has very little restrictions on the level of employee they can seek and find with little difficulty - especially when you include a weak economy in this whole mix.
I dont know, not something i've thought a lot about i could be totally off base, but seems like it could be part of a wider trend to me
You're right, it is self-evident to me that group productivity is not the aggregate of individual productivity. We have companies and teams for a reason.
Secondly, the goal of a company is rarely to maximize aggregate employee productivity. If you can realize the same profits with fewer LOF, stories delivered, or whatever your employees "produce", why not?
Economic productivity is usually defined as the value of what is produced. Again, my point was that if you have some measure of productivity P such that decreasing sum(P(employees)) increases P(sum(employees)), your P is not a good measure.
It's good to be explicit like this. Probably we agree more than I realized, because what I'm trying to get at is that so many of these blog posts complaining about open offices use the wrong P by focusing on a subset what they produce on a daily basis. In fact, I'm suggesting that the employee is using a different P than the employer, and that it's plausible that the employee is the using the wrong P (whether the employer is or not).
Then again, where we might disagree is to what degree the same P can and should be applied to both individual employees and to companies. So often individuals project what they (like to) do onto the team as its primary purpose. "I make widgets, and there are less widgets at the end of the day than there could be, so the company must not be optimized." It might be, or you may overvalue widgets.
I'm only saying that when engineers discuss how they are affected by their work environment, they should keep perspective on the actual value they provide to their employer, which is not equivalent to what and how much they produce on a daily basis.
Of course, but engineers are expensive. When is it ever in the employer's interest to get less output from them? Meet with them in person, do code reviews and such as often as you'd like, but once you've defined a problem for them to solve, what possible value could you get by keeping them in a distracting work environment?
In the very rare case that there is a clearly defined Problem which has been presented to from Above, which you can work on in isolation, and which you should prioritize to the exclusion of all else, then sure, distracting them will cause the Solution to be delayed. Many engineers like to view this kind of situation as their only responsibility and only avenue for adding value. I'm not sure it's accurate.
We have to use a more broad and nuanced perspective on the value we provide when we talk about how that value is affected by our work environment.
> ...more broad and nuanced perspective on the value we provide ...
This sounds like "leveraging synergies". If it was possible to explain any tangible benefit to such an environment, the conversation would move forward. Really, with the Very Rare Case? Developers don't have enough work to do where they don't need to stop and talk every 10 minutes?
No, I'm not saying that you should accept someone else's valuation of your contribution. Instead, I am questioning this particular person's valuation because, like those of so many other employees, it is too narrow.
If you do not even address the long-term value you provide to the company, then you can't possibly make a reasonable argument that open-offices are a detriment to that long-term value. For the record, I believe they probably are a detriment, so I have an incentive for the discourse and arguments to improve.
Personally I like open offices and working with people. I think the problem is not with the idea but with the implementation. Great open offices have some place where you can go and where nobody will bother you. So you can have the alone time you need!
Unfortunately most open office have no such space. Not even the toilets (multiple cramped stalls? hell no!).
Everywhere you go and at all times there are going to be people. I think that is the problem.
I think that depends what kind of introvert you are. I like big open offices because I feel "anonymous" there and it's very easy for me to disconnect from what's going on around me and do my own thing for hours, and then speak with anybody from my team when I need to. If there's some person around with whom you don't connect very well no bad atmosphere is going to occur unless you force it. I don't need a big bubble for myself, I can create a small one easily.
For me a small office with 2 or 3 people is more stressful, it makes me more aware about what's going on around me because it's a more personal space. If you don't connect with somebody in there, the bad atmosphere is going to fill the small office very fast. My bubble is going to trip onto somebody else's bubble more often than in a big space full of people because there's less neutral space around.
Not sure if that makes sense at all... but it does to me.
It really comes down to this. One of my psychology textbooks characterized the basic distinction between extrovert and introvert as the following:
Extroverts become uncomfortable with a lack of external stimuli and thus prefer to surround themselves with stimuli.
Introverts become uncomfortable with too much external stimulus and thus prefer to isolate themselves from excessive stimulus.
To me that basically directly quantifies why some people thrive in an open office and other suffer. I'm not sure what the solution is to requiring essentially two different environments depending on a person's makeup.
There's no shortage of studies that have been done on the issue. Here's a great New Yorker article[1] from last year that references a number of studies. This blog post[2] links to a couple others and addresses costs/alternatives. A "study of over 40,000 survey responses collected over a decade has found that the benefits for workers are quickly outweighed by the disadvantages"[3]. Here's a Washington Post article[4] discussing Facebook's huge open office layout, another Post piece[5] on office furniture designers realizing (citing multiple studies) "open-plan spaces are actually lousy for workers." A TIME article[6] highlighted decades of research that associated open layouts with "greater employee stress, poorer co-worker relations and reduced satisfaction with the physical environment."
Yes I know there is a massive amount of studies that have shown this, but management often believes that these don’t apply to their company. What is needed is a simple experiment that can be performed at your company that show your management how much money they are losing by using an open office layout. Even the most obtuse pointy haired boss can understand that losing money is bad.
I find this argument slanted towards the introverted programmer. Fact is, the introverted programmer is not the standard anymore. Our open office is a mix of different types of people.
I chose an open office because I worked at a company for 9 years. In the first 5 it was an open office. I'm sure some work was suffered, but we had a culture, which is one thing the author seemingly deliberately avoids talking about. The remaining 4 years were stuffed in a cubicle and the office culture died. It became much easier to blame other people for faults because you no longer had a relationship with your co-workers.
I wouldn't say the author is wrong, just that he presents a slanted viewpoint that caters to other programmer introvert types and what their preferences are - but definitely not a fair assessment of the open office.
Everything I've read says that, in order to be maximally productive, software developers needs to have long periods of quiet time where they can focus (I've never seen anything to say this is true only of programmers who are introverts). This, not supporting programmers' introversion, is the reason people criticize open-office plans for software development work.
See, for me that's the problem. This sense that productivity rules all. Why did you become a programmer, or a designer? Was it to be "maximumly productive"? I sure as heck didn't. Productivity is important, but I have to wonder where you decided it was the absolute most important thing. If you dig a bit, I bet that idea didn't come from you - I bet that idea comes from your industry and the social structure around it.
On paper, the development of Color was productive, but in actuality it was a poorly thought out blunder that costed millions of dollars.
Office culture is more important than the product or service you're developing. There, I said it. And I stand by it. The fact of the matter is, I got into this work so I had something to wake to everyday that isn't a complete anti-social "productivity" sinkhole. Every cubicle, or siloed worker represents attrition to company culture. When we were the happiest in our work, we had community tied to it. Think about the classic examples of first economies - what did they all have in common? It wasn't productivity. It was a desire to build, or bake, or design - and it is these markets where happiness was at its height. Don't fool yourselves. You're being productive towards someone else's idea of happiness - not your own.
Last weekend I was at the Amiga's 30th Anniversary event at the Computer History Museum here in Mountain View. A lot of the original Amiga engineers spoke. A couple of things stood out that are relevant here. First, the culture of the company was so positive, and what they built was so impactful, that even 30 years later, they considered it the highlight of their careers.
Second was that the engineers had offices with doors that closed.
As a programmer, I want to build solid, easy-to-use, impactful products that delight people. In order to do that I need to focus, and in order to focus I need long periods free of distraction. It's also satisfying to be able to get into a state of flow. I don't need that all the time, but I need and want it some of the time.
As a (hopefully!) future tech CEO, I want to work with people who want the same. As as Amiga Inc (and many others) have proved, small groups of people with private offices can produce wonderful things and still have a great corporate culture. I also want to make sure I spend my company's or my investors' money wisely and efficiently.
(As an aside, the reason that the Amiga had line-drawing built into hardware is because one of the software developers had experience with line-draw algorithms, recognized that it could be easily put into hardware, and convinced the hardware designer to do so. Which I offer as a counter to the people who think that private offices means people won't collaborate).
Finally, I'm rather taken aback that the director of a design agency would get onto a public forum and announce that he doesn't really care about productivity. Productivity relates to money spent; don't your customers care about how much it costs to engage you?
> Second was that the engineers had offices with doors that closed.
And sometimes that works well. Othertimes it's a hinderance. In this case, I'm arguing that it removes culture from the workplace. There's hardly anyway to deny that. There are also indicators that suggest lowering social interaction is linked with depression; which would have some effect on company culture.
> As a programmer,...
You mean, "you, as an individual person with individual needs", right?
> Finally, I'm rather taken aback that the director of a design agency would get onto a public forum and announce that he doesn't really care about productivity. Productivity relates to money spent; don't your customers care about how much it costs to engage you?
That's an oversimplification of my statement. Our clients pay for us to take our time, to welcome new ideas and research ways to innovate their message, goals or vision. They do care that we spend the extra time – in fact they knowingly pay for it.
Agree, I take the opposite side of this argument as well, and the author is very biased toward a certain personality type. I love open office plans and I think they have intangible benefits that many people discount.
But, I'm tired of arguing against the hivemind on this one. So be it.
Well, he's talking about productivity. You seem to be in a good position to answer that question. Where were you more productive, in the open office, or in the cubicle?
However, a cubicle might not be the best comparison. A cubicle is not the same as an office with a door that closes. As the IBM example elsewhere in these comments said, not all cubicles provide the benefit. A minimum height of the walls being one requirement.
I've seen dozens of these discussions over the years, and it never ceases to amaze me when I see that the tax angle is being left out. At a prior company, the CFO flat-out told me that tax concerns were the driving factor. To put it simply, in many places the "open" space is subject to lower real estate taxes, and depreciation on the partitions etc. within that space can lower tax bills even further (out of all proportion to their actual cost). It's totally FUBAR, like most tax-related stuff, but it looks good on a balance sheet. When the benefits are in black and white but the drawbacks are nebulous - even if they're practically certain to be greater - it's pretty predictable which way execs and investors will go.
Communication is probably the most important factor to success for both small and large companies. Most things that go wrong is a direct consequence of bad or no communication.
If everyone is inside his or her own office, your company depends on having good email or message communication. Everyone should include the right persons when sending emails, planning meetings, etc ... . This is not impossible of course.
When you have an open office plan, your communication becomes 'in your face'. You might overhear things that are also relevant for you. You see or hear a lot more than through official communication channels alone. And that is probably the reason to their success. Of course interruptions etc are disadvantages, but nothing beats having good communication.
> You might overhear things that are also relevant for you.
The key word is might. I remember a similar HN discussion a few months back someone mentioned that because they happened to overhear a conversation between two engineers they were able to save weeks of them working on the wrong thing.
Anecdotes like that aren't signs that open offices are successful or result in "having good communication" - it means you have horrible communication and you got lucky. What if that conversation happened while the person who serendipitously overheard it was in the bathroom? Or on vacation? Or in a meeting? Or wearing headphones? Good communication means not relying on chance to prevent engineers from wasting their time.
You are correct, but how many companies you worked for had 'good' communication? I worked for 5 companies, and have been outsourced to an extra 2. Communication is a real issue in companies, and the ones that already get this partly right will win over those who don't get it right.
Like I said, it is possible to get communication right without having an open office plan. But at least with an open office plan, you have the bare minimum set up. And for big companies, the bare minimum is already better than average.
The same for small companies. If you are with 5 to 10 people, not everything will be officially communicated through email. You know stuff because it happened in the same room as everyone is sitting in.
You won't catch me defending open offices, but they're even worse when they're full and the meeting rooms are booked all the time. I even frequently can't attend a standup or other meeting because someone is in the room who shouldn't be. And of course they're always on a phone call with a customer, or some other thing that can't be interrupted.
I used to work from home, for 4 years. It's hard to deal with all the stupid shit you put up with in offices.
> And yet everyone, including some exceptionally profitable companies, use them. Why?
Because they are cheaper to build and outfit.
This will always be attractive to management incapable of looking beyond the current expense period. All other reasons are simply marketing to the captive audience.
Sadly, I think a factor of why smaller orgs do this is simply because larger orgs do.
I was working at a smallish company that had a wonderful layout. The customer support team had an open layout of mini cubes (walls between desks but not extending past desk), which worked great for them, and the engineers had a closed off room that was basically off limits to anyone who wasn't an engineer. It was wonderful and productive. Then, the owners decided to change everything to intermingled open for everyone. No more walls, just flat desks everywhere. Why? All because they visited a couple other companies such as Facebook, etc. And that is how they saw it done there. If it works for them it must work for us, right?
I pretty much stay out of the whole Open Office debate since I seem to be pretty much unaffected. My powers of concentration are obscenely high so even without headphones I have to work to be interruptible.
I actually liked Open Office plans because there were more interruptions. When you have a family to get home to there is something wrong with realizing it's nearly 10pm and the last thing you remember before you started coding was eating lunch. The open office plans gave just enough background distraction to allow me to surface every once and a while.
It really doesn't matter. Successful startup X could have began in a coffee shop, apartment, dorm room, open office...it was still going to be successful. People with the right drive are going to create and be productive despite their working environment, not because of it.
It's only later on in the process when the company can buy office space and do some hires, when people start to get picky about where they work. I've had some of my most productive days on a borrowed old laptop and a folding table in a breakroom.
Also, add total hot desking to this for another bump down the productivity scale. Then, to make things worse, they usually oversubscribe the people to desks because at any one time quite a few people would be in meetings or not in the office. Net result, huge saving on rent. After a few days of sharing a desk with another guy and swapping power points to keep both our laptops charged, I complained and they looked at me like I was a freak of nature.
Open offices are not my favorite - I prefer silence or music of my choosing when I'm working. There's nothing worse than a loud sales meeting going on 10 feet away from you when you're trying to code.
Maybe it's because developers don't run companies in general. For management cost and control would be more important considerations. I know of a local software company that's owned by 2 developers where they have individual offices. While 37 Signals kind of have an open office, it's meant to be as quiet as a library to avoid disruptions. People who manage probably don't need to concentrate as much as the people who create the stuff. I suppose my point is that people at the top call the shots, not the people at the bottom.
> People who manage probably don't need to concentrate as much as the people who create the stuff.
That may be true but it's probably more of a case of the difference between the "Maker's Schedule" and the "Manager's Schedule"[1] - i.e. managers probably don't realize how much distractions or interruptions are bad for "maker's" productivity because a manager's day is mostly distractions or interruptions.
That comment is just a way of saying 'they' are not like 'me' (but trying to make it sound general).
Facebook, led by a (former) dev, has gone all-in on open plan offices - which is a direct counterpoint. That people at the top call the shots is tautological.
> Facebook, led by a (former) dev, has gone all-in on open plan offices
As another comment[1] pointed out it may have been founded by a developer, but it was one who never worked anywhere other than the company he founded and grew. Things that work well for a company of ten or twenty don't necessarily work best when it grows into the hundreds or thousands of employees.
My point was to refute the assertion that it's "because developers don't run companies in general." If we're going to add lots of caveats to what 'developer' means, then we're essentially making a 'No true scotsman argument' -- (i.e. Zuck hasn't worked anywhere else, and is ostensibly a manager now).
Oh sorry, I certainly didn't mean to suggest he isn't a "true" developer - I'm sure he could code circles around me. I meant only that you were pointing to Facebook as a "direct counterpoint" to the suggestion that "developers don't run companies in general" - and it's not a great counterpoint because he's not going into the office every day and writing code, and hasn't since Facebook was a small/early-stage startup (where open layouts make the most sense).
i.e. he isn't making the decision as a developer, he's making it as a manager and the comment you were originally replying to was suggesting "for management cost and control would be more important considerations." [Though I'm not sure I entirely agree with that, as pointed out in my own reply to that comment]
Followed in some number of years by people in Africa doing the former jobs of people in San Francisco. If you in America can work remotely, there's no reason why someone with equivalent skills and 1/20 the cost can't work really remotely.
You might point to stories of Indian contractors like Tata, and how badly that works out for anything that's not Telco CRUD. (Note, I'm talking about the business model, not individual Indians.) But that was just the first attempt. Software will figure it out, just as textiles figured it out.
Personally I feel that with varying degrees of effort I can work in either end of the spectrum: closed offices, cubicle farm, open office, coffeeshop, etc. Some people can't. Some people need a desk and a door to close.
So for me from a founders perspective, the best kind of environment to get things started would be a company space that encourages setups like a cafe/lounge/coffeetable-style area for laptop users and those who need to be comfortable, focused, and relatively available with ease for discussion .. as well as areas of the workfloor where there are offices, for example, for those who need them, or hackylab environment for where its needed, etc.
Alas, this kind of mixed-format personal workspace organization is not so economical, and it usually requires someone be responsible for the estate, rather than just having a cubicle-cleaning service, etc. etc. All kinds of legal hoops, too, whatever.
But in my experience, this seems to be the case: the nicest 'kind' of workspace for a modern startup company in my opinion, is your average McMansion.
(If you can fit and find one local to everyone else who has to commute to it.)
I've seen great mixed-format startup scenarios in your average American mega-house, cubicle farm in the 4-car garage, massive lounge/rec-room space for the hackers, room #? of (4..6) bedrooms converted to shared office rooms, 2 or 3 to a room, for those who want 'a desktop' and a way to isolate themselves from the rest of the team, and so on.
Of course, this works if the house is clean and maintained as a work environment, and indeed if people don't live there but rather treat it as their place of work. That can be difficult to do in some neighborhoods.
Wouldn't it then be interesting to see some sort of 'office space plan' revolution occur in the broader sense, which accommodates all of these modes of operation, to implement a model you could name "Big Family Home"?
Instead of cubical farms, you find interior spaces analog to a modern home plan, s/beds/desktops/, centered around a common room/kitchen/meeting area, etc.
Generally the way I feel when I walk into a very large organization and its just endless row of cubicle, mind-numbing soviet khaki walls lit by brain-sucking illumination standards, all I want to do is go home.
That is a hard impulse to resist all day. If, instead, I'm in a place that accommodates a large number of peoples needs with some consistent attention to variety, usually its pretty hard to go home at the end of the day .. I really like going to a business where people are working and comfortable because they know where the kitchen is, and someone is making lunch for everyone while code gets written, clients are met, teams are having meetings out on the deck in the sunshine, etc. I'm sure there are movements towards this; but the cubicle-farm seems to have persisted for decades, alas, and for this I guess we must thank the bean-counters.
I had a flashback of playing the board game Clue (http://www.hasbro.com/common/instruct/clue_%282002%29.pdf) with my siblings when I was younger. Specifically, I imagined a small team of developers in a large McMansion-Office trying to debug a large application saying things like, "I suggest the bug was commit'd in the kitchen, by Tom with his Custom Arch Laptop."
One factor of open offices that is seldom counted, but remains very important, is "fitting the organization to an image". If you're bringing bankers or investors or other external parties through, what is more "impressive": A floor full of hard working bees, a buzz of energy filling the air, whiteboards full of complex diagrams (that probably haven't been erased in months if not years) and quirky and crazy "personalizations" on all of the desks, or walking down a hall of quiet closed doors. Or, worse still (and the #1 reason why many companies don't allow remote work), no office really at all because everyone is everywhere.
That "show off the empire" thing is absolutely a function of decisions. I was recently part of such a decision process where a group was looking to become startup-y, and cargo culting the media image of what such organizations look like reigned supreme.
It seems to be common to do that... A friend of mine who was at a unicorn company in SF before it was a unicorn, had nice tables, furniture, and overpowered computers everywhere around the pathway from the entrance to the conference room. So when VCs and business angels came it looked very nice but everybody out of the pathway had nothing like that.
Ya I know... Like people in the front were able to have a budget of ~2000 dollars per person (hello Thunderbolt Displays!) but for others it was not even more than 300 dollars.
Having, in the past, worked in a mandatory open office, and currently being on a remote team, this is right on target. The decision to have your employees work in an open office plan is not a decision of productivity.
The factor you mentioned is indeed seldom counted, but it implies that such decisions about having closed rooms/ cubicles or open spaces are strongly related to the company evolution stage. This factor is much more substantial for small seeds or young start-up companies. This show-off sometimes tells a nice story, just like a nice demo, but for the long run you might need more than a good story. In that sense the "long run" sometimes forces you to treat your work environment to be more solid, or to support more consistent operations, and from my experience productive operations do sometimes require at least a cubicle.
If I'm bringing bankers or investors, I think the most important factor will still be, "show me the money".
For example, why have an office full of workers when you can outsource most of your work and save a ton?
I've seen a lot of articles arguing against the productivity losses of an open office. Unfortunately I can't remember seeing solid numbers on whether or not those productivity losses (or possibly gains) translate to more or less profit.
The major takeaways I have from reading about this when I started in tech entrepreneurship are:
1) Knowledge workers need 30 square feet of desk to get optimal efficiency
2) They need 100 square feet of space
3) cubicles are as good as an office if the walls are over 8 feet high (!)
4) small teams tend to want to talk at the same time, and be quiet at the same time.
These together round up into recommended 400 or so square foot offices with doors that close and house teams of four people, all working on the same thing at the same time.
If teams are 2 or 3, they should have a smaller space. That's enough space for desks, a couch, a whiteboard area, etc.
I have found over the years that this tracks fairly closely with my own preferences. I really want to be able to talk and interrupt a team-mate when we're working if we're working on the same thing. We also often are hacking away and need silence, but instant access to whoever is in the know about a certain bit of code is undeniably great, and provided we aren't breaking each-other out of flow, is fine.
This system works really well when it's a small team and the team is working on the same stuff. It breaks down when there are multiple things happening or more than four or so people: interruptions abound, killing productivity.
I think you'll often see workers getting this situation when they really need to be productive by commandeering a conference room and shutting the door. But, it makes sense to provide it for teams in general, in my opinion.